WO2020071072A1

WO2020071072A1 - Information provision system, mobile terminal, information provision method, and computer program

Info

Publication number: WO2020071072A1
Application number: PCT/JP2019/035744
Authority: WO
Inventors: 良明林
Original assignee: 住友電気工業株式会社
Priority date: 2018-10-02
Filing date: 2019-09-11
Publication date: 2020-04-09
Also published as: JPWO2020071072A1

Abstract

This invention is provided with: an acquisition unit for acquiring, on the basis of dynamic map information, movement information indicating the movement state of a plurality of moving bodies, and outer appearance information for the plurality of moving bodies; a computation unit for performing, on the basis of the movement information, a collision prediction of predicting whether or not there will be a collision between an object moving body having a mobile terminal, from among the plurality of moving bodies, and another moving body other than the object moving body, and obtaining an evaluation value indicating the possibility of collision between the object moving body and another moving body; and an assessment unit for assessing whether or not to communicate a collision prediction result indicating the result of the collision prediction performed by the computation unit to the mobile terminal. The computation unit obtains the evaluation value on the basis of: the collision prediction time length until the object moving body and another moving body collide, the collision prediction time length being included in the collision prediction result; outer appearance attribute values corresponding to the respective outer appearance information for the object moving body and each of the other moving bodies; and behavior attribute values corresponding to behavior information indicating the specifics of the respective behavior of the object moving body and each of the other moving bodies, obtained from the movement information. The assessment unit assesses, on the basis of the evaluation value, whether or not to communicate the collision prediction result to the mobile terminal.

Description

Information providing system, mobile terminal, information providing method, and computer program

The present disclosure relates to an information providing system, a mobile terminal, an information providing method, and a computer program.
This application claims the priority based on Japanese Patent Application No. 2018-187530 filed on Oct. 2, 2018, and incorporates all the contents described in the Japanese application.

Patent Document 1 discloses a traffic system that reports information about another vehicle to the own vehicle.

JP 2013-109746 A

The information providing system according to an embodiment includes, based on dynamic map information in which dynamic information of a plurality of moving objects is superimposed on map information, moving information indicating a moving state of the plurality of moving objects, Body appearance information, an acquisition unit that acquires, a target mobile having a mobile terminal among the plurality of mobiles, a collision prediction of whether or not a mobile other than the target mobile collides; While performing based on the movement information, a calculation unit that calculates an evaluation value indicating the possibility of collision between the target mobile object and the other mobile object, and a collision prediction result indicating a result of the collision prediction by the calculation unit, A determination unit that determines whether or not to notify a mobile terminal, wherein the calculation unit includes a collision prediction time, which is included in the collision prediction result, until the target mobile unit and the other mobile unit collide with each other. And the target moving body, and the other Based on an appearance attribute value corresponding to the appearance information of each moving object, and an action attribute value obtained from the movement information, the action attribute value corresponding to action information indicating the action content of each of the target moving object and the other moving object. The evaluation unit obtains the evaluation value, and the determination unit determines whether to notify the collision prediction result to the mobile terminal based on the evaluation value.

An information providing system according to another embodiment includes an acquisition unit configured to acquire movement information indicating a movement state of the plurality of moving objects based on dynamic map information in which dynamic information of the plurality of moving objects is superimposed on map information. And performing a collision prediction of whether or not a target mobile having a mobile terminal among the plurality of mobiles collides with another mobile other than the target mobile based on the movement information, A calculating unit that calculates an evaluation value indicating a possibility that the moving body collides with the other moving body, and determines whether to notify the mobile terminal of a collision prediction result indicating a result of the collision prediction by the calculating unit; A determination unit, wherein the calculation unit is configured to determine, based on position information of the plurality of moving objects included in the movement information, a relative distance between the target moving object and each of the other moving objects, and the relative distance. Of the relative distance And the calculated evaluation value, the evaluation unit based on the displacement of the relative distance, whether to notify the collision prediction result to the mobile terminal determines on the basis of the evaluation value.

In addition, a mobile terminal according to another embodiment is a mobile terminal that receives the collision prediction result from the information providing system and outputs the collision prediction result to a user.

An information providing method according to another embodiment is an information providing method for providing information to a mobile terminal, based on dynamic map information in which dynamic information of a plurality of moving objects is superimposed on map information, An acquisition step of acquiring movement information indicating a moving state of a moving body and appearance information of the plurality of moving bodies, a target moving body having the mobile terminal among the plurality of moving bodies, and a target other than the target moving body Performing a collision prediction of whether or not to collide with another moving body based on the movement information, and calculating an evaluation value indicating a possibility of collision between the target moving body and the other moving body; and Determining whether or not to notify the mobile terminal of a collision prediction result indicating a result of the collision prediction by the calculation step, wherein in the calculation step, the target movement included in the collision prediction result is included. A collision prediction time until the object and the other moving object collide with each other, the target moving object, an appearance attribute value according to the appearance information of each of the other moving objects, and the object obtained from the movement information. The evaluation value is obtained based on a behavior attribute value corresponding to behavior information indicating a behavior of each of the moving body and the other moving body, and in the determining step, the collision prediction result is notified to the mobile terminal. Is determined based on the evaluation value.

Further, a computer program according to another embodiment is a computer program for causing a computer to execute an information providing process for providing information to a mobile terminal, and the computer converts dynamic information of a plurality of moving objects into map information. An acquisition step of acquiring movement information indicating a movement state of the plurality of moving objects, and appearance information of the plurality of moving objects, based on the superimposed dynamic map information; A target mobile having a terminal and a collision prediction of whether or not a mobile other than the target mobile collides is performed based on the movement information, and the target mobile and the other mobile are A calculating step of obtaining an evaluation value indicating a possibility of collision; and determining whether or not to notify the mobile terminal of a collision prediction result indicating a result of the collision prediction by the calculating step. Determination step, wherein in the calculation step, the collision prediction time, which is included in the collision prediction result, until the target mobile body and the other mobile body collide with each other, the target mobile body, and the other An appearance attribute value corresponding to the appearance information of each moving object, and an action attribute value corresponding to action information indicating the action content of each of the target moving object and the other moving objects obtained from the movement information, A computer program that determines whether or not to notify the collision prediction result to the mobile terminal based on the evaluation value in the determining step.

FIG. 1 is a schematic diagram illustrating an overall configuration of a wireless communication system according to an embodiment. FIG. 2 is a block diagram illustrating an example of an internal configuration of the edge server and the core server. FIG. 3 is a block diagram illustrating an example of an internal configuration of an in-vehicle device of a vehicle equipped with a communication terminal. FIG. 4 is a block diagram illustrating an example of an internal configuration of the pedestrian terminal. FIG. 5 is a block diagram illustrating an example of an internal configuration of a roadside sensor on which a wireless communication device as a communication terminal is mounted. FIG. 6 is an overall configuration diagram of the information providing system according to the present embodiment. FIG. 7 is a sequence diagram illustrating an example of dynamic information update processing and distribution processing. FIG. 8 is a functional block diagram of the edge server showing a function for providing a collision prediction result. FIG. 9 is a diagram illustrating an example of the moving object database. FIG. 10 is a diagram showing a part of the part for registering the attribute information in the moving object database, and shows a column for registering the attribute information when the moving object is a vehicle. FIG. 11 is a diagram showing a part of the part for registering the attribute information in the moving object database, and shows a column for registering the attribute information when the moving object is a pedestrian. FIG. 12 is a flowchart illustrating an example of calculation processing of the evaluation value of the collision prediction result by the calculation unit. FIG. 13 is a flowchart illustrating an example of the individual calculation process in step S55 in FIG. FIG. 14 is a flowchart illustrating an example of the process of adding a collision prediction target in FIG. FIG. 15 is a flowchart showing an example of the vehicle addition process in FIG. FIG. 16 is a flowchart showing an example of the pedestrian addition process in FIG. FIG. 17 is a diagram illustrating an example of the evaluation value database. FIG. 18 is a flowchart illustrating an example of a determination process performed by the determination unit. FIG. 19 is a diagram showing a situation around an intersection according to scenario 1. FIG. 20 is a diagram showing a situation around an intersection according to scenario 2. FIG. 21 is a diagram showing a situation around an intersection according to scenario 3. FIG. 22 is a diagram showing a situation around an intersection according to scenario 4. FIG. 23 is a diagram showing a situation around an intersection according to scenario 5. FIG. 24 is a diagram showing a situation around an intersection according to scenario 6. FIG. 25 is a diagram illustrating a storage unit of the edge server according to the second embodiment. FIG. 26 is a diagram illustrating an example of the relative distance database. FIG. 27 is a flowchart illustrating an example of a calculation process performed by the calculation unit 31a according to the present embodiment. FIG. 28 is a flowchart showing an example of the adding process based on the relative distance in step S59 in FIG. FIG. 29 is a flowchart illustrating an example of an addition process for an evaluation value. FIG. 30 is a flowchart illustrating an example of a process corresponding to vehicle-to-pedestrian or pedestrian-to-vehicle. FIG. 31 is a flowchart illustrating an example of a process corresponding to a vehicle-to-vehicle process. FIG. 32 is a diagram illustrating an example of the evaluation value database after the addition processing based on the relative distance. FIG. 33 is a diagram for explaining a specific example of the addition process based on the relative distance. FIG. 34 is a diagram for describing another specific example of the addition process based on the relative distance. FIG. 35 is a diagram for describing another specific example of the addition process based on the relative distance. FIG. 36A is a diagram for describing another specific example of the addition processing based on the relative distance. FIG. 36B is a diagram for describing another specific example of the addition processing based on the relative distance.

[Problems to be solved by the present disclosure]
In the above conventional example, the central device of the system is configured to determine the presence or absence of an abnormal event of each vehicle based on vehicle information obtained from each vehicle, and to notify each vehicle of the determination result.

In the above conventional example, the system is configured to notify a result of occurrence of an abnormal event. For example, using such a system, a result of predicting a collision occurring between vehicles (moving bodies) is notified. Can be considered.
Here, if the current speed, azimuth, and the like of the moving bodies are used to predict the collision between the moving bodies, a collision prediction time until the moving bodies that are predicted to collide can collide can be obtained.
It can be said that the shorter the collision prediction time is, the higher the possibility of collision is and the more urgent it is required. Therefore, if it is determined whether or not to notify the mobile body of the collision prediction result according to the collision prediction time, it is possible to notify the collision prediction result according to necessity.

By the way, the moving objects include not only vehicles but also traffic weak people such as pedestrians, and the possibility of collision may differ depending on the attributes of each moving object.
For this reason, if the possibility of a collision is evaluated only based on the collision prediction time and it is determined whether or not to notify the mobile body of the collision prediction result, the possibility of the collision evaluated based on the collision prediction time and the actual collision It is conceivable that a divergence occurs from the possibility and the notification of the collision prediction result is not properly performed.
Therefore, it is desired to appropriately provide a collision prediction result for which a notification is highly necessary to the mobile terminal.

The present disclosure has been made in view of such circumstances, and has as its object to provide a technology capable of appropriately providing a collision prediction result to a mobile terminal.

[Effects of the present disclosure]
According to the present disclosure, it is possible to appropriately provide a collision prediction result to a mobile terminal.

First, the contents of the embodiment will be listed and described.
[Overview of Embodiment]
(1) An information providing system according to one embodiment is configured to include, based on dynamic map information in which dynamic information of a plurality of moving objects is superimposed on map information, movement information indicating a moving state of the plurality of moving objects, An acquisition unit that acquires appearance information of a plurality of moving objects, and whether or not a target moving object having a mobile terminal among the plurality of moving objects collides with another moving object other than the target moving object; A computing unit that performs a collision prediction based on the movement information, and calculates an evaluation value indicating a possibility of collision between the target moving object and the other moving object; and a collision prediction indicating a result of the collision prediction by the computing unit. A determination unit that determines whether or not to notify a result to the mobile terminal, wherein the calculation unit is included in the collision prediction result until the target mobile unit and the other mobile unit collide with each other. Collision prediction time, the target moving body, and Appearance attribute value according to the appearance information of each of the moving objects, obtained from the movement information, the target moving object, and an action attribute value according to the action information indicating the action content of each of the other moving objects, And determines the evaluation value based on the evaluation value. The determination unit determines whether to notify the collision prediction result to the mobile terminal.

According to the above configuration, the evaluation value for determining whether or not to notify the collision prediction result is determined based on the collision prediction time, the appearance attribute value, and the action attribute value. The appearance and behavior of the target moving object and other moving objects can be reflected in the determination as to whether or not to notify the result.
Therefore, for example, for a collision prediction regarding a moving object having a high possibility of collision in appearance or behavior, an evaluation value can be set so as to be weighted in a direction in which a collision prediction result is notified, and the necessity of notification is required. In determining, the appearance and behavior of the moving object can be taken into account.
As a result, the necessity of notification can be appropriately determined, and a collision prediction result with high necessity of notification can be appropriately provided to the mobile terminal.

(2) The information providing system may further include a notification unit that notifies the collision prediction result to the mobile terminal that the determination unit determines to notify.
In this case, information can be provided only to the mobile terminals for which it is determined that the collision prediction result is necessary based on the evaluation value.

(3) Further, in the information providing system, the calculation unit specifies a moving body that is predicted to collide with the target moving body among the other moving bodies based on the collision prediction result, The evaluation value may be obtained using the collision prediction time, the appearance attribute value, and the action attribute value of the moving object predicted to collide.
In this case, it is possible to obtain an evaluation value relating to the collision prediction result of the moving body predicted to collide.

(4) The information providing system further includes a control unit that controls the mobile terminal and causes the mobile terminal to execute a process of outputting the collision prediction result to a user of the mobile terminal, wherein the control unit includes: The mobile terminal may be controlled so that an output mode of the collision prediction result is different depending on the appearance information and the behavior information of the moving body predicted to collide.
In this case, it is possible to output to the user in an output mode according to the characteristics of the moving object predicted to collide.

(5) In the information providing system, when the other moving object is a vehicle, the appearance attribute value includes an attribute value corresponding to a size of the vehicle, an attribute value corresponding to a shape of the vehicle, At least one of an attribute value according to a color of the vehicle and an attribute value according to information described on a license plate attached to the vehicle may be included.

(6) In the information providing system, when the other moving object is a pedestrian, the appearance attribute value includes an attribute value corresponding to the height of the pedestrian, an attribute value corresponding to the clothes of the pedestrian, And at least one of attribute values according to the pedestrian's equipment.

(7) In the information providing system, when the other moving object is a vehicle, the action attribute value includes at least an attribute value corresponding to a current driving state and an attribute value corresponding to a past driving history. One may be included.

(8) In the information providing system, when the other moving object is a pedestrian, the action attribute value includes an attribute value corresponding to a walking location of the pedestrian and an attribute corresponding to a walking locus of the pedestrian. At least one of a value and an attribute value according to the operation of the traffic light with respect to the lamp color may be included.

(9) In the information providing system, the calculation unit is configured to determine a relative distance between the target moving object and each of the other moving objects based on position information of the plurality of moving objects included in the movement information; The relative distance may be determined, and an additional value based on the relative distance and the relative distance may be added to the evaluation value.
In this case, the collision prediction result is not provided in the collision prediction time, but the collision prediction result is provided to the mobile terminal of the target moving body when the possibility of collision occurs due to the approach of the target moving body and another moving body. In the evaluation value.

(10) An information providing system according to another embodiment acquires movement information indicating a movement state of the plurality of moving objects based on dynamic map information in which dynamic information of the plurality of moving objects is superimposed on map information. And performing a collision prediction of whether or not a target mobile having a mobile terminal among the plurality of mobiles collides with another mobile other than the target mobile based on the movement information. An arithmetic unit for calculating an evaluation value indicating a possibility of collision between the target mobile object and the other mobile object, and whether to notify the mobile terminal of a collision prediction result indicating a result of a collision prediction by the arithmetic unit. A determination unit that determines the relative distance between the target mobile unit and each of the other mobile units based on the position information of the plurality of mobile units included in the movement information, and Determine the displacement of the relative distance, Obtains the evaluation value based on the displacement of the pair distance and the relative distance, the determining unit determines whether to notify the collision prediction result to the mobile terminal determines on the basis of the evaluation value.

According to the above configuration, the evaluation value for determining whether or not to notify the collision prediction result is obtained based on the relative distance between the target moving object and each of the other moving objects, and the displacement of the relative distance. If the target moving body and the other moving body are within the range of the predetermined relative distance and the relative distance decreases so as to approach each other, it can be determined that the possibility of collision has occurred. . As a result, regardless of the traveling directions of the target moving object and the other moving objects, other moving objects having a possibility of collision with the target moving object can be widely evaluated, and the necessity of notification can be appropriately determined. it can.

(11) A mobile terminal according to another embodiment receives the collision prediction result from the information providing system according to any one of (1) to (10) and outputs the collision prediction result to a user. It is.

(12) An information providing method according to another embodiment is an information providing method for providing information to a mobile terminal, based on dynamic map information in which dynamic information of a plurality of moving objects is superimposed on map information. Moving information indicating a moving state of the plurality of moving objects, appearance information of the plurality of moving objects, an obtaining step of obtaining, a target moving object having the mobile terminal among the plurality of moving objects, An evaluation value indicating the possibility of collision between the target moving object and the other moving object, while performing a collision prediction of whether or not the moving object collides with another moving object other than the target moving object based on the movement information. And a determining step of determining whether or not to notify the mobile terminal of a collision prediction result indicating a result of the collision prediction by the calculation step, wherein the calculation step includes the collision prediction result. The A collision prediction time until the target moving object and the other moving object collide with each other, an appearance attribute value according to the appearance information of the target moving object and the other moving object, and the movement information. Determining the evaluation value based on the target mobile object and an action attribute value corresponding to action information indicating the action content of each of the other mobile objects. Whether to notify the terminal is determined based on the evaluation value.

(13) Further, a computer program according to another embodiment is a computer program for causing a computer to execute an information providing process for providing information to a mobile terminal, and stores the dynamic information of a plurality of moving objects in the computer. An acquisition step of acquiring movement information indicating a moving state of the plurality of moving bodies, and appearance information of the plurality of moving bodies, based on the dynamic map information superimposed on the map information; The target mobile having the mobile terminal and a mobile other than the target mobile perform a collision prediction based on the mobile information as to whether or not a collision occurs, and the target mobile and the other mobile are used. A calculating step of calculating an evaluation value indicating a possibility of collision with the body, and determining whether to notify the mobile terminal of a collision prediction result indicating a result of the collision prediction by the calculation step. Determination step, wherein in the calculation step, the collision prediction result included in the collision prediction result, the collision prediction time until the target moving body and the other moving body collides, the target moving body, and the An appearance attribute value according to the appearance information of each of the other moving objects, and an action attribute value obtained from the movement information and corresponding to action information indicating an action content of each of the target moving object and the other moving object; And a computer program that determines whether or not to notify the collision prediction result to the mobile terminal based on the evaluation value in the determining step.

[Details of Embodiment]
Hereinafter, preferred embodiments will be described with reference to the drawings.
In addition, at least a part of each embodiment described below may be arbitrarily combined.

[Overall configuration of wireless communication system]
FIG. 1 is a schematic diagram illustrating an overall configuration of a wireless communication system according to an embodiment. Referring to FIG. 1, a wireless communication system includes a plurality of communication terminals 1A to 1D capable of wireless communication, one or a plurality of base stations 2 wirelessly communicating with communication terminals 1A to 1D, and a wired or wireless communication with base station 2. It includes one or more edge servers 3 that communicate with each other and one or more core servers 4 that communicate with the edge servers 3 by wire or wirelessly. The communication terminals 1A to 1D are also referred to as communication terminals 1.

(4) The core server 4 is installed in a core data center (DC) of the core network. The edge server 3 is installed in a distributed data center (DC) of a metro network. The metro network is, for example, a communication network constructed for each city. Each metro network is connected to the core network. The base station 2 is communicably connected to any one of the edge servers 3 of the distributed data center included in the metro network.

(4) The core server 4 is communicably connected to the core network. The edge server 3 is communicably connected to a metro network. Therefore, the core server 4 can communicate with the edge server 3 and the base station 2 belonging to each metro network via the core network and the metro network. The base station 2 includes at least one of a macro cell base station, a micro cell base station, and a pico cell base station.

In the wireless communication system of the present embodiment, the edge server 3 and the core server 4 are general-purpose servers capable of performing SDN (Software-Defined Networking). The relay device such as the base station 2 and a repeater (not shown) is a transport device capable of SDN. Therefore, a plurality of virtual networks (network slices) S1 to S4 satisfying conflicting service requirements such as low-delay communication and large-capacity communication can be defined as physical devices of the wireless communication system by the network virtualization technology.

The above-described network virtualization technology is a basic concept of a “fifth generation mobile communication system” (hereinafter abbreviated as “5G” (5th Generation)) that is currently being standardized. Therefore, the wireless communication system according to the present embodiment is compliant with, for example, 5G.
However, the wireless communication system according to the present embodiment is a mobile communication system capable of defining a plurality of network slices (hereinafter, also referred to as “slices”) S1 to S4 according to predetermined service requirements such as delay time. It is not limited to 5G. The number of slices to be defined is not limited to four, but may be five or more.

In the example of FIG. 1, each of the network slices S1 to S4 is defined as follows. The slice S1 is a network slice defined so that the communication terminals 1A to 1D communicate directly. The communication terminals 1A to 1D that directly communicate in the slice S1 are also referred to as “node N1”.

Slice S2 is a network slice defined so that communication terminals 1A to 1D communicate with base station 2. The highest communication node in the slice S2 (the base station 2 in the illustrated example) is also referred to as “node N2”.

The slice S3 is a network slice defined so that the communication terminals 1A to 1D communicate with the edge server 3 via the base station 2. The highest communication node in the slice S3 (the edge server 3 in the illustrated example) is also referred to as “node N3”. In the slice S3, the node N2 becomes a relay node. That is, data communication is performed by an uplink route from the node N1 to the node N2 to the node N3 and a downlink route from the node N3 to the node N2 to the node N1.

Slice S4 is a network slice defined so that communication terminals 1A to 1D communicate with core server 4 via base station 2 and edge server 3. The highest-order communication node in the slice S4 (the core server 4 in the illustrated example) is also referred to as “node N4”. In the slice S4, the nodes N2 and N3 become relay nodes. That is, data communication is performed by an uplink route of the node N1 → node N2 → node N3 → node N4 and a downlink route of the node N4 → node N3 → node N2 → node N1.

ルーティング In the slice S4, there is a case where the routing is performed without using the edge server 3 as a relay node. In this case, data communication is performed by an uplink route from the node N1 to the node N2 to the node N4 and a downlink route from the node N4 to the node N2 to the node N1.

In the case where a plurality of base stations 2 (nodes N2) are included in the slice S2, routing for tracing communication between the

base stations

2 and 2 is also possible. Similarly, when a plurality of edge servers 3 (nodes N3) are included in the slice S3, routing for tracing communication between the

edge servers

3 and 3 is also possible. In the case where a plurality of core servers 4 (node N4) are included in the slice S4, routing for following the communication of the

core servers

4 and 4 is also possible.

The communication terminal 1A is a wireless communication device mounted on the vehicle 5. The vehicles 5 include not only ordinary passenger cars but also public vehicles such as route buses and emergency vehicles. The vehicle 5 may be not only a four-wheeled vehicle but also a two-wheeled vehicle (motorcycle). The drive system of the vehicle 5 may be any of an engine drive, an electric motor drive, and a hybrid system. The driving method of the vehicle 5 may be any of normal driving in which a passenger performs operations such as acceleration / deceleration and steering of a steering wheel, and automatic driving in which the operation is executed by software.

The communication terminal 1A of the vehicle 5 may be an existing wireless communication device of the vehicle 5, or may be a portable terminal carried by the passenger into the vehicle 5. The passenger's mobile terminal is temporarily connected to the in-vehicle LAN (Local Area Network) of the vehicle 5 to temporarily become a wireless communication device mounted on the vehicle.

The communication terminal 1B is a portable terminal (pedestrian terminal) carried by the pedestrian 7. The pedestrian 7 is a person who walks outdoors on roads and parking lots, and indoors in buildings and underground malls. The pedestrian 7 includes not only a person walking but also a person riding a bicycle having no power source.

The communication terminal 1C is a wireless communication device mounted on the roadside sensor 8. The roadside sensor 8 is an image-type vehicle sensor installed on a road, a security camera installed outdoors or indoors, and the like. The communication terminal 1D is a wireless communication device mounted on the traffic signal controller 9 at the intersection.

Service requirements for slices S1 to S4 are as follows. Delay times D1 to D4 allowed for slices S1 to S4 are defined such that D1 <D2 <D3 <D4. For example, D1 = 1 ms, D2 = 10 ms, D3 = 100 ms, and D4 = 1s. Data traffic C1 to C4 per predetermined period (for example, one day) allowed for slices S1 to S4 are defined so as to satisfy C1 <C2 <C3 <C4. For example, C1 = 20 GB, C2 = 100 GB, C3 = 2 TB, and C4 = 10 TB.

As described above, in the wireless communication system of FIG. 1, the direct wireless communication in the slice S1 (for example, “inter-vehicle communication” in which the communication terminal 1A of the vehicle 5 directly communicates) and the slice through the base station 2 Wireless communication of S2 is possible. However, in the present embodiment, an information providing service for users included in a relatively wide service area (for example, an area including municipalities and prefectures) using the slices S3 and S4 in the wireless communication system of FIG. 1 is provided. is assumed.

[Internal configuration of edge server and core server]
FIG. 2 is a block diagram illustrating an example of an internal configuration of the edge server 3 and the core server 4. With reference to FIG. 2, the edge server 3 includes a control unit 31 including a CPU (Central Processing Unit), a ROM (Read Only Memory) 32, a RAM (Random Access Memory) 33, a storage unit 34, and a communication unit 35. including.

The control unit 31 controls the operation of each hardware by reading out and executing one or a plurality of programs stored in the ROM 32 in advance in the RAM 33, so that the computer can communicate with the core server 4, the base station 2, and the like. It functions as the edge server 3.

The RAM 33 is composed of a volatile memory element such as an SRAM (Static RAM) or a DRAM (Dynamic RAM), and temporarily stores a program executed by the control unit 31 and data necessary for the execution.

The storage unit 34 is configured by a nonvolatile memory element such as a flash memory or an EEPROM (Electrically Erasable Programmable Read Only Only Memory), or a magnetic storage device such as a hard disk.

The communication unit 35 has a function of communicating with the core server 4, the base station 2, and the like via the metro network. The communication unit 35 transmits the information provided from the control unit 31 to an external device via the metro network, and provides the control unit 31 with the information received via the metro network.

The storage unit 34 stores a dynamic information map (hereinafter, also simply referred to as a “map”) M1 as dynamic map information. The map M1 is an aggregate (virtual database) of data in which dynamic information that changes every moment is superimposed on a high-definition digital map that is static information.
The information constituting the map M1 includes the following “dynamic information” and “static information”.

“Dynamic information” refers to dynamic data that requires a delay time within one second. For example, dynamic information includes position information and signal information of a moving object (vehicle and pedestrian, etc.) used as ITS (Intelligent Transport Systems) look-ahead information.
Note that the position information of the moving object included in the dynamic information does not have a wireless communication function in addition to the position information of the vehicle 5 and the pedestrian 7 capable of wireless communication by having the

communication terminals

1A and 1B. The position information of the vehicle 5 and the pedestrian 7 is also included.

"Static information" refers to static data with a delay of one month or less. For example, road surface information, lane information, three-dimensional structure data, and the like correspond to the static information.

The control unit 31 of the edge server 3 updates the dynamic information of the map M1 stored in the storage unit 34 at a predetermined update cycle (update processing). Specifically, the control unit 31 collects, from the communication terminals 1A to 1D, various types of sensor information acquired by the vehicle 5, the roadside sensor 8, and the like in the service area of the own device for each predetermined update cycle, The dynamic information of the map M1 is updated based on the collected sensor information.

When receiving the dynamic information request message from the

communication terminals

1A and 1B of the predetermined user, the control unit 31 sends the latest dynamic information to the

communication terminals

1A and 1B of the transmission source of the request message at every predetermined distribution cycle. Distribute (distribution processing).
The control unit 31 collects traffic information and weather information of various places in the service area from the traffic control center, the private weather service support center, and the like, and updates the dynamic information or the static information of the map M1 based on the collected information. May be.

2, core server 4 includes a control unit 41 including a CPU, a ROM 42, a RAM 43, a storage unit 44, and a communication unit 45.

The control unit 41 controls the operation of each hardware by reading and executing one or a plurality of programs stored in the ROM 32 in advance in the RAM 43, and functions as a core server 4 capable of communicating the computer device with the edge server 3. Let it.

The RAM 43 is composed of a volatile memory element such as an SRAM or a DRAM, and temporarily stores a program executed by the control unit 41 and data necessary for the execution.

The storage unit 44 is configured by a nonvolatile memory element such as a flash memory or an EEPROM, or a magnetic storage device such as a hard disk.

The communication unit 45 has a function of communicating with the edge server 3, the base station 2, and the like via the core network. The communication unit 45 transmits the information provided from the control unit 41 to an external device via the core network, and provides the control unit 41 with the information received via the core network.

(2) As shown in FIG. 2, the storage unit 44 of the core server 4 stores an information map M2. The data structure of the map M2 (data structure including dynamic information and static information) is the same as the data structure of the map M1. The map M2 may be a map of the same service area as the map M1 of the specific edge server 3, or may be a wider area map obtained by integrating the maps M1 held by the plurality of edge servers 3.

Like the edge server 3, the control unit 41 of the core server 4 updates the dynamic information of the map M2 stored in the storage unit 44, and distributes the dynamic information in response to the request message. And may be performed. That is, the control unit 41 can perform the update process and the distribution process based on the map M2 of the own device independently of the edge server 3.
That is, the control unit 41 can independently execute the update process and the distribution process of the dynamic information based on the map M2 of the own device, separately from the edge server 3.

However, the core server 4 belonging to the slice S4 has a longer communication delay time with the communication terminals 1A to 1D than the edge server 3 belonging to the slice S3.
For this reason, even if the core server 4 independently updates the dynamic information of the map M2, the real-time property is inferior to the dynamic information of the map M1 managed by the edge server 3.
Therefore, for example, the control unit 31 of the edge server 3 and the control unit 41 of the core server 4 may perform the update processing and the distribution processing of the dynamic information in a distributed manner according to the priority defined for each predetermined area. .

[Internal configuration of in-vehicle device]
FIG. 3 is a block diagram illustrating an example of an internal configuration of the vehicle-mounted device 50 of the vehicle 5 equipped with the communication terminal 1A.
Referring to FIG. 3, on-vehicle device 50 mounted on vehicle 5 includes a control unit (ECU: Electronic Control Unit) 51, a GPS receiver 52, a vehicle speed sensor 53, a gyro sensor 54, a storage unit 55, a display 56, and a speaker. 57, an input device 58, a vehicle-mounted camera 59, a radar sensor 60, and a communication unit 61.

The communication unit 61 is the above-described communication terminal 1A (a wireless communication device capable of performing communication conforming to 5G). Therefore, the in-vehicle device 50 of the vehicle 5 can communicate with the edge server 3 as a kind of mobile terminal belonging to the slice S3. Further, the in-vehicle device 50 of the vehicle 5 can communicate with the core server 4 as a kind of mobile terminal belonging to the slice S4.

The control unit 51 is a computer device that searches for a route of the vehicle 5 and controls other electronic devices 52 to 61. The control unit 51 obtains the vehicle position of the own vehicle from the GPS signals periodically acquired by the GPS receiver 52. Further, the control unit 51 complements the vehicle position and direction based on the input signals of the vehicle speed sensor 53 and the gyro sensor 54, and grasps the accurate current position and direction of the vehicle 5.

The GPS receiver 52, the vehicle speed sensor 53, and the gyro sensor 54 are sensors for measuring the current position, speed, and direction of the vehicle 5.
Storage unit 55 includes a map database. The map database provides the control unit 51 with road map data. The road map data includes link data and node data, and is stored in a recording medium such as a DVD, CD-ROM, memory card, or HDD. The storage unit 55 reads necessary road map data from the recording medium and provides the read road map data to the control unit 51.

The display 56 and the speaker 57 are output devices for notifying the user who is the occupant of the vehicle 5 of various information generated by the control unit 51. Specifically, the display 56 displays an input screen at the time of a route search, a map image around the own vehicle, route information to a destination, and the like. The speaker 57 outputs an announcement or the like for guiding the vehicle 5 to the destination by voice. These output devices can also notify the passenger of the provided information received by the communication unit 61.

The input device 58 is a device for the passenger of the vehicle 5 to perform various input operations. The input device 58 is an operation switch provided on a steering wheel, a touch panel provided on the joystick display 56, a combination thereof, or the like. The input device 58 may be a voice recognition device that receives an input by voice recognition of a passenger. The input signal generated by the input device 58 is transmitted to the control unit 51.

The on-vehicle camera 59 is an image sensor that captures an image in front of the vehicle 5. The in-vehicle camera 59 may be either a monocular or a compound eye. The radar sensor 60 is a sensor that detects an object existing in front of or around the vehicle 5 using a millimeter-wave radar, a LiDAR method, or the like.
The control unit 51 executes a driving support control to output an alert to the occupant during driving on the display 56 or to perform a forced brake intervention based on the measurement data by the on-vehicle camera 59 and the radar sensor 60. be able to.

The control unit 51 is configured by an arithmetic processing device such as a microcomputer that executes various control programs stored in the storage unit 55.
The control unit 51 has a function of displaying a map image on the display 56 and a route from the departure place to the destination (including the position of the stopover point, if any) as functions realized by executing the control program. And a function of guiding the vehicle 5 to the destination according to the calculated route.

The control unit 51 performs an object recognition process of recognizing an object in front of or around the own vehicle based on at least one measurement data of the in-vehicle camera 59 and the radar sensor 60, and calculates a distance to the recognized object. It has a function of performing distance processing.
The control unit 51 can calculate the position information of the object recognized by the object recognition process from the distance calculated by the distance measurement process and the sensor position of the own vehicle.

The control unit 51 can execute the following processes in communication with the edge server 3 (or the core server 4).
1) Request message transmission processing 2) Dynamic information reception processing 3) Change point information generation processing 4) Change point information transmission processing

The request message transmission process is a process of transmitting, to the edge server 3, a control packet requesting distribution of the dynamic information of the map M1 sequentially updated by the edge server 3. The control packet includes the vehicle ID of the own vehicle.
When receiving the request message including the predetermined vehicle ID, the edge server 3 distributes the dynamic information at a predetermined distribution cycle to the communication terminal 1A of the vehicle 5 having the transmission source vehicle ID.

The dynamic information receiving process is a process of receiving dynamic information distributed by the edge server 3 to the own device.
The generation process of the change point information by the control unit 51 calculates the change between the received dynamic information and the sensor information of the own vehicle at the time of reception, based on the comparison result, and obtains information on the difference between the two information. This is the process of generating change point information.
The change point information generated by the control unit 51 is, for example, the following information examples a1 to a2.

Information example a1: Change point information related to a recognized object The control unit 51 detects an object X (a moving object such as a vehicle or a pedestrian or an obstacle) not included in the received dynamic information by its own object recognition processing. In this case, the image data and the position information of the detected object X are used as change point information. If the position information of the object X included in the received dynamic information and the position information of the object X obtained by its own object recognition processing are shifted by a predetermined threshold or more, the control unit 51 And the difference value between the two pieces of position information are set as change point information.

Information Example a2: Change Point Information Regarding the Own Vehicle The control unit 51 shifts the position information of the own vehicle included in the received dynamic information from the vehicle position of the own vehicle calculated by the GPS signal by a predetermined threshold or more. In this case, the difference value between the two is used as change point information. When the azimuth of the own vehicle included in the received dynamic information and the azimuth of the own vehicle calculated from the measurement data of the gyro sensor 54 deviate by a predetermined threshold or more, the control unit 51 determines the difference between the two. The value is used as change point information.

When generating the change point information as described above, the control unit 51 generates a communication packet addressed to the edge server 3 including the generated change point information and the vehicle ID of the own vehicle.
The transmission process of the change point information is a process of transmitting the communication packet including the change point information to the edge server 3. The transmission process of the change point information is performed within the dynamic information distribution cycle by the edge server 3.

Based on the dynamic information received from the edge server 3 or the like, the control unit 51 outputs a warning to the occupant during driving on the display 56 or executes driving support control for forcibly intervening braking. You can also.

(Internal configuration of pedestrian terminal)
FIG. 4 is a block diagram illustrating an example of an internal configuration of the pedestrian terminal 70 (the communication terminal 1B).
The pedestrian terminal 70 in FIG. 4 is a wireless communication device capable of performing communication processing conforming to, for example, 5G. Therefore, the pedestrian terminal 70 can communicate with the edge server 3 as a kind of mobile terminal belonging to the slice S3. Further, the pedestrian terminal 70 can also communicate with the core server 4 as a kind of mobile terminal belonging to the slice S4.

を Referring to FIG. 4, pedestrian terminal 70 includes a control unit 71, a storage unit 72, a display unit 73, an operation unit 74, and a communication unit 75.

The communication unit 75 is a communication interface that performs wireless communication with the base station 2 of the carrier that provides the 5G service. The communication unit 75 converts the RF signal from the base station 2 into a digital signal and outputs the digital signal to the control unit 71. The communication unit 75 converts the digital signal input from the control unit 71 into an RF signal, and transmits the RF signal to the base station 2.

The control unit 71 includes a CPU, a ROM, and a RAM. The control unit 71 reads out and executes the program stored in the storage unit 72, and controls the entire operation of the pedestrian terminal 70.

The storage unit 72 includes a hard disk, a nonvolatile memory, and the like, and stores various computer programs and data. The storage unit 72 stores a mobile ID, which is identification information of the pedestrian terminal 70. The mobile ID is, for example, a unique user ID or MAC address of a carrier contractor.

(4) The storage unit 72 stores various application software arbitrarily installed by the user. The application software stored in the storage unit 72 includes, for example, application software for enjoying an information providing service for receiving dynamic information of the map M1 through communication with the edge server 3 (or the core server 4). .

The operation unit 74 includes various operation buttons and a touch panel function of the display unit 73. The operation unit 74 outputs an operation signal according to a user operation to the control unit 71.

The display unit 73 is, for example, a liquid crystal display. The display unit 73 presents various information to the user. For example, the display unit 73 displays the image data of the information maps M1 and M2 transmitted from the

servers

3 and 4 on a screen.

The control unit 71 has a time synchronization function of acquiring the current time from the GPS signal, a position detection function of measuring the current position (latitude, longitude, and altitude) of the vehicle from the GPS signal, and detects the direction of the pedestrian 7 by using the direction sensor. And a azimuth detecting function for measuring.

The control unit 71 can execute the following processes in communication with the edge server 3 (may be the core server 4).
1) Request message transmission processing 2) Dynamic information reception processing 3) Change point information generation processing 4) Change point information transmission processing

The request message transmission process is a process of transmitting, to the edge server 3, a control packet requesting distribution of the dynamic information of the map M1 sequentially updated by the edge server 3. The control packet includes the mobile ID of the pedestrian terminal 70.
When receiving the request message including the predetermined mobile ID, the edge server 3 distributes the dynamic information to the communication terminal 1B of the pedestrian 7 having the transmission source mobile ID at a predetermined distribution cycle.

The dynamic information receiving process is a process of receiving dynamic information distributed by the edge server 3 to the own device.
The generation process of the change point information by the control unit 71 calculates the change between the received dynamic information and the sensor information of the own vehicle at the time of reception, based on the comparison result, and obtains information on the difference between the two information. This is the process of generating change point information.
The change point information generated by the control unit 71 is, for example, the following information example.

Information example: change point information regarding the own vehicle The control unit 71 determines that the position information of the own pedestrian 7 included in the received dynamic information and the position of the own pedestrian 7 calculated by the GPS signal are a predetermined threshold value. If there is a deviation, the difference between the two is used as change point information. When the direction of the self-pedestrian 7 included in the received dynamic information and the direction of the self-pedestrian 7 calculated by the direction sensor deviate by a predetermined threshold or more, the control unit 71 determines the difference value between the two. Change point information.

When generating the change point information as described above, the control unit 71 generates a communication packet addressed to the edge server 3 including the generated change point information and the mobile ID of the own terminal 70.
The transmission process of the change point information is a process of transmitting the communication packet including the change point information to the edge server 3. The transmission process of the change point information is performed within the dynamic information distribution cycle by the edge server 3.

As described above, the control unit 71 transmits the state information including the position and orientation information of the own terminal 70 to the edge server 3 by performing the change point information generation processing and the change point information transmission processing.

[Internal configuration of roadside sensor]
FIG. 5 is a block diagram illustrating an example of an internal configuration of the roadside sensor 8 including the wireless communication device as the communication terminal 1C. Referring to FIG. 5, roadside sensor 8 includes a control unit 81, a storage unit 82, a roadside camera 83, a radar sensor 84, and a communication unit 85.

The communication unit 85 is the above-described communication terminal 1C, that is, a wireless communication device capable of performing communication processing conforming to, for example, 5G. Therefore, the roadside sensor 8 can communicate with the edge server 3 as a kind of fixed terminal belonging to the slice S3. The roadside sensor 8 can also communicate with the core server 4 as a kind of fixed terminal belonging to the slice S4.

The control unit 81 includes a CPU, a ROM, and a RAM. The control unit 81 reads and executes the program stored in the storage unit 82, and controls the entire operation of the roadside sensor 8.

The storage unit 82 includes a hard disk, a nonvolatile memory, and the like, and stores various computer programs and data. Further, the storage unit 82 stores a sensor ID which is identification information of the roadside sensor 8. The sensor ID is, for example, a user ID or a MAC address unique to the owner of the roadside sensor 8.

The roadside camera 83 is an image sensor that captures an image of a predetermined shooting area. The roadside camera 83 may be a single eye or a compound eye. The radar sensor 60 is a sensor that detects an object existing in front of or around the vehicle 5 using a millimeter-wave radar, a LiDAR method, or the like.

When the roadside sensor 8 is a security camera, the control unit 81 transmits the captured video data and the like to the computer device of the security administrator. When the roadside sensor 8 is an image type vehicle sensor, the control unit 81 transmits the captured image data and the like to the traffic control center.

The control unit 81 performs an object recognition process of recognizing an object in the shooting area based on at least one measurement data of the roadside camera 83 and the radar sensor 84, and a distance measurement process of calculating a distance to the recognized object. , Has the function of executing. The control unit 51 can calculate the position information of the object recognized by the object recognition process from the distance calculated by the distance measurement process and the sensor position of the own vehicle.

The control unit 81 can execute the following processes in communication with the edge server 3 (may be the core server 4).
1) Change point information generation processing 2) Change point information transmission processing

The generation process of the change point information in the roadside sensor 8 is performed based on a comparison result between the previous sensor information and the current sensor information for each predetermined measurement cycle (for example, the distribution cycle of the dynamic information by the edge server 3). This is a process of calculating a change between sensor information and generating change point information that is information on a difference between the two information.
The change point information generated by the roadside sensor 8 is, for example, the following information example b1.

Information example b1: change point information related to a recognized object The control unit 81 detects an object Y (a moving object such as a vehicle or a pedestrian or an obstacle) that was not detected in the previous object recognition process by the current object recognition process. In this case, the detected image data and position information of the object Y are used as change point information.
When the position information of the object Y obtained by the previous object recognition process and the position information of the object Y obtained by the current object recognition process are shifted by a predetermined threshold or more, the control unit 81 And the difference value between them are defined as change point information.

When generating the change point information as described above, the control unit 81 generates a communication packet addressed to the edge server 3 including the generated change point information and the sensor ID of the own device.
The transmission process of the change point information is a process of transmitting the communication packet including the change point information in the data to the edge server 3. The transmission process of the change point information is performed within the dynamic information distribution cycle by the edge server 3.

[Overall configuration of information provision system]
FIG. 6 is an overall configuration diagram of the information providing system according to the present embodiment.
Referring to FIG. 6, the information providing system according to the present embodiment includes a large number of vehicles 5, pedestrian terminals 70, and roadside sensors 8 scattered in a service area (real world) of edge server 3 which is relatively wide. And an edge server 3 that can perform wireless communication with low delay by 5G-compliant communication and the like performed via these communication nodes and the base station 2 and that functions as an information providing device. That is, the information providing system is configured to include part or all of the above-described wireless communication system.
In addition, the mobile object existing in the service area of the edge server 3 includes the vehicle 5 capable of wireless communication by mounting the communication terminal 1A and the in-vehicle device 50, and the pedestrian 7 carrying the pedestrian terminal 70. And a pedestrian 7 who does not carry the pedestrian terminal 70 without a wireless communication function.

The edge server 3 collects the above-mentioned change point information at a predetermined cycle from the in-vehicle device 50 of the vehicle 5 in the service area, the pedestrian terminal 70, the roadside sensor 8, and the like (step S31).
The edge server 3 integrates the collected change point information by map matching (integration processing), and updates the dynamic information of the information map M1 being managed (step S32).

If there is a request from the in-vehicle device 50 or the pedestrian terminal 70 of the vehicle 5, the edge server 3 transmits the latest dynamic information to the requesting communication node (step S33). Thus, for example, the vehicle 5 that has received the dynamic information can utilize the dynamic information for driving assistance of a passenger and the like.
The edge server 3 may transmit the map M1 updated in step S32 to the requesting communication node as dynamic information.

When the vehicle 5 receiving the dynamic information detects the change point information from the sensor information of the own vehicle based on the dynamic information, the vehicle 5 transmits the detected change point information to the edge server 3 (Step S34).

As described above, in the information providing system of the present embodiment, collection of change point information (step S31) → update of dynamic information (step S32) → distribution of dynamic information (step S33) → detection of change point information by vehicle The information processing in each communication node circulates in the order of (step S34) → collection of change point information (step S31).

FIG. 6 illustrates an information providing system including only one edge server 3, but the information providing system may include a plurality of edge servers 3, or may be used instead of the edge server 3 or in addition to the edge server 3. One or more core servers 4 may be included.
The information map M1 managed by the edge server 3 may be any map as long as at least dynamic information of an object is superimposed on map information such as a digital map. This is the same for the information map M2 of the core server.

[Dynamic information update processing and distribution processing]
FIG. 7 is a sequence diagram illustrating an example of a dynamic information update process and a distribution process performed by cooperation of the pedestrian terminal 70, the vehicle-mounted device 50 of the vehicle 5, the roadside sensor 8, and the edge server 3.
In the following description, the execution subject is the pedestrian terminal 70, the in-vehicle device 50 of the vehicle 5, the roadside sensor 8, and the edge server 3, but the actual execution subject is those

control units

71, 51, 81, 31. It is.
.. In FIG. 7 are the distribution periods of the dynamic information.

Referring to FIG. 7, when receiving a request message for dynamic information from pedestrian terminal 70 and on-vehicle device 50 of vehicle 5 (step S1), edge server 3 transmits the latest dynamic information at the time of reception to the transmission source. To the pedestrian terminal 70 and the in-vehicle device 50 of the vehicle 5 (step S2).

In step S1, the edge server 3 analyzes the received request message. If the information indicating the request source included in the message is information indicating the communication terminal 1 registered in advance, the edge server 3 determines that the request message has been transmitted. Alternatively, dynamic information may be transmitted.

In step S1, if there is a request message from either the pedestrian terminal 70 or the in-vehicle device 50, in step S2, the dynamic information is distributed to only one of the communication terminals that transmitted the request message. Is done.

When the pedestrian terminal 70 that has received the dynamic information distributed in step S2 generates the change point information within the distribution cycle U1 (step S3), the pedestrian terminal 70 transmits the generated change point information to the edge server 3 (step S6). ).
When the in-vehicle device 50 that has received the dynamic information distributed in step S2 generates change point information from the comparison result between the dynamic information and its own sensor information within the distribution cycle U1 (step S4), the generated change The point information is transmitted to the edge server 3 (Step S6).
When the roadside sensor 8 generates the change point information of its own sensor information within the distribution cycle U1, the roadside sensor 8 transmits the generated change point information to the edge server 3 (step S6).

When the edge server 3 receives the change point information from the pedestrian terminal 70, the in-vehicle device 50, and the roadside sensor 8 within the update period U1, the edge server 3 updates the change information with dynamic information reflecting the change point information (Step S7). The updated dynamic information is distributed to the pedestrian terminal 70 and the in-vehicle device 50 (Step S8).

For example, when only the in-vehicle device 50 generates the change point information within the distribution cycle U1, only the change point information generated by the in-vehicle device 50 in step S4 is transmitted to the edge server 3 (step S6), and the change point Update of the dynamic information reflecting only the information is performed (step S7).
If the pedestrian terminal 70, the in-vehicle device 50, and the roadside sensor 8 do not perform the change point information within the distribution cycle U1, the processes of steps S3 to S7 are not performed, and the dynamic information of the previous transmission is not performed. The same dynamic information as in (Step S2) is distributed to the pedestrian terminal 70 and the in-vehicle device 50 (Step S8).
Thus, the edge server 3 updates the dynamic information in step S7 based on the change point information transmitted within the distribution cycle U1.

The pedestrian terminal 70 that has received the dynamic information distributed in step S8 generates the change point information within the distribution cycle U2 (step S9), and transmits the generated change point information to the edge server 3 (step S12). ).
The in-vehicle device 50 that has received the dynamic information distributed in step S8 generates change point information from the comparison result between the dynamic information and its own sensor information within the distribution cycle U2 (step S10). The point information is transmitted to the edge server 3 (Step S12).
When the roadside sensor 8 generates the change point information of its own sensor information within the distribution cycle U2, the roadside sensor 8 transmits the generated change point information to the edge server 3 (step S12).

When receiving the change point information from the vehicle-mounted device 50 and the roadside sensor 8 within the distribution cycle U2, the edge server 3 updates the change information with dynamic information reflecting the change point information (Step S13), and updates the updated dynamic information. Is distributed to the pedestrian terminal 70 and the in-vehicle device 50 (step S14).
Thus, the edge server 3 updates the dynamic information in step S13 based on the change point information transmitted within the distribution cycle U2.

The processing after step S14 is performed until a request message for stopping distribution of dynamic information is received from both the pedestrian terminal 70 and the vehicle 5 or the communication between the pedestrian terminal 70 and the vehicle 5 is interrupted. Is repeated by the same sequence as.

[About the first embodiment]
[Provision of information to mobile terminals]
The information providing system according to the present embodiment has a function of providing a collision prediction result to the pedestrian terminal 70 mounted on one or a plurality of moving objects located in the service area and the vehicle-mounted device 50 of the vehicle 5. ing. Note that the collision prediction result is information indicating a prediction result when predicting whether or not a mobile body including the in-vehicle device 50 or the pedestrian terminal 70 collides with another mobile body other than the mobile body. is there. The collision prediction result includes information such as the attribute of the moving body predicted to collide, the direction in which the moving body predicted to collide approaches, the collision prediction time, and the like.

FIG. 8 is a functional block diagram of the edge server 3 showing a function for providing a collision prediction result according to the first embodiment.
The control unit 31 of the edge server 3 functionally includes a calculation unit 31a, a determination unit 31b, a notification unit 31c, and a detection unit 31d. Each of these functions is realized by the control unit 31 executing a program stored in the storage unit 34.

The calculation unit 31a is configured to perform a mobile terminal (pedestrian terminal 70 and vehicle-mounted) mounted on a moving object (pedestrian 7, vehicle 5) located in the service area represented by the dynamic information map M1 based on the dynamic information map M1. Each of the devices 50) has a function of performing a collision prediction and obtaining an evaluation value of the collision prediction result. The calculation unit 31a refers to the mobile object database 34a (described later) in which information obtained from the dynamic information map M1 is registered, and obtains an evaluation value for a collision prediction result.

The determining unit 31b has a function of determining whether or not to notify a collision prediction result of a mobile object equipped with a mobile terminal (the pedestrian terminal 70 and the in-vehicle device 50) to the mobile terminal based on the evaluation value. doing.
The notifying unit 31c has a function of notifying the mobile terminal of a collision prediction result based on the determination result of the determining unit 31b.
The detection unit 31d detects a plurality of moving objects (pedestrians 7, vehicles 5) in which the position information is included in the dynamic information of the dynamic information map M1, and generates moving object information indicating the status of each moving object. It has the function to do.

The storage unit 34 stores a mobile object database 34a and an evaluation value database 34b in addition to the above-described dynamic information map M1.
FIG. 9 is a diagram illustrating an example of the mobile object database 34a.
As shown in FIG. 9, mobile object information generated by the detection unit 31d is registered in the mobile object database 34a.
The mobile database 34a is managed and updated by the detection unit 31d.
The detecting unit 31d refers to the dynamic information map M1, and when the position information of the new moving object (pedestrian 7, vehicle 5) is registered in the dynamic information, the detecting unit 31d assigns the moving object ID to the moving object. The moving body information corresponding to the moving body ID is generated and registered in the moving body database 34a. As described above, the detection unit 31d detects a moving object whose position information is registered in the dynamic information map M1, assigns a moving object ID to each moving object, and generates moving object information.

The moving object information includes information such as information on the presence or absence of a communication function, a vehicle ID (mobile ID), attribute information of the moving object, position information, azimuth information indicating a moving direction, and speed information indicating a moving speed. The attribute information is information indicating an attribute of each moving object. The attributes of the moving object include a plurality of attributes such as an attribute related to the appearance of the moving object and an attribute related to the behavior of the moving object. Note that the attribute information is distinguished between a vehicle and a pedestrian. The attribute information includes information indicating the attribute content of each of the plurality of attributes in the moving object.

The mobile object database 34a has columns for registering the mobile object ID, information on the presence or absence of a communication function, a vehicle ID (mobile ID), attribute information of the mobile object, position information, azimuth information, and speed information. ing.
Each piece of information in the moving object database 34a is registered in association with the moving object ID.

The detection unit 31d (acquisition unit) refers to the dynamic information map M1 and generates moving body information of each moving body. The detection unit 31d determines, among the moving object information, information on the presence or absence of a communication function included in the dynamic information map M1, a vehicle ID (mobile ID), and position information, information included in the dynamic map information M1. Is acquired as it is and registered in the mobile object database 34a.
The detecting unit 31d calculates the azimuth information and the speed information based on the change over time of the position information of each moving object included in the dynamic information map M1.

FIGS. 10 and 11 are diagrams showing a part of a part for registering attribute information in the mobile object database 34a. FIG. 10 shows a column for registering attribute information when the moving object is a vehicle. FIG. 11 shows a column for registering attribute information when the moving object is a pedestrian.

As shown in FIG. 10, the attribute information of the vehicle includes an appearance attribute and an action attribute.
The appearance attributes include four types of attributes of “size”, “shape”, “color”, and “display license plate”. The contents of these attributes are set in advance, and the attribute contents (appearance information) corresponding to the target vehicle are registered in association with the moving object ID from among a plurality of preset attribute contents.

The attribute content in “size” includes large, medium, and small. Buses and trucks are determined to be large, light vehicles are determined to be small, and other general passenger vehicles are determined to be medium.
The attribute contents in the “shape” include a sedan, a coupe, a wagon, a minivan, an SUV, a convertible car, and a truck.
The attribute contents of “color” include high brightness and low brightness.
The attribute content in the “display of license plate” includes the display of the place name of the license plate attached to the target vehicle 5 (the place name of the transportation branch office that controls the home of use of the vehicle).

The detection unit 31d refers to the image data of the moving object captured by the camera or the like included in the dynamic information map M1, and acquires the information on the appearance attribute of the vehicle 5 described above.

中 In FIG. 10, the behavior attribute includes a past traveling history. More specifically, the behavior attributes include four numbers of “past lane change number”, “past sudden steering sharp brake number”, “lane change number after warning”, and “fast steering sharp brake number after warning”. Contains type attributes. The “number of past lane changes” indicates the number of times the target vehicle 5 has changed lanes during a fixed period in the past. The “number of past sudden steering wheel sudden brakes” indicates the number of times that the target vehicle 5 has performed the sudden steering operation or the sudden braking operation in a past fixed period. The “number of lane changes after warning” indicates the number of times the vehicle 5 has changed lanes after a warning is issued to the in-vehicle device 50 of the target vehicle 5 by the edge server 3 described below. The “number of sudden steering wheel sudden brakes after the alarm” indicates the number of times the vehicle 5 performs the sudden steering wheel sudden braking operation after the alarm.

The detecting unit 31d stores past dynamic information (position information included therein) of each moving body in the storage unit 34 and generates trajectory information of each moving body that has moved in the past based on the past dynamic information. It has the function to do. The trajectory information also includes speed information of the moving object. Therefore, in the case of the vehicle 5, it is possible to detect that the sudden steering wheel sudden braking operation is performed based on the trajectory information. Further, based on the trajectory information, it is also possible to detect that a lane change that changes lanes when traveling straight ahead has been performed.

The detecting unit 31d detects a lane change and a sudden steering sudden braking operation by each vehicle 5 based on the trajectory information of each vehicle 5, and determines the number of times each vehicle 5 has changed lanes and the sudden steering sudden braking operation. It has a function to count the number of times it has been performed.

The edge server 3 has a function of notifying a warning to the in-vehicle device 50 based on the number of times the lane change counted by the detection unit 31d and the number of times of the sudden steering wheel sudden braking operation are performed. . The edge server 3 notifies the in-vehicle device 50 that has performed lane change and sudden steering sudden braking operation a certain number of times or more within a predetermined period.
The detection unit 31d also counts, for the in-vehicle device 50 that has been notified of the alarm, the number of times that each vehicle 5 has changed lanes and the number of times that the driver has performed a sudden steering wheel sudden braking operation after the alarm.

Based on these functions, the detecting unit 31d uses the information on “the number of past lane changes”, “the number of past sudden steering sudden brakes”, “the number of lane changes after warning”, and “the number of sudden steering sudden brakes after warning” ( Behavior information).

As shown in FIG. 11, the attribute information of the pedestrian also includes an appearance attribute and an action attribute, like the attribute information of the vehicle.
The appearance attributes include four types of attributes of “height”, “clothes”, “earphones or aside”, and “crutches or wheelchairs”.
The contents of the attributes of "clothes", "earphones or aside", and "crutches or wheelchairs" are set in advance, and an attribute corresponding to the target pedestrian from a plurality of preset attribute contents (appearance information). The contents are registered.

The height of the pedestrian is registered in the attribute content of “height”. The attribute contents in “clothes” include students and non-students. If the attribute content of “clothes” is a student, it indicates that the pedestrian's clothes are wearing school uniforms and hats. If the attribute content of "clothes" is other than student, it indicates that the pedestrian's clothes are not school uniforms.
The attribute content in “earphones or aside” includes applicable and non-applicable. If the attribute content in “earphones or looking away” is applicable, it indicates that the pedestrian is wearing earphones or looking at a terminal such as a smartphone while walking. When the attribute content of “earphone or side view” is not applicable, it indicates that the user is walking normally.
The attribute content in “crutch or wheelchair” includes applicable and non-applicable. When the attribute content in “crutches or wheelchairs” is applicable, it indicates that the pedestrian is using crutches or wheelchairs. If the attribute content of “Crutch or Wheelchair” is not applicable, it indicates that the user is walking normally.

中 In FIG. 11, the behavior attributes include four types of attributes, “walking place”, “ignore signal”, and “walking locus”. The attribute content of “walking place” includes a road, a sidewalk, and others. When the attribute content of “walking place” is a road, it indicates that the target pedestrian 7 is walking on the road. When the attribute content in “walking place” is a sidewalk, it indicates that the target pedestrian 7 is walking on a sidewalk or a crosswalk. If the attribute content of “walking place” is “other”, it indicates that the target pedestrian 7 is walking in a building or the like.

The attribute content in “ignore signal” includes applicable and non-applicable. If the attribute content in “ignore signal” is applicable, it indicates that the signal is currently ignored. If the attribute content in "ignore signal" is not applicable, it indicates that the signal is not currently ignored.
The attribute content in the “walking locus” includes straight and meandering. If the attribute content in the “walking locus” is meandering, it indicates that the nearest walking locus of the target pedestrian 7 is meandering. When the attribute content in the “walking locus” is straight, it indicates that the nearest walking locus of the target pedestrian 7 is straight.

The detecting unit 31d refers to the image data of the moving object captured by the camera or the like included in the dynamic information map M1 and refers to the above-described trajectory information, and obtains information (behavior information) regarding each attribute of each pedestrian. get.
Note that when the dynamic information map M1 does not include image data of a moving object (vehicle and pedestrian), the detection unit 31d determines the image data and the like included in the change point information for updating the dynamic information map M1. Can be used.
The detecting unit 31d generates attribute information of each moving object based on the acquired information on the attributes.

The detection unit 31d repeatedly generates the moving body information of each moving body and registers the moving body information in the moving body database 34a, and updates the moving body database 34a as needed. As a result, the moving object information registered in the moving object database 34a is maintained at the latest information.

The evaluation value database 34b in FIG. 8 is a database for registering the evaluation value of the collision prediction result obtained by the calculation unit 31a. The evaluation value database 34b will be described later.

[Evaluation value calculation processing]
FIG. 12 is a flowchart illustrating an example of the calculation process of the evaluation value of the collision prediction result by the calculation unit 31a.
As shown in FIG. 12, first, the calculation unit 31a reads the mobile object database 34a (step S51), and specifies a mobile object having a communication function as an evaluation target (step S52). The collision prediction result can be provided to a moving object having the pedestrian terminal 70 or the in-vehicle device 50. Therefore, the calculation unit 31a specifies a moving object having the pedestrian terminal 70 and the in-vehicle device 50 as an evaluation target for obtaining an evaluation value.
That is, the evaluation target indicates a mobile terminal for which an evaluation value is obtained by the calculation unit 31a.
In the following description, the evaluation target (target moving object) is a pedestrian terminal 70 or a pedestrian 7 or a vehicle 5 which is a moving object having the in-vehicle device 50, or is carried or mounted on a moving object. Pedestrian terminal 70 or on-vehicle device 50 that is running. In addition, the vehicle 5 on which a person carrying the pedestrian terminal 70 boards is also included in the evaluation target.

Next, the calculation unit 31a performs an evaluation value calculation process (Step S53).
In the evaluation value calculation processing, the calculation unit 31a sequentially executes the processing on each of the specified evaluation targets, and repeats the processing until the processing is performed on all of the evaluation targets.

First, the calculation unit 31a proceeds to step S55, and performs an individual calculation process. The calculation unit 31a obtains an evaluation value to be evaluated in the individual calculation process (Step S55).

FIG. 13 is a flowchart illustrating an example of the individual calculation process in step S55 in FIG.
As illustrated in FIG. 13, the calculation unit 31a specifies a mobile body (another mobile body) other than the evaluation target (Step S61), and calculates a collision prediction time between the evaluation target (the target mobile body) and another mobile body. Is calculated for each of the other moving objects (step S62).
It should be noted that other moving objects include moving objects equipped with the pedestrian terminal 70 or the vehicle-mounted device 50.

The calculation unit 31a refers to the mobile object database 34a and, based on the position information, the azimuth information, and the speed information of the evaluation object and the other mobile objects, a collision when the evaluation object collides with another mobile object. Find the estimated time.
When there is no possibility of collision from the position information and the azimuth information, the calculation unit 31a sets the collision prediction time to an extremely large predetermined value (for example, 5 minutes). The calculation unit 31a also sets the collision prediction time to the predetermined value when the calculated collision prediction time is equal to or longer than the predetermined value.

As described above, the calculation unit 31a predicts a collision between the evaluation target and another moving body based on the dynamic information map M1, and obtains a collision prediction time.

The calculation unit 31a specifies, as the collision prediction target, the moving object having the shortest collision prediction time among the collision prediction times calculated for the other moving objects (step S63).
Here, when the possibility of a collision is determined based on the predicted collision time, it can be determined that the moving object having the shortest predicted collision time has the highest possibility of colliding with the evaluation target. Therefore, the calculation unit 31a specifies a moving object having the shortest collision prediction time as a collision prediction target.
As a result, the calculation unit 31a performs a collision prediction on each of the other moving objects, and obtains a collision prediction result.

Next, the calculation unit 31a obtains a base value for obtaining an evaluation value of the collision prediction result based on the calculated collision prediction time (Step S64).
The base value is a value serving as a basis for obtaining an evaluation value of the collision prediction result, and an evaluation value is obtained by adding an additional value to the base value in a later process.

The calculation unit 31a obtains a base value for the predicted collision time according to the following rule. For example, the settings are as follows, but are not limited thereto.
Collision prediction time is 10 seconds or more: Base value = 0
Collision prediction time less than 10 seconds: Base value = 20
Collision prediction time less than 5 seconds: Base value = 100
Collision prediction time less than 3 seconds: Base value = 200
Collision prediction time less than 1 second: Base value = 500
The evaluation value of the collision prediction result is set such that the larger the value, the higher the possibility of collision and the higher the need for notification. Accordingly, the base value is also set such that the larger the value, the higher the possibility of collision and the higher the need for notification.
That is, the evaluation value of the collision prediction result is a value indicating a possibility that the collision prediction target collides with a moving object other than the collision prediction target.
The calculation unit 31a sets “0” as a base value for a moving object other than the collision prediction target. The added value is not added to the base value of the moving object other than the collision prediction target by the subsequent processing. Therefore, the evaluation value of the collision prediction result of the moving object other than the collision prediction target (evaluation value of the collision prediction result between the target moving object and the moving object other than the collision prediction target) is “0”.
As described above, the calculation unit 31a obtains the base value for each moving object based on the predicted collision time.

Next, the calculation unit 31a proceeds to step S65, and determines whether there is a blind spot factor that causes a blind spot between the evaluation target and the collision prediction target (step S65).
The presence or absence of a blind spot factor between the evaluation target and the collision prediction target is determined by the calculation unit 31a by referring to the dynamic information map M1. The calculation unit 31a refers to the dynamic information map M1, and determines whether there is a building or another moving object that blocks the line of sight between the evaluation target and the collision prediction target. When such a building or another moving object exists between the evaluation target and the collision prediction target, the calculation unit 31a determines that there is a blind spot factor between the evaluation target and the collision prediction target. On the other hand, if there is no obstacle between the evaluation target and the collision prediction target, the calculation unit 31a determines that there is no blind spot factor.

If it is determined in step S65 that there is a blind spot factor between the evaluation target and the collision prediction target, the calculation unit 31a adds the base value obtained in step S64 (step S66), and proceeds to step S67.
On the other hand, when it is determined that there is no blind spot factor between the evaluation target and the collision prediction target (step S65), the calculation unit 31a proceeds to step S67.

As described above, the evaluation value is set such that the larger the value, the higher the factor that impairs the safety of the evaluation target.
When there is a blind spot factor between the evaluation target and the collision prediction target, safety is further impaired in the evaluation target. Therefore, when it is determined that there is a blind spot factor, the calculation unit 31a adds the base value.
The added value added to the base value in step S66 is, for example, "100". This addition value is an example and is not limited to this. The same applies to the following addition values.

As described above, the calculation unit 31a determines the presence or absence of a blind spot factor that causes a blind spot between the evaluation target and the collision prediction target, and adds the determination result of the presence or absence of the blind spot factor to the evaluation value.
In this case, the presence or absence of a blind spot factor can be reflected in the evaluation value, and can be reflected in the execution determination of providing information to the evaluation target described later.

Next, in step S67, the calculation unit 31a performs an addition process of adding a value to the basic value according to the content of the attribute of the collision prediction target (step S67).

FIG. 14 is a flowchart illustrating an example of the process of adding a collision prediction target in FIG.
First, the calculation unit 31a determines whether (the moving object of) the collision prediction target is a vehicle (step S70).
If it is determined that the collision prediction target is the vehicle 5, the calculation unit 31a proceeds to step S71, executes the vehicle addition processing, and ends the processing.

FIG. 15 is a flowchart showing an example of the vehicle addition process in FIG.
In the vehicle addition process, the calculation unit 31a first refers to the attribute information of the collision prediction target (vehicle 5) in the mobile object database 34a.
The calculation unit 31a performs addition regarding “size” of the attribute information (step S81).

An addition value is set in advance for each of the attribute contents of “size”, that is, large, medium, and small.
The arithmetic unit 31a adds “50” to the base value as an added value when the attribute content in “size” is large, and adds “30” to the base value as an added value when the attribute content is medium-sized. In some cases, “20” is added to the base value as an additional value.
Generally, it is considered that the greater the size of the vehicle 5, the greater the possibility of collision. Therefore, the addition value is set to increase as the size of the vehicle 5 increases.
As described above, the calculation unit 31a adds the added value corresponding to the “size” of the collision prediction target to the base value (Step S81).

Next, the calculation unit 31a performs addition regarding “shape” in the attribute information (step S82).
An addition value is set in advance for each of the attributes of the “shape”, that is, the sedan, coupe, wagon, minivan, SUV, open car, and truck. For example, when there is a difference in the accident rate or the like according to the shape of the vehicle 5, a larger added value can be set for the attribute content having a high accident rate.
For example, when the attribute content in the “shape” is coupe, open car, and truck, the addition value “30” is added, and when the attribute content is sedan, wagon, minivan, and SUV, the addition value “10” is added.
As described above, the calculation unit 31a adds the added value corresponding to the “shape” of the collision prediction target to the base value (Step S82).

Next, the calculation unit 31a performs addition regarding “color” in the attribute information (step S83).
An addition value is set in advance for each of the high brightness and the low brightness, which are the attribute contents of “color”. For example, when there is a difference in the accident rate or the like between the high brightness and the low brightness, it is possible to set the attribute content having the high accident rate to have a larger addition value.
For example, when the attribute content of “color” is high lightness, the added value is not added, and when the attribute content is “low lightness”, the added value “10” is added.
As described above, the calculation unit 31a adds the addition value corresponding to the “color” of the collision prediction target to the base value (Step S83).

Next, the calculation unit 31a determines whether or not the place name display in the “display of license plate” in the attribute information is not local (step S84).
If the place name display on the license plate is different from the place name of the jurisdiction of the transport branch office at the current position of the vehicle 5, the arithmetic unit 31a determines that the place name display is not local.
If it is determined in step S84 that the place name display on the license plate is not local, the calculation unit 31a adds the added value to the base value (step S85), and proceeds to step S86.
On the other hand, if it is determined in step S84 that the place name display on the license plate is local, the calculation unit 31a proceeds to step S86.
For example, if the attribute content in “display license plate” is local, the added value is not added, and if it is not local, the added value “10” is added.
Steps S81 to S85 are processing relating to appearance attributes, and step S86 is processing relating to action attributes.

Next, the calculation unit 31a performs addition on the past travel history that is the action attribute of the attribute information (step S86).
Each of the "past lane change count", "past sudden steering sharp brake count", "lane change count after warning", and "fast steering sharp brake count after warning" included in the behavior attribute includes A corresponding addition value is set in advance. As the added value set for these action attributes, a larger added value is set as the number of times, which is the attribute content, increases.
When a lane change or a sudden steering wheel sudden braking operation is frequently performed, it is conceivable that there is a high probability that an action having a high possibility of a collision is performed. For this reason, as the number of lane changes and the number of sudden steering brakes increase, a large addition value is set.

Further, between "the number of lane changes in the past" and "the number of lane changes after the warning", a relatively larger added value is set in the "number of lane changes after the warning".
Similarly, between "the number of sudden steering sudden brakes in the past" and "the number of sudden steering sudden brakes after the alarm", a relatively larger added value is set in the "number of sudden steering sudden brakes after the alert" Is done.
This is because, in the case where the number of lane changes or the sudden steering wheel sudden braking operation is repeated despite the notification of the warning from the edge server 3, it is considered that the probability of performing an action having a high possibility of collision is higher.

As described above, the calculation unit 31a adds the addition value according to the past traveling history to the base value (step S86), and ends the addition processing.
Here, the case where the addition value according to the past traveling history is added is illustrated, but the addition value according to the current traveling state such as the current traveling speed may be further added to the base value.

Returning to FIG. 14, when it is determined in step S70 that the collision prediction target is not the vehicle 5, the calculation unit 31a proceeds to step S72, executes a pedestrian addition process, and ends the process.

FIG. 16 is a flowchart showing an example of the pedestrian addition process in FIG.
In the pedestrian addition process, the calculation unit 31a first refers to the attribute information of the collision prediction target (pedestrian 7) in the mobile object database 34a.
The calculation unit 31a determines whether “height” of the attribute information is equal to or less than a predetermined value (step S91).
If it is determined in step S91 that the “height” is equal to or smaller than the predetermined value, the calculation unit 31a adds the added value to the base value (step S92), and proceeds to step S93.
On the other hand, if it is determined in step S91 that “height” is not smaller than or equal to the predetermined value, the calculation unit 31a proceeds to step S93.

The predetermined value is set to, for example, 120 cm. In this case, in step S91, it can be determined whether the pedestrian is a child younger than the lower grade of elementary school or a pedestrian other than that. Therefore, when the pedestrian is a child younger than the lower grade of elementary school, an additional value can be added. For example, the added value in step S92 is “20”.

Next, the calculation unit 31a determines whether or not the attribute content of “clothes” is a student (step S93).
If it is determined in step S93 that the attribute content of “clothes” is a student, the calculation unit 31a adds the added value to the base value (step S94), and proceeds to step S95.
On the other hand, if it is determined in step S93 that the attribute content of “clothes” is not a student, the calculation unit 31a proceeds to step S95.
In steps S93 and S94, when the pedestrian is a student, an additional value can be added. For example, the added value in step S94 is “10”.

Next, the calculation unit 31a determines whether or not the attribute content in “earphone or side-by-side” is applicable (step S95).
If it is determined in step S95 that the attribute content of “earphone or aside” is applicable, the calculation unit 31a adds the added value to the base value (step S96), and proceeds to step S97.
On the other hand, if it is determined in step S95 that the attribute content in “earphone or aside” is not applicable (not applicable), the operation unit 31a proceeds to step S97.
In steps S95 and S96, when the pedestrian is wearing earphones or looking at a terminal such as a smartphone while walking, the added value can be added. For example, the added value in step S96 is “10”.

Next, the calculation unit 31a determines whether or not the attribute content of “crucible or wheelchair” is applicable (step S97).
If it is determined in step S97 that the attribute content of “crutch or wheelchair” is applicable, the calculation unit 31a adds the added value to the base value (step S98), and proceeds to step S99.
On the other hand, if it is determined in step S97 that the attribute content of “crutch or wheelchair” is not applicable (not applicable), the calculation unit 31a proceeds to step S99.
In steps S97 and S98, when the pedestrian is using a crutch or a wheelchair, the added value can be added. For example, the added value in step S98 is “30”.
Steps S91 to S98 are processing relating to appearance attributes, and processing from step S99 is processing relating to action attributes.

The calculation unit 31a determines whether or not the attribute content of the “walking place” is a road (Step S99).
If it is determined in step S99 that the attribute content of the “walking place” is a road, the calculation unit 31a adds the added value to the base value (step S100), and proceeds to step S101.
On the other hand, if it is determined in step S99 that the attribute content of “walking place” is not a road, the calculation unit 31a proceeds to step S101.
In steps S100 and S101, when the pedestrian is walking on the road, the added value can be added. For example, the added value in step S100 is “30”.

Next, the calculation unit 31a determines whether the walking speed is lower than a predetermined value (Step S101).
The determination as to whether the walking speed of the pedestrian is lower than the predetermined value can be made by referring to the moving object information of the moving object database 34a.
The predetermined value to be compared with the walking speed is set, for example, every 3.6 km as a general pedestrian speed.

If it is determined in step S101 that the walking speed is lower than the predetermined value, the calculation unit 31a adds the added value to the base value (step S102), and proceeds to step S103.
On the other hand, if it is determined in step S101 that the walking speed is not lower than the predetermined value, the calculation unit 31a proceeds to step S103.
In steps S101 and S102, when the walking speed of the pedestrian is low, the added value can be added. For example, the added value in step S102 is “10”.

Next, the arithmetic unit 31a determines whether or not the attribute content in "ignore signal" is applicable (step S103).
If it is determined in step S103 that the attribute content of "ignore signal" is applicable, the operation unit 31a adds the added value to the base value (step S104), and proceeds to step S105.
On the other hand, if it is determined in step S103 that the attribute content of “ignore signal” is not applicable (not applicable), the operation unit 31a proceeds to step S105.
In step S103 and step S104, when the pedestrian ignores the signal, the added value can be added. For example, the added value in step S100 is “50”.

Next, the calculation unit 31a determines whether the attribute content in the “walking locus” is meandering (Step S105).
If it is determined in step S105 that the attribute content in the “walking locus” is meandering, the calculation unit 31a adds the addition value to the base value (step S106), and ends the addition processing.
On the other hand, if it is determined in step S105 that the attribute content in the “walking locus” is not meandering (is a straight line), the calculation unit 31a ends the addition processing.
In step S105 and step S106, for example, since the pedestrian is in a drunken state due to drinking or the like, the added value can be added when the pedestrian is meandering. For example, the added value in step S106 is “50”.

As described above, the arithmetic unit 31a performs the addition process after distinguishing between the vehicle and the pedestrian (FIG. 14).
Returning to FIG. 13, after completing the addition processing in step S67, the calculation unit 31a proceeds to step S68, and performs an addition processing of adding a value to a base value according to the attribute content of the evaluation target (step S68).
The addition processing in step S68 is the same as the addition processing in step S67 except that the addition processing is performed on the evaluation target, and thus the description is omitted here.
After completing the addition processing in step S68, the calculation unit 31a ends the individual calculation processing.

In the addition processing in steps S67 and S68, the calculation unit 31a calculates an addition value (appearance attribute value) corresponding to the attribute content of the appearance attribute of the evaluation target and the collision prediction target, and an addition value (action value) corresponding to the attribute content of the action attribute. Attribute value) to the base value.
As a result, it is possible to reflect the appearance of the evaluation target and the collision prediction target and the past or current behavior in the evaluation value.
In other words, factors that affect the collision prediction, which appear in the appearance and behavior of the evaluation target and the collision prediction target, can be reflected in the evaluation value.

Returning to FIG. 12, when the individual calculation processing in step S55 is completed, the calculation unit 31a proceeds to step S57, and uses the sum of the added value added in the individual calculation processing and the base value as the evaluation value of the collision prediction result. Is registered in the evaluation value database 34b (step S57).

The calculation unit 31a sequentially processes Steps S55 and S57 in FIG. 12 for the specified evaluation targets, and repeats the processing until all the evaluation targets are processed.
After obtaining the evaluation values for all of the specified evaluation targets and registering them in the database 34b, the calculation unit 31a returns to step S51 again and repeats the same processing.

Therefore, the arithmetic unit 31a repeats the calculation processing, repeatedly calculates and registers the evaluation value of each of the specified evaluation targets, and updates the registered contents of the evaluation value database 34b as needed.

FIG. 17 is a diagram illustrating an example of the evaluation value database 34b.
As shown in FIG. 17, in the evaluation value database 34b, a moving object ID to be evaluated and evaluation values of moving objects other than the evaluation object (other moving objects) are registered in association with each other.

The evaluation value database 34b is provided with a column of a mobile unit ID to be evaluated, a column of a vehicle ID (mobile ID), and a column of evaluation values of mobile units other than the evaluation target.
In the column of the mobile object ID to be evaluated, the mobile object ID of the mobile object specified as the evaluation object is registered.
In the column of vehicle ID (mobile ID), a vehicle ID (mobile ID) of a mobile terminal (pedestrian terminal 70 or in-vehicle device 50) included in the mobile object to be evaluated is registered.

The column of the evaluation value of the mobile unit other than the evaluation target includes a column of the mobile unit ID of each mobile unit other than the evaluation target. Thereby, the evaluation values of the plurality of moving objects are registered in the column of the evaluation values of the moving objects other than the evaluation target.
As described above, the evaluation value for the moving object other than the collision prediction target is set to “0”.
Therefore, a moving object other than the evaluation target whose evaluation value is registered with a value other than “0” is a collision prediction target to be evaluated.

For example, in FIG. 17, in the evaluation value of the evaluation target whose mobile object ID is “1006”, among the columns of the evaluation values of the mobile objects other than the evaluation target, the column corresponding to the mobile object ID of “1004” is the evaluation value. “500” is registered as a value, and “0” is registered as the other values. From this, it can be seen that the moving object with the moving object ID “1004” is specified as the collision prediction target of the evaluation object with the moving object ID “1006”.
In this manner, by referring to the evaluation value database 34b, it is possible to specify the collision prediction target of each evaluation target.

[Evaluation value judgment processing]
FIG. 18 is a flowchart illustrating an example of the determination processing by the determination unit 31b.
As illustrated in FIG. 18, the determination unit 31b refers to the evaluation value database 34b and determines whether there is an evaluation target whose evaluation value is equal to or greater than a preset threshold (threshold for notification determination) (Step S110). ).
If it is determined in step S110 that there is no evaluation target whose evaluation value is equal to or greater than the notification determination threshold, the determination unit 31b repeats step S110. Therefore, the determination unit 31b repeats step S110 until it determines that there is an evaluation target whose evaluation value is equal to or greater than the notification determination threshold.

If it is determined in step S110 that there is an evaluation target whose evaluation value is equal to or larger than the notification determination threshold, the determination unit 31b proceeds to step S111 and determines to notify the evaluation target of the collision prediction result of the collision prediction target. Then, the process returns to step S110.

The determination unit 31b refers to the evaluation value of each evaluation target, and if there are a plurality of evaluation targets whose evaluation values are equal to or greater than the notification determination threshold, determines the notification of the collision prediction result of the collision prediction target to all of the plurality of evaluation targets. I do.
As described above, the determination unit 31b determines, for each evaluation target, whether to notify the evaluation target of the collision prediction result of each evaluation target based on the evaluation value.

In the present embodiment, whether or not to notify the collision prediction result is determined for each evaluation target based on the evaluation value of each of the evaluation targets (the pedestrian terminal 70, the in-vehicle device 50, or the moving object including these). Therefore, necessary information can be appropriately provided to each evaluation target.
In addition, since the dynamic information map M1 on which the dynamic information on a plurality of moving objects is superimposed is used to obtain the evaluation value of the collision prediction result of each evaluation target, the moving objects including no mobile terminals are included. Collision prediction can be performed, and a collision prediction result appropriately predicted for each evaluation target can be represented as an evaluation value.

[Notification processing]
When the determination unit 31b determines to notify the evaluation target of the collision prediction result, the notification unit 31c notifies the evaluation target of which the determination unit 31b has determined the notification of the collision prediction result between the evaluation target and the collision prediction target. I do.
Thereby, information can be provided only to the mobile terminals for which it is determined that the collision prediction result is necessary based on the evaluation value.
As described above, the collision prediction result notified to the evaluation target by the notifying unit 31c includes the attribute of the collision prediction target, the direction in which the collision prediction target approaches, the collision prediction time, and the like.

The mobile terminal to be evaluated (the pedestrian terminal 70 and the in-vehicle device 50) that has received the notification by the notification unit 31c outputs the notified collision prediction result to the user of the mobile terminal.

According to the above configuration, the evaluation value for determining whether or not to notify the collision prediction result is determined based on the collision prediction time, the appearance attribute value, and the action attribute value. The appearance and behavior of the evaluation target and the collision prediction target can be reflected in the determination of whether or not to notify the result.
Therefore, for example, for a collision prediction regarding a moving object having a high possibility of collision in appearance or behavior, an evaluation value can be set so as to be weighted in a direction in which a collision prediction result is notified, and the necessity of notification is required. In determining, the appearance and behavior of the moving object can be taken into account.
As a result, the necessity of notification can be appropriately determined, and a collision prediction result with high necessity of notification can be appropriately provided to the mobile terminal.

[About Scenario 1]
Next, the operation of the information providing system of the present embodiment will be described according to a scenario assumed on a road.
FIG. 19 is a diagram showing a situation around an intersection according to scenario 1.
In FIG. 19, the

pedestrians

7A and 7B get off the vehicle 5C parked at the position immediately after passing the pedestrian crossing P and cross the pedestrian crossing P. The pedestrian 7A is an adult, and the pedestrian 7B is a child (having a height of 120 cm or less).
At the timing when the

pedestrians

7A and 7B started to cross the pedestrian crossing P, the light color of the signal of the pedestrian crossing P was flashing blue.

On the other hand, at the timing when the

pedestrians

7A and 7B start to cross the pedestrian crossing P, the traffic light on the route on which the vehicle 5A runs is red, but the traffic light on the pedestrian crossing P is blinking blue. The vehicle 5A approaching the pedestrian crossing P recognizes that it will soon turn blue. Therefore, it is assumed that the vehicle 5A is trying to pass through the intersection without stopping.
It is assumed that the

pedestrians

7A and 7B cannot see the vehicle 5A due to the presence of the vehicle 5C.
Further, a vehicle 5B is running behind the vehicle 5A.

It is assumed that the

vehicles

5A, 5B, 5C and the

pedestrians

7A, 7B are registered as moving objects in the dynamic information map M1 and the moving object database 34a by the system.
In addition, it is assumed that the

vehicles

5A and 5B have the in-vehicle device 50 mounted thereon, and the

pedestrians

7A and 7B carry the pedestrian terminal 70.

Here, it is assumed that the light color of the traffic light of the pedestrian crossing P turns red in the middle of the pedestrian crossing P, and the

pedestrians

7A and 7B cross the pedestrian crossing P in a hurry.

At this time, there is a possibility that the vehicle 5A and the

pedestrians

7A and 7B will collide on the pedestrian crossing P. Therefore, (the arithmetic unit 31a of) the edge server 3 specifies the

pedestrians

7A and 7B as the collision prediction targets of the vehicle 5A, and specifies the vehicle 5A as the collision prediction targets of the

pedestrians

7A and 7B. The attributes of the

pedestrians

7A and 7B are children (only the pedestrian 7B) and the signals are ignored. Further, a vehicle 5C that is a blind spot factor exists between the vehicle 5A and the

pedestrians

7A and 7B.

In this case, when calculating the evaluation value when the vehicle 5A is to be evaluated, the edge server 3 obtains a basic value based on the collision prediction time, and in addition to the addition value obtained by the addition processing regarding the vehicle 5, the blind spot factor and the

pedestrian

7A , 7B are added to the base value to obtain an evaluation value. At this time, in the addition process for the

pedestrians

7A and 7B, the height of the pedestrian and the added value due to ignoring the signal are added.
The addition process for the vehicle 5 is appropriately added depending on the attribute contents of the appearance attribute and the behavior attribute of each vehicle 5. The same applies to the vehicle 5 appearing in the following description.

(4) The notification determination threshold used for determining whether or not to notify a collision prediction result is set to be smaller than an evaluation value in a case where factors that impair safety overlap as in this scenario. Therefore, in the case of the present scenario, the evaluation values of the

pedestrians

7A and 7B in the vehicle 5A are larger than the notification determination threshold. For this reason, the edge server 3 notifies the vehicle 5A of the collision prediction result of the

pedestrians

7A and 7B in the vehicle 5A.

Further, the edge server 3 performs the addition process on both the evaluation target and the collision prediction target when obtaining the evaluation value. Therefore, the evaluation value when the

pedestrians

7A and 7B are the evaluation targets The evaluation value is the same as the target. Therefore, the evaluation value of the vehicle 5A in the

pedestrians

7A and 7B becomes larger than the notification determination threshold, and the edge server 3 notifies the

pedestrians

7A and 7B of the collision prediction result of the vehicle 5A in the

pedestrians

7A and 7B. I do.

{Circle around (5)} Also, in the vehicle 5B, the speed decreases as the vehicle 5A approaches the pedestrian crossing P, and when both vehicles approach, the edge server 3 specifies the vehicle 5A as a collision prediction target of the vehicle 5B. As a result of the collision prediction, if the collision prediction time is short, the evaluation value is set to a large value. As a result, the edge server 3 notifies the vehicle 5B of the collision prediction result of the vehicle 5A in the vehicle 5B.

In this way, the edge server 3 can appropriately provide necessary information to each mobile terminal.
The in-vehicle devices 50 of the

vehicles

5A and 5B and the pedestrian terminals 70 of the

pedestrians

7A and 7B output a collision prediction result to the user of the own device based on the notification from the edge server 3.

In FIG. 19, an output screen V1 of the vehicle-mounted device 50 of the vehicle 5A includes a display D1 indicating that a pedestrian appears from the right side of the pedestrian crossing P in front of the own vehicle, an arrow D2 indicating the traveling direction of the pedestrian, and the like. Is displayed.
Thereby, the edge server 3 can make the user of the vehicle 5A recognize in advance that a pedestrian appears from the right side of the pedestrian crossing P ahead.
Here, although the

pedestrians

7A and 7B are displayed without being specified, for example, the control unit 51 of the vehicle 5A displays whether the attribute of the pedestrian 7 is a child or an adult. The method may be different. This is because more attention is required when the pedestrian is a child.

That is, in the present system, the output mode of the collision prediction result may be controlled to be different depending on the attribute contents of the appearance attribute and the action attribute in the collision prediction target. Thus, information can be output to the user in an output mode according to the characteristics of the attribute of the collision prediction target.

The output screen V2 of the pedestrian terminal 70 of the

pedestrians

7A and 7B includes a display D3 indicating that the vehicle A appears from the left side of the pedestrian crossing P ahead, an arrow D4 indicating the traveling direction of the vehicle 5A, and the like. Is displayed.
Thereby, the edge server 3 can make the

pedestrians

7A and 7B recognize in advance that the vehicle appears from the left side of the pedestrian crossing P in front.

In addition, on the output screen V3 of the in-vehicle device 50 of the vehicle 5B, a display D5 indicating that the vehicle 5A traveling ahead of the own vehicle is warned, an arrow D6 indicating the traveling direction of the vehicle 5A, and the like are displayed.
Thereby, the edge server 3 can make the user of the vehicle 5B recognize in advance that the vehicle 5A traveling ahead stops and approaches.

In this way, the edge server 3 can avoid collision between the moving bodies by causing the in-vehicle device 50 of the

vehicles

5A and 5B and the pedestrian terminals 70 of the

pedestrians

7A and 7B to output to the user. it can.

The output of the collision prediction result by the pedestrian terminal 70 and the in-vehicle device 50 is controlled by an output control unit of the control unit 71 and the control unit 51 of the pedestrian terminal 70 and the in-vehicle device 50.
When the control unit 31 of the edge server 3 has an output control unit that can control the output of the pedestrian terminal 70 and the in-vehicle device 50 to the user, the output control unit of the edge server 3 The output to the user by the pedestrian terminal 70 and the in-vehicle device 50 may be controlled.

In addition, FIG. 19 shows the case where the position of the collision prediction target and the traveling direction thereof are displayed on the output screens V1, V2, V3, but may be displayed together with the collision prediction time.

Further, in this scenario, a case was shown in which the factors that impede safety in pedestrians overlap. For example, there are no factors that impede safety such as signal ignorance in the

pedestrians

7A and 7B, and the evaluation value is If the value is not larger than the notification determination threshold, the edge server 3 does not notify the collision prediction result to the vehicle 5A and the

pedestrians

7A and 7B.

On the other hand, even if the

pedestrians

7A and 7B do not have any obstacles to safety such as ignoring the signal, the added value is added according to the attribute content of the vehicle 5A, so that the evaluation value exceeds the notification determination threshold. There is also.
In the vehicle addition process (FIG. 15), the added value is added based on the size, shape, color, license plate display, and past traveling history of the vehicle 5A.
Therefore, although the

pedestrians

7A and 7B are set as the collision prediction targets in the vehicle 5A, even when the collision prediction result is not notified, the edge server 3 may determine the collision prediction result depending on the attribute of the vehicle 5A. The

pedestrians

7A and 7B and the vehicle 5A may be notified.

As described above, the attribute of the vehicle 5A can be taken into account in determining the necessity of the notification of the collision prediction result. As a result, the necessity of notification of the collision prediction result can be appropriately determined, and the notification of the collision prediction result can be appropriately performed.

[About scenario 2]
FIG. 20 is a diagram showing a situation around an intersection according to scenario 2.
In FIG. 20, the settings of the

vehicles

5A and 5B and the

pedestrians

7A and 7B are the same as in the scenario 1. Further, the scenario 2 does not include the vehicle 5C of the scenario 1.
In scenario 2, the

pedestrians

7A and 7B are crossing the pedestrian crossing P at a speed lower than a general walking speed (3.6 km per hour).

Since the

pedestrians

7A and 7B are slowly crossing the pedestrian crossing P, at the timing when the

pedestrians

7A and 7B start to cross the pedestrian crossing P, the light color of the signal of the pedestrian crossing P is flashing blue. In the middle of the pedestrian crossing P, it is assumed that the light color of the signal of the pedestrian crossing P changes to red, and the

pedestrians

7A and 7B slowly cross the pedestrian crossing P at the same walking speed.

In this case, there is no blind spot factor between the vehicle 5A and the

pedestrians

7A and 7B as in the scenario 1.
Further, since the

pedestrians

7A and 7B are slowly crossing the pedestrian crossing P, the light color of the signal of the pedestrian crossing P which has been blinking blue changes to red in the middle of the pedestrian crossing P, and the

pedestrians

7A and 7B Is slowly crossing the crosswalk P at the same walking speed.
Here, there is a possibility that the vehicle 5A and the

pedestrians

7A and 7B will collide on the crosswalk. Therefore, the edge server 3 specifies the

pedestrians

7A and 7B.

At this time, when obtaining the evaluation value when the vehicle 5A is to be evaluated, the edge server 3 obtains a basic value from the collision prediction time, and adds the

pedestrians

7A and 7B in addition to the addition value by the addition processing for the vehicle 5. The value added by the processing is added to the base value to obtain an evaluation value. At this time, in the addition processing relating to the

pedestrians

7A and 7B, the height of the pedestrian (120 cm or less), the speed of the pedestrian, and the added value due to ignoring the signal are added.
As a result, the evaluation value of the

pedestrians

7A and 7B in the vehicle 5A becomes larger than the notification determination threshold, and the edge server 3 notifies the vehicle 5A of the collision prediction result of the

pedestrians

7A and 7B in the vehicle 5A.
Further, the edge server 3 notifies the

pedestrians

7A and 7B of the prediction result of the collision prediction of the vehicle 5A in the

pedestrians

7A and 7B.

As described above, the in-vehicle devices 50 of the

vehicles

5A and 5B and the pedestrian terminals 70 of the

pedestrians

7A and 7B output the collision prediction result to the user of the own device based on the notification from the edge server 3.
Thereby, the edge server 3 can make the user of the vehicle 5A recognize in advance that a pedestrian appears from the right side of the pedestrian crossing P in front.
In addition, the edge server 3 can cause the

pedestrians

7A and 7B to recognize in advance that a vehicle will appear from the left side of the pedestrian crossing P in front.

Thereby, the edge server 3 can avoid collision between the moving objects by causing the in-vehicle device 50 of the

vehicles

5A and 5B and the pedestrian terminals 70 of the

pedestrians

7A and 7B to output to the user. .

[About Scenario 3]
FIG. 21 is a diagram showing a situation around an intersection according to scenario 3.
In FIG. 21, when the

pedestrians

7A and 7B are walking on the sidewalk H, there is an obstacle G1 on the sidewalk H, and the

pedestrians

7A and 7B run off the roadway while avoiding the obstacle G1. The pedestrian 7A is an adult, and the pedestrian 7B is a child.

Here, the vehicle 5A and the

pedestrians

7A and 7B may collide on the road. Therefore, the edge server 3 specifies the

pedestrians

7A and 7B. In addition, the

pedestrians

7A and 7B have a child attribute (only the pedestrian 7B) and do not walk on the sidewalk.

At this time, when obtaining the evaluation values of the

pedestrians

7A and 7B when the vehicle 5A is the evaluation target, the edge server 3 obtains the base value from the predicted collision time, The addition value obtained by the addition processing for the

persons

7A and 7B is added to the base value to obtain an evaluation value. At this time, in the addition process for the

pedestrians

7A and 7B, an added value based on the height (120 cm or less) of the pedestrian and the walking place (roadway) of the pedestrian is added.
As a result, the evaluation values of the

pedestrians

7A and 7B in the vehicle 5A become larger than the notification determination threshold, and the edge server 3 notifies the vehicle 5A of the collision prediction result of the

pedestrians

7A and 7B in the vehicle 5A.
Further, the edge server 3 notifies the

pedestrians

7A and 7B of the result of the collision prediction of the vehicle 5A in the

pedestrians

7A and 7B.

If the vehicle 5A attempts to protrude into the opposite lane avoiding the

pedestrians

7A and 7B, the vehicle 5B running on the opposite lane may collide with the vehicle 5A. Therefore, in this case, the edge server 3 specifies the vehicle 5A as a collision prediction target of the vehicle 5B. Further, a vehicle 5D which is a blind spot factor exists between the vehicle 5B and the vehicle 5A.

When calculating the evaluation value when the vehicle 5B is to be evaluated, the edge server 3 adds an addition value based on the presence or absence of a blind spot factor to the evaluation value obtained from the collision prediction time.
When the evaluation value of the vehicle 5A in the vehicle 5B is larger than the notification determination threshold, the edge server 3 notifies the vehicle 5B of the collision prediction result of the vehicle 5A in the vehicle 5B.

{Circle around (5)} Also, in the vehicle 5C, the speed of the vehicle 5B is reduced by the protrusion of the vehicle 5A, and when the two vehicles approach, the edge server 3 specifies the vehicle 5B as a collision prediction target of the vehicle 5C. As a result of the collision prediction, if the collision prediction time is short, the evaluation value is set to a large value. As a result, the edge server 3 notifies the collision prediction result of the vehicle 5B in the vehicle 5C to the vehicle 5C.

Thus, the edge server 3 causes the in-vehicle devices 50 of the

vehicles

5A, 5B, 5C and the pedestrian terminals 70 of the

pedestrians

7A, 7B to output to the user, thereby avoiding collision between the moving objects. Can be.

[About Scenario 4]
FIG. 22 is a diagram showing a situation around an intersection according to scenario 4.
In FIG. 22, the

pedestrians

7A and 7B are walking on the sidewalk H, pass through between the stopped vehicle 5C and the vehicle 5D, and walk on a place other than the pedestrian crossing to cross the road. The pedestrian 7A is an adult, and the pedestrian 7B is a child.

Here, there is a possibility that the vehicle 5A running in the opposite lane and the

pedestrians

7A and 7B will collide on the road. Therefore, the edge server 3 specifies the

pedestrians

7A and 7B. In addition, the

pedestrians

7A and 7B have a child attribute (only the pedestrian 7B) and do not walk on the sidewalk and the crosswalk. Further, a vehicle 5D which is a blind spot factor exists between the vehicle 5A and the

pedestrians

7A and 7B.

At this time, when calculating the evaluation values of the

pedestrians

7A and 7B when the vehicle 5A is to be evaluated, the edge server 3 obtains a base value from the predicted collision time, The factors and the added value by the adding process for the

pedestrians

7A and 7B are added to the base value to obtain an evaluation value. At this time, in the addition process for the

pedestrians

7A and 7B in the vehicle 5A.
Further, the edge server 3 notifies the

pedestrians

7A and 7B of the result of the collision prediction of the vehicle 5A in the

pedestrians

7A and 7B.

{Circle around (5)} Also, in the vehicle 5B, when the vehicle 5A slows down due to the

pedestrians

7A and 7B crossing the road, and the two vehicles approach, the edge server 3 specifies the vehicle 5A as a collision prediction target of the vehicle 5B. As a result of the collision prediction, if the collision prediction time is short, the evaluation value is set to a large value. As a result, the edge server 3 notifies the vehicle 5B of the collision prediction result of the vehicle 5A in the vehicle 5B.

As a method of detecting a moving object (pedestrian) passing between the vehicles 5 as in this scenario, in addition to the detection by the roadside sensor 8, an in-vehicle device having an in-vehicle camera 59 in the vehicle 5 in which a pedestrian passes. If 50 is mounted, it can also be detected by the in-vehicle camera 59.

[About Scenario 5]
FIG. 23 is a diagram showing a situation around an intersection according to scenario 5.
In FIG. 23,

pedestrians

7A and 7B are walking meandering on sidewalk H. The pedestrian 7A is an adult, and the pedestrian 7B is a child.

If the

pedestrians

7A and 7B meander the sidewalk H and run off the road, the vehicle 5A may collide with the

pedestrians

7A and 7B on the road. Therefore, the edge server 3 specifies the

pedestrians

7A and 7B. In addition, the

pedestrians

7A and 7B have a child attribute (only the pedestrian 7B), and if the pedestrians run off the road, they are not walking on the sidewalk.

At this time, when obtaining the evaluation values of the

pedestrians

persons

pedestrians

7A and 7B, an added value based on the height (120 cm or less) of the pedestrian, the walking locus (meandering) of the pedestrian, and the walking place (roadway) is added.
As a result, the evaluation values of the

pedestrians

7A and 7B in the vehicle 5A.
Further, the edge server 3 notifies the

pedestrians

7A and 7B of the result of the collision prediction of the vehicle 5A in the

pedestrians

7A and 7B.

If the vehicle 5A attempts to protrude into the opposite lane avoiding the

pedestrians

7A and 7B, the vehicle 5B running on the opposite lane may collide with the vehicle 5A. Therefore, in this case, the edge server 3 specifies the vehicle 5A as a collision prediction target of the vehicle 5B.

The edge server 3 obtains an evaluation value when the vehicle 5B is an evaluation target, based on a predicted collision time between the

pedestrians

7A and 7B.
When the evaluation value of the vehicle 5A in the vehicle 5B is larger than the notification determination threshold, the edge server 3 notifies the vehicle 5B of the collision prediction result of the vehicle 5A in the vehicle 5B.

[About Scenario 6]
FIG. 24 is a diagram showing a situation around an intersection according to scenario 6.
In FIG. 24,

pedestrians

7A and 7B are crossing a pedestrian crossing P that crosses a route R1. At the timing when the

pedestrians

7A and 7B start crossing the pedestrian crossing P, the light color of the signal of the pedestrian crossing P is blinking blue, but the light color of the signal of the pedestrian crossing P is It turns red, and it is assumed that the

pedestrians

7A and 7B cross the pedestrian crossing P in a hurry. The pedestrian 7A is an adult, and the pedestrian 7B is a child.

The vehicle 5A enters the intersection from the route R2, turns left, and travels in the intersection toward the route R1. The vehicle 5B is running following the vehicle 5A. Since the light color of the signal on the pedestrian crossing P changes from blinking blue to red, the light color of the

vehicles

5A and 5B and the traffic light ahead also changes from blue to yellow and red. Therefore, the

vehicles

5A and 5B are also in a hurry to pass the intersection.
Further, there is a building G2 between the route R2 and the route R1 and at a corner of the intersection, which blocks the line of sight between the route R1 and the route R2.

Here, there is a possibility that the vehicle 5A that turns left at the intersection toward the route R1 and the

pedestrians

7A and 7B collide on the road. In particular, since there is a building G2 obstructing the line of sight between the route R1 and the route R2, it is difficult for the vehicle 5A and the

pedestrians

7A and 7B to notice each other until they approach each other. .

When the vehicle 5A turns left and runs toward the

pedestrians

7A and 7B, the edge server 3 specifies the

pedestrians

7A and 7B as the collision prediction targets of the vehicle 5A and the collision prediction targets of the

pedestrians

7A and 7B. , The vehicle 5A is specified. Further, the attributes of the

pedestrians

7A and 7B are children (only the pedestrian 7B), and the signals are ignored. Further, a building G2 that is a blind spot factor exists between the vehicle 5A and the

pedestrians

7A and 7B.

At this time, when obtaining the evaluation value when the vehicle 5A is the evaluation target, the edge server 3 obtains the base value from the collision prediction time, and in addition to the addition value by the addition processing regarding the vehicle 5, the blind spot factor and the

pedestrian

7A , 7B are added to the base value to obtain an evaluation value. At this time, in the addition processing for the

pedestrians

7A and 7B, the height of the pedestrian (120 cm or less) and an added value due to ignoring the signal are added.
As a result, the evaluation value of the

pedestrians

7A and 7B in the vehicle 5A.
Further, the edge server 3 notifies the

pedestrians

7A and 7B.

{Circle around (5)} Also, in the vehicle 5B, when the vehicle 5A notices the

pedestrians

7A and 7B on the pedestrian crossing P and slows down, and the two vehicles approach, the edge server 3 specifies the vehicle 5A as a collision prediction target of the vehicle 5B. As a result of the collision prediction, if the collision prediction time is short, the evaluation value is set to a large value. As a result, the edge server 3 notifies the vehicle 5B of the collision prediction result of the vehicle 5A in the vehicle 5B.

As described above, the in-vehicle devices 50 of the

vehicles

5A and 5B and the pedestrian terminals 70 of the

pedestrians

7A and 7B output the collision prediction result to the user of the own device based on the notification from the edge server 3.
Thereby, the edge server 3 can make the user of the vehicle 5A recognize in advance that the pedestrian is crossing the pedestrian crossing P.
In addition, the edge server 3 can make the

pedestrians

7A and 7B recognize in advance that the vehicle 5A will appear from the right side of the pedestrian crossing P.

vehicles

5A and 5B and the pedestrian terminals 70 of the

pedestrians

7A and 7B to output to the user. .

[About the second embodiment]
FIG. 25 is a diagram illustrating the storage unit 34 of the edge server 3 according to the second embodiment.
The storage unit 34 of the present embodiment stores a dynamic information map M1, a moving object database 34a, an evaluation value database 34b, and a relative distance database 34c.
In the relative distance database 34c, the relative distance between the moving objects registered in the moving object database 34a is registered.

The calculation unit 31a of the present embodiment evaluates using the movement information (position information, azimuth information, and speed information) of the evaluation target (target moving object) and the moving objects other than the evaluation target (other moving objects). The third embodiment is different from the first embodiment in that an addition process based on the relative distance between the moving objects registered in the relative distance database 34c is further performed in addition to the individual calculation process for obtaining the value.
The configuration other than the configuration described below is the same as that of the first embodiment.

FIG. 26 is a diagram illustrating an example of the relative distance database 34c.
As shown in FIG. 26, the relative distance between the moving objects registered in the moving object database 34a is registered in the relative distance database 34c.
In the relative distance database 34c, a relative distance from another moving object ID is registered for each moving object ID. Further, a past relative distance is also registered in the relative distance database 34c.
For example, in the relative distance database 34c, as the relative distance between the mobile unit ID “1001” and the mobile unit ID “1002”, “132 m” is registered as the relative distance obtained from the latest updated position information, “152 m” is registered as the relative distance obtained from the positional information at the time, and “172 m” is registered as the relative distance obtained from the positional information at the time of updating two times before. That is, in the relative distance database 34c, a plurality of relative distances arranged in chronological order at each update interval are registered.

Each time the detecting unit 31d updates the moving object database 34a in which the position information of each moving object is registered, the detecting unit 31d calculates the relative distance between the moving objects and registers the relative distance in the relative distance database 34c.
Therefore, the detecting unit 31d updates the relative distance database 34c at the same time as updating the moving object database 34a as needed.

FIG. 27 is a flowchart illustrating an example of a calculation process performed by the calculation unit 31a according to the present embodiment.
In the calculation processing of the present embodiment, an addition processing (step S59) based on the relative distance is performed after the individual calculation processing in step S55.
In the adding process based on the relative distance, a process of further adding the added value to each evaluation value registered in step S57 is performed.
Therefore, here, step S59 will be mainly described.

FIG. 28 is a flowchart showing an example of the adding process based on the relative distance in step S59 in FIG.
As illustrated in FIG. 28, the calculation unit 31a first specifies a mobile object other than the evaluation target (step S120), and among the mobile objects other than the evaluation target, the relative distance to the target mobile object is within 100 m and the relative distance to the target mobile object is less than 100 m. A moving body having a negative distance displacement is specified as an approaching target (step S121).

The displacement of the relative distance is a value obtained by subtracting the past relative distance from the current relative distance. When the displacement of the relative distance is positive, it indicates that the relative distance is increasing, and the evaluation target and the moving object other than the evaluation target are moving in a direction relatively away from each other. In addition, when the displacement of the relative distance is negative, it indicates that the relative distance is decreasing, and it can be seen that the evaluation target and the moving object other than the evaluation target are moving in a direction relatively approaching. .
The calculation unit 31a obtains the displacement of the relative distance by subtracting the relative distance based on the position information updated earlier than the latest one from the relative distance based on the latest updated position information.
The arithmetic unit 31a subtracts the relative distance based on the position information updated earlier than the latest one from the relative distance based on the latest updated position information, and further calculates a unit based on the update interval of the relative distance database 34c. The displacement of the relative distance per time may be obtained, and this may be used as the displacement of the relative distance.

A moving object located at a predetermined distance around the target moving object and approaching the target moving object has a possibility of colliding with or contacting the target moving object.
Therefore, the calculation unit 31a specifies, as an approaching target, a moving object that is located within 100 m around the target moving object and whose relative distance decreases so as to approach the target moving object.
That is, the moving object specified as the approaching target is a moving object that has a possibility of colliding or contacting the target moving object.

After step S121, the calculation unit 31a performs an addition process on the evaluation value of the moving object specified as an approaching target (hereinafter, also referred to as an approaching moving object) (step S122).
FIG. 29 is a flowchart illustrating an example of an addition process for an evaluation value.
The addition processing for the approaching moving body is processing for further adding an addition value to the evaluation value of the approaching moving body in the target moving body.
Note that the evaluation value of the approaching moving body in the target moving body is obtained by an individual calculation process.

The calculation unit 31a repeats the processing from step S131 to step S137 until all the approaching moving objects identified in step S121 are processed.
Hereinafter, in the description of steps S132 to S136, the processing of the target moving object and one of the specified approaching moving objects will be described.

First, in step S132, the calculation unit 31a determines whether the relationship between the target moving body and the approaching moving body is pedestrian-to-pedestrian (step S132). If it is determined in step S132 that the relationship between the target moving body and the approaching moving body is pedestrian-to-pedestrian, the calculation unit 31a proceeds to step S137 without performing the adding process on the approaching moving body. Accordingly, when the relationship between the target moving body and the approaching moving body is pedestrian-to-pedestrian, the addition processing is not performed.

On the other hand, if it is determined in step S132 that the relationship between the target moving object and the approaching moving object is not a pedestrian-to-pedestrian, the calculation unit 31a proceeds to step S134, where the relationship between the target moving object and the approaching moving object is determined. It is determined whether or not the vehicle is a vehicle-to-vehicle (step S134).
If it is determined in step S134 that the relationship between the target moving body and the approaching moving body is not a vehicle-to-vehicle, the calculation unit 31a proceeds to step S136 and performs a process corresponding to the vehicle-to-pedestrian or pedestrian-to-vehicle.

FIG. 30 is a flowchart illustrating an example of a process corresponding to vehicle-to-pedestrian or pedestrian-to-vehicle.
In this case, one of the target moving body and the approaching moving body is a vehicle, and the other is a pedestrian. Here, a case where the target moving body is a vehicle and the approaching moving body is a pedestrian will be described. However, the same processing is performed when the target moving body is a pedestrian and the approaching moving body is a vehicle.

As shown in FIG. 30, first, the calculation unit 31a refers to the attribute “walking place” of the pedestrian who is the approaching moving body, and determines whether or not the content is in the sidewalk or in the building (step S141).

The detection unit 31d of the present embodiment acquires whether the pedestrian's walking place is a roadway, a sidewalk, other (in a building), or a pedestrian crossing, and describes the content of the pedestrian's behavior attribute “walking place”. , One of a roadway, a sidewalk, other (inside a building), and a crosswalk is registered.

When it is determined that the content of the “walking place” is inside the sidewalk or the building, the calculation unit 31a ends the process without performing the adding process on the approaching moving body (pedestrian), and proceeds to step S137 (FIG. 29). Proceed to. When the pedestrian is in the sidewalk or the building, it is considered that the possibility that the pedestrian will collide with the vehicle as the target moving body is low. Therefore, when the pedestrian is on the sidewalk or in the building, the calculation unit 31a does not perform the addition processing.

If it is determined in step S141 that the content of “walking place” is not in the sidewalk or in the building, the calculation unit 31a refers to the attribute “speed” of the target moving object vehicle and determines whether the speed of the vehicle is 5 km / h or less. It is determined whether or not it is (step S143).
When determining that the speed of the vehicle is 5 km / h or less, the calculation unit 31a ends the process without performing the addition process on the pedestrian who is the approaching moving body, and proceeds to step S137 (FIG. 29). The reason why the addition process is not performed in this case is that if the speed of the vehicle is 5 km / h or less, the possibility that the pedestrian will collide with the vehicle as the target moving body is considered to be low.

If it is determined in step S143 that the speed of the vehicle is not lower than 5 km / h, the calculation unit 31a refers to the attribute “walking place” of the pedestrian and determines whether or not the content is a pedestrian crossing (step S144). .
When determining that the content of “walking place” is a pedestrian crossing, the calculation unit 31a proceeds to step S145, and determines whether a pedestrian who is an approaching moving body has passed the center of the road traversed by the pedestrian crossing. (Step S145).
The calculation unit 31a determines whether the pedestrian has passed the center of the road. The calculation unit 31a refers to the map M1 and the attribute information of the pedestrian, and based on the position and the moving direction of the pedestrian, the position of the pedestrian has walked on the pedestrian crossing and has passed through the center of the road which is an intermediate point of the pedestrian crossing. It is determined whether or not it is a position.

When determining that the pedestrian has passed the center of the road, the calculation unit 31a determines whether or not the position of the pedestrian is within 2 m until reaching the sidewalk (step S146).
The calculation unit 31a also determines whether the position of the pedestrian is within 2 meters before reaching the sidewalk. The calculation unit 31a refers to the map M1 and the attribute information of the pedestrian, and determines whether the pedestrian walks on the pedestrian crossing from the pedestrian's position and the moving direction and the pedestrian's position is within 2 m until reaching the sidewalk. Is determined.

(4) If it is determined that the position of the pedestrian is within 2 m before reaching the sidewalk, the calculation unit 31a ends the process without performing the addition process on the pedestrian who is the approaching moving body, and proceeds to step S137 (FIG. 29). If the position of the pedestrian is within 2 m before reaching the sidewalk, it is considered that the pedestrian quickly finishes crossing the pedestrian crossing, and the possibility that the pedestrian will collide with the vehicle as the target moving body is low. Therefore, if it is determined in step S142 that the pedestrian is within 2 m until the pedestrian reaches the sidewalk, the calculation unit 31a does not perform the addition process.

On the other hand, if it is determined in step S144 that the content of the pedestrian's behavior attribute “walking place” is not a pedestrian crossing, the operation unit 31a proceeds to step S147, and adds the addition value to the evaluation value of the approaching moving body in the target moving body. (Step S147), the process ends.
In step S147, the calculation unit 31a refers to the evaluation value database 34b, and specifies a column in which an evaluation value between the moving object ID corresponding to the target moving object and the moving object ID corresponding to the approaching moving object is registered. I do. The calculation unit 31a adds the addition value to the evaluation value registered in the specified column.
The added value added to the evaluation value in step S147 is, for example, “200”.

In this case, the pedestrian is likely to be walking on a road other than the crosswalk, and the pedestrian's behavior attribute “walking location” is also a road. Since the pedestrian is located on the roadway and the relative distance between the pedestrian and the vehicle that is relatively approaching is within 100 m, it can be considered that the possibility that the pedestrian and the vehicle collide is relatively high. . Therefore, the added value added here has a high necessity of notification, and is set to the same value as the base value when the predicted collision time is less than 3 seconds in the individual calculation processing.

If it is determined in step S145 that the pedestrian has not passed through the center of the road, the calculation unit 31a also proceeds to step S147, and if it is determined that the pedestrian is not within 2 meters before reaching the sidewalk, the target movement The addition value is added to the evaluation value of the approaching moving body in the body (step S147), and the process ends.
In this case, the pedestrian is walking on the pedestrian crossing, but it still requires a certain amount of time to finish crossing the pedestrian crossing, during which the approaching vehicle may collide with the pedestrian. It can be considered that there is. Therefore, also in this case, the addition value is added to the evaluation value of the approaching moving body in the target moving body.

If it is determined in step S146 that the pedestrian is not within 2 meters before reaching the sidewalk, the calculation unit 31a proceeds to step S148, and adds an addition value to the evaluation value of the approaching moving body in the target moving body (step S148). Finish the process.
In step S148, the calculation unit 31a refers to the evaluation value database 34b and specifies a column in which an evaluation value between the moving object ID corresponding to the target moving object and the moving object ID corresponding to the approaching moving object is registered. I do. The calculation unit 31a adds the addition value to the evaluation value registered in the specified column.
The added value added to the evaluation value in step S148 is, for example, "100".

In this case, the pedestrian has crossed the pedestrian crossing halfway, but it still requires a certain amount of time to complete the pedestrian crossing, during which the approaching vehicle may collide with the pedestrian. Can be considered to be. Therefore, also in this case, the addition value is added to the evaluation value of the approaching moving body in the target moving body.
It should be noted that the time required for the pedestrian to cross the pedestrian crossing is shorter than the case where it is determined that the pedestrian has not passed through the center of the road. Therefore, the added value in step S148 is set to a value smaller than the added value in step S147.

Returning to FIG. 29, when it is determined in step S134 that the relationship between the target moving body and the approaching moving body is vehicle-to-vehicle, the calculation unit 31a proceeds to step S135 and performs processing corresponding to the vehicle-to-vehicle. .

FIG. 31 is a flowchart illustrating an example of a process corresponding to a vehicle-to-vehicle process.
As shown in FIG. 31, first, the calculation unit 31a determines whether the vehicle of the target moving body and the vehicle of the approaching moving body are running in parallel on the same road in the same direction, or are traveling on the same lane. Is determined (step S151).
The determination unit 31a determines whether the two vehicles are running in parallel in the same direction on the same road or in the same lane.
The calculation unit 31a refers to the map M1 and the attribute information of the two vehicles, and based on the positions and the moving directions of the two vehicles, the two vehicles are running in parallel on the same road in the same direction, or traveling in the same lane. Is determined.

If it is determined that the two vehicles are not traveling in the same direction on the same road and are not traveling in the same lane, the arithmetic unit 31a determines whether or not the road on which both vehicles travel has a median strip. Is determined (step S152).

判定 The calculation unit 31a also determines whether or not there is a median strip on the road on which both vehicles travel. The calculation unit 31a refers to the map M1 and the attribute information of both vehicles, and determines whether there is a median strip on the road on which both vehicles travel.

When determining that there is a median strip on the road on which both vehicles travel, the calculation unit 31a ends the process without adding the added value to the evaluation value of the approaching moving body in the target moving body, and proceeds to step S137 (FIG. 29). move on.
If both vehicles are not running side by side on the same road in the same direction and are not traveling in the same lane when both vehicles are approaching each other, both vehicles are traveling in opposite lanes It is believed that there is. At this time, if there is a median strip, it is considered that the possibility of collision between the two is extremely low. For this reason, if it is determined in step S152 that there is a median strip on the road on which both vehicles travel, the calculation unit 31a does not perform the addition processing.

On the other hand, if it is determined in step S152 that there is no median strip on the road on which both vehicles are traveling, the calculation unit 31a proceeds to step S153, adds the addition value to the evaluation value of the approaching moving body in the target moving body, and performs the processing. Finish.
In this case, the two vehicles are traveling on the opposite lane and do not collide in principle. However, since the two vehicles are approaching each other, the calculation unit 31a performs an addition process.
In step S153, the calculation unit 31a refers to the evaluation value database 34b and specifies a column in which an evaluation value between the moving object ID corresponding to the target moving object and the moving object ID corresponding to the approaching moving object is registered. I do. The calculation unit 31a adds the addition value to the evaluation value registered in the specified column.
The added value added to the evaluation value in step S153 is, for example, “50”.

In step S151, when it is determined that the vehicle of the target moving body and the vehicle of the approaching moving body are running in parallel on the same road in the same direction or are running on the same lane, the arithmetic unit 31a It is determined whether or not the road curves within 100 m in front of the moving vehicle (step S154).

判定 The calculation unit 31a determines whether the road is curved within 100m ahead. The calculation unit 31a refers to the map M1 and the attribute information of the vehicle of the target moving body, and determines whether the road is curved within a range of 100 m ahead of the target moving body. The curve determined in step S154 indicates a curve that can be turned at a legal speed without greatly reducing the speed.

If it is determined that the road is not curved within 100 m ahead, the calculation unit 31a ends the processing without adding the added value to the evaluation value of the approaching moving body in the target moving body, and proceeds to step S137 (FIG. Proceed to 29).
If both vehicles are running side by side, or are traveling in the same lane, such a condition may occur if both vehicles are traveling normally on a straight road, even if both vehicles are gradually approaching. It is not necessary to actively notify the vehicle of the collision prediction result. Therefore, if it is determined in step S154 that the road is not curved within 100 m ahead, the calculation unit 31a does not perform the addition processing.

If it is determined in step S154 that the road is curved within 100 m ahead, the operation unit 31a proceeds to step S155, adds the added value to the evaluation value of the approaching moving body in the target moving body, and ends the processing.
When both vehicles enter the curve, both vehicles need to turn the steering wheel according to the curve. When both vehicles are running side by side or on the same lane and approach each other, the possibility of contact with each other increases if both vehicles turn the steering wheel. Therefore, the arithmetic unit 31a performs an addition process.

In step S155, the calculation unit 31a refers to the evaluation value database 34b and specifies a column in which an evaluation value between the moving object ID corresponding to the target moving object and the moving object ID corresponding to the approaching moving object is registered. I do. The calculation unit 31a adds the addition value to the evaluation value registered in the specified column.
The added value added to the evaluation value in step S155 is, for example, “200”. In step S155, since both vehicles need to turn the steering wheel, it is considered that there is a higher possibility that both vehicles will contact each other than in the case of step S155. Therefore, the added value in step S155 is set to a value larger than the added value in step S153.

Returning to FIG. 29, after completing the process of step S135 or the process of step S136, the calculation unit 31a proceeds to step S137.
The calculation unit 31a repeats the processing from step S131 to step S137 until all the approaching moving objects identified in step S121 are processed.

When the processing for all approaching moving objects is completed, the calculation unit 31a ends the addition processing for the evaluation value (Step S122 in FIG. 28), and ends the addition processing based on the relative distance.
As described above, in the addition processing based on the relative distance, the calculation unit 31a treats the approaching moving body as a moving body having a possibility of colliding with or contacting the target moving body, and performs the addition processing on the evaluation value for the approaching moving body.

Here, in the individual calculation processing, “0” is set as an evaluation value for a moving object other than the collision prediction target. However, in the addition processing based on the relative distance, a plurality of approaching moving bodies are specified, and an addition value is added to the evaluation value of each of the plurality of approaching moving bodies.
Therefore, in the individual calculation process, even in the case of a moving object whose evaluation value is “0” without being specified as a collision prediction target for a certain target moving object in the individual calculation process, the calculation unit 31a has the above-mentioned in the addition process based on the relative distance. It may be specified as a moving body approaching the target moving body and added to the evaluation value.

After completing the addition process for the evaluation value, the calculation unit 31a returns to the calculation process of FIG. The calculation unit 31a sequentially processes steps S55, S57, and S59 in FIG. 27 for the specified target moving objects, and repeats the processing until all of the target moving objects are processed.
The calculation unit 31a obtains evaluation values for all of the specified target moving objects, and after completing the addition processing on the evaluation values, returns to step S51 again and repeats the same processing.

Therefore, the arithmetic unit 31a repeatedly calculates and registers the evaluation value of each specified target moving object by repeating the arithmetic processing, and updates the registered contents of the evaluation value database 34b as needed.

FIG. 32 is a diagram illustrating an example of the evaluation value database 34b after the addition processing based on the relative distance.
In FIG. 32, in the evaluation value of the target moving object having the moving object ID “1005”, in the column of the evaluation values of the moving objects other than the evaluation target, the column corresponding to the moving object ID “1001” has “ "200" is registered, and "200" is registered as an evaluation value in a column corresponding to the mobile unit ID "1003", and "0" is registered in other fields.

In the evaluation value database 34b of FIG. 32, for example, the moving object with the moving object ID “1003” is specified as a collision prediction target with respect to the target moving object with the moving object ID “1005”, whereby the evaluation value is set to “200”. Suppose It is also assumed that the moving object with the moving object ID “1001” is specified as a moving object approaching the target moving object with the moving object ID “1005”, and the evaluation value is “200”. Further, it is assumed that the notification determination threshold value for the evaluation value used to determine whether the determination unit 31b notifies the collision prediction result is set to “200”. That is, when the evaluation value is “200” or more, the determination unit 31b determines that the collision prediction result should be notified to the evaluation target.

As described above, in the individual calculation processing, “0” is set as an evaluation value for a moving object other than the collision prediction target. However, in the addition processing based on the relative distance, a plurality of approaching moving bodies are specified, and an addition value is added to the evaluation value of the target moving body for each of the plurality of approaching moving bodies.
For this reason, as shown in FIG. 32, a plurality of moving objects (the moving object with the moving object ID “1001” and the moving object with the moving object ID “1003”) correspond to the moving object with the moving object ID “1005”. The evaluation value is registered.

In this case, the target mobile unit (the mobile unit with the mobile unit ID “1005”) has a collision prediction result for the mobile unit with the mobile unit ID “1001” and a collision prediction result for the mobile unit with the mobile unit ID “1003”. Will be notified.
As described above, in the present embodiment, not only the collision prediction result related to the collision prediction target specified by the individual calculation processing but also the collision prediction result related to the approaching moving body specified by the addition processing based on the relative distance is transmitted to the target moving body. Notified.
The content of the collision prediction result for the approaching moving body is the same as the collision prediction result for the collision prediction target, and includes information such as the attribute of the approaching moving body and the direction in which the approaching moving body approaches.

The edge server 3 according to the present embodiment identifies a moving object having a relative distance of 100 m or less and a negative displacement of the relative distance as an approaching moving object among other moving objects, and approaches the moving object at the target moving object. The additional value is added to the evaluation value of the moving object.
In other words, the edge server 3 adds the relative distance between the target moving object and another moving object and an added value based on the displacement of the relative distance to the evaluation value of the approaching moving object in the target moving object.

The collision prediction time calculated by the individual calculation processing is based on the movement information (position information, azimuth information, and speed information) of the target moving body and other moving bodies, and the target moving body and the other moving bodies are used. The collision prediction time when a collision occurs is obtained. Therefore, the edge server 3 determines that the traveling directions (azimuth information) of the target mobile unit and the other mobile units intersect or overlap with each other, and that the relationship between the speed of the target mobile unit and the speed of the other mobile unit is different. It is determined whether or not a collision occurs.
Therefore, even when the target moving body is approaching another moving body, the edge server 3 determines that the possibility of collision between the approaching target moving body and the other moving body is low. There is.

On the other hand, in the present embodiment, the edge server 3 uses the relative distance between the target moving object and another moving object and the added value based on the displacement of the relative distance as the evaluation value of the approaching moving object in the target moving object. to add. Thereby, the collision prediction result is provided to the mobile terminal of the target moving object when the possibility of a collision occurs due to the approach of the target moving object and another moving object, which does not appear in the collision prediction time. In the evaluation value.

In step S121 in FIG. 28 of the present embodiment, among the moving objects other than the evaluation object, a moving object whose relative distance to the target moving object is within 100 m and whose relative distance is negative is specified as an approaching moving object. However, the threshold of the relative distance for determining the approaching moving body is not limited to 100 m.

In the present embodiment, assuming that the threshold of the relative distance is 100 m, even if the target moving body and the other moving body are both vehicles and run at 50 km / h along the facing directions, the relative moving distance If a notification is made immediately after the vehicle is within 100 m and both vehicles start a stop operation, they can stop each other. For this reason, in this embodiment, the threshold value of the relative distance is set to 100 m.

On the other hand, when one of the target moving object and the other moving objects is a vehicle and the other is a pedestrian, the threshold of the relative distance can be made smaller.
Therefore, the threshold value of the relative distance may be changed according to the attributes (vehicle or pedestrian) of the target moving object and other moving objects.
Furthermore, the threshold of the relative distance may be configured to change according to the displacement of the relative distance.

Also, in step S154 in FIG. 31 of the present embodiment, a case has been illustrated in which it is determined whether or not the road curves within 100 m ahead of the vehicle of the target moving object. However, the threshold value of the distance for determining whether there is a curve ahead is not limited to 100 m.

In the present embodiment, assuming that the vehicle is traveling at 50 km / h, assuming that the vehicle is traveling at 50 km / h, it takes several seconds until the vehicle reaches the curve, assuming that the threshold for determining the presence or absence of the curve ahead is 100 m. . In the present embodiment, the threshold value of the distance for determining the presence or absence of a forward curve is set to 100 m because the notification can be made in advance during these several seconds.
Therefore, the threshold may be set to a longer distance if the notification can be made in advance before the vehicle reaches the curve.
Further, the threshold value of the distance for determining the presence or absence of a forward curve may be changed according to the speed of the vehicle.

Further, in the present embodiment, in addition to the individual calculation processing for calculating the evaluation value using the movement information (position information, azimuth information, and speed information) of the target moving body and the other moving bodies, Although the case where the addition process based on the relative distance between the moving objects registered in the relative distance database 34c is further performed is illustrated, the calculation unit 31a does not perform the individual calculation process and based on the relative distance between the moving objects. The evaluation value between the target moving object and another moving object may be obtained only by the addition processing.

In this case, the evaluation value for determining whether or not to notify the collision prediction result is determined based on the relative distance between the target moving object and each of the other moving objects, and the displacement of the relative distance. If the relative distance to another moving object is within a range of a predetermined relative distance and the relative distance decreases so as to approach each other, it can be determined that a possibility of collision has occurred. As a result, regardless of the traveling directions of the target moving object and the other moving objects, other moving objects having a possibility of collision with the target moving object can be widely evaluated, and the necessity of notification can be appropriately determined. it can.

[Specific example by addition processing based on relative distance]
[Specific example 1]
FIG. 33 is a diagram for describing a specific example of the addition process based on the relative distance, and is a diagram illustrating a pedestrian 7 crossing a crosswalk 90 at an intersection and a vehicle 5 turning right at the intersection.
In FIG. 33, the pedestrian 7 and the vehicle 5 are both traveling in accordance with the traffic light.
It is assumed that the pedestrian 7 and the vehicle 5 are registered as moving objects in the dynamic information map M1 and the moving object database 34a.
The vehicle 5 is equipped with the on-vehicle device 50, and the pedestrian 7 carries the pedestrian terminal 70.
In the following description, the notification determination threshold is set to “200”, and the edge server 3 determines that the evaluation value between the target mobile unit and another mobile unit is the notification determination threshold value (“200”). If so, it is assumed that the collision prediction result is notified to the target moving body.

(A), (b), (c), and (d) in FIG. 33 show, in chronological order, a case where the vehicle 5 enters an intersection and turns right when the pedestrian 7 starts to cross the pedestrian crossing 90. ing. Further, the distance d between the pedestrian 7 and the vehicle 5 gradually decreases in the order of (a) to (d) in FIG. Further, it is assumed that the relative distance between the pedestrian 7 and the vehicle 5 is within 100 m in any of the cases (a) to (d). Further, the speed of the vehicle 5 is assumed to be higher than 5 km / h.

In (a), (b), (c), and (d) of FIG. 33, when the above condition is satisfied, the pedestrian 7 is specified as the approaching moving body when the vehicle 5 is set as the target moving body. When the pedestrian 7 is set as the target moving body, the vehicle 5 is specified as the approaching moving body.
In the following description, a case will be mainly described in which the vehicle 5 is set as the target moving body and the pedestrian 7 is set as the approaching moving body. Similar processing is performed.

In (a) of FIG. 33, the pedestrian 7 and the vehicle 5 do not collide in the collision prediction based on the position information, the azimuth information, and the speed information because the traveling directions (arrows in the figure) are parallel to each other. The edge server 3 (the calculation unit 31a thereof) does not specify the pedestrian 7 as the collision prediction target of the vehicle 5 in the individual calculation processing. The edge server 3 does not specify the vehicle 5 as a collision prediction target for the pedestrian 7. Therefore, the edge server 3 sets the evaluation value for the vehicle 5 when the pedestrian 7 is the target moving object to “0” in the individual calculation processing. Similarly, the edge server 3 sets the evaluation value for the pedestrian 7 when the vehicle 5 is the target moving object to “0” in the individual calculation processing.

Also, in the addition processing based on the relative distance, since the pedestrian 7 is on the sidewalk 91, the edge server 3 does not add the addition value to the evaluation value for the pedestrian 7 of the vehicle 5 as the target moving body. (Step S141 in FIG. 30).
Therefore, at the stage (a) in FIG. 33, the edge server 3 does not notify the vehicle 5 of the collision prediction result regarding the pedestrian 7, and notifies the pedestrian 7 of the collision prediction result regarding the vehicle 5. No notification.

Next, in (b) of FIG. 33, the pedestrian 7 walks on the pedestrian crossing from the sidewalk 91 to the sidewalk 92, and the vehicle 5 enters the intersection and is about to start a right turn.
Further, in (b) of FIG. 33, the pedestrian 7 has not yet passed the center of the road (the middle point of the pedestrian crossing 90), and the distance to reach the sidewalk 92 ahead in the traveling direction is 2 m or more. Is also walking at a long point.

At this time, since the pedestrian 7 and the vehicle 5 cross each other in the traveling directions (arrows in the figure), there is a possibility that the pedestrian 7 and the vehicle 5 are specified as collision prediction targets by collision prediction based on position information, direction information, and speed information. However, the intersection position T where the traveling directions intersect each other is relatively distant, and it may be predicted that no collision will occur depending on the speed of each other. Therefore, whether or not the edge server 3 mutually specifies the vehicle 5 and the pedestrian 7 as the collision prediction targets in the individual calculation processing is a case-by-case basis.
Here, it is assumed that the edge server 3 does not mutually identify the vehicle 5 and the pedestrian 7 as collision prediction targets. Therefore, the edge server 3 sets the evaluation value of the vehicle 5 of the pedestrian 7 as the target moving object to “0” in the individual calculation processing. Similarly, the edge server 3 sets the evaluation value of the pedestrian 7 in the vehicle 5 as the target moving object to “0” in the individual calculation processing.

In the case of (b) in FIG. 33, as shown in (c) and (d) in FIG. 33, if the pedestrian 7 continues to cross the pedestrian crossing 90 and the vehicle 5 continues to turn right, The possibility that the pedestrian 7 and the vehicle 5 collide on the sidewalk 90 is high.

Here, the pedestrian 7 is on the pedestrian crossing 90 and has not yet passed the center of the road. Therefore, in the addition processing based on the relative distance, the edge server 3 adds the added value “200” to the evaluation value of the pedestrian 7 in the vehicle 5 as the target moving body (Step S147 in FIG. 30). Therefore, the evaluation value of the pedestrian 7 in the vehicle 5 as the target moving body is “200”.
Therefore, the evaluation value of the pedestrian 7 in the vehicle 5 is equal to or greater than the notification determination threshold, and the edge server 3 notifies the vehicle 5 of the collision prediction result regarding the pedestrian 7 to the stage (b) in FIG. . The edge server 3 performs the same processing on the pedestrian 7 as the target moving body. Therefore, the edge server 3 notifies the pedestrian 7 of the collision prediction result regarding the vehicle 5 at the stage (b) in FIG.

As described above, even if the vehicle 5 and the pedestrian 7 are not mutually specified as the collision prediction targets in the individual calculation processing, the edge server 3 performs the calculation based on the relative distance in the stage of (b) in FIG. By the addition processing, the evaluation value of the vehicle 5 in the pedestrian 7 as the target moving object and the evaluation value of the pedestrian 7 in the vehicle 5 as the target moving object are increased to be equal to or more than the notification determination threshold value. Thereby, the edge server 3 notifies both the pedestrian 7 and the vehicle 5 of the collision prediction result.
As a result, the vehicle 5 and the pedestrian 7 can be made to recognize each other before the situation as shown in (c) and (d) in FIG. 33, and a collision can be avoided. .

[Specific example 2]
FIG. 34 is a diagram for describing another specific example of the addition processing based on the relative distance, and is a diagram illustrating a pedestrian 7 crossing a crosswalk 90 at an intersection and a vehicle 5 turning right at the intersection.
(A), (b), (c), and (d) in FIG. 34 show, in chronological order, a case where the vehicle 5 enters an intersection and turns right while the pedestrian 7 is crossing the crosswalk 90. Is shown.

In the specific example 1, the pedestrian 7 is located on the sidewalk 91 immediately before the vehicle 5 enters the intersection ((a) in FIG. 33). On the other hand, in the specific example 2, as shown in FIG. 34A, immediately before the vehicle 5 enters the intersection, the pedestrian 7 is located on the pedestrian crossing 90 at a position passing through the center of the road. Is shown. That is, the specific example 2 shows a case where the pedestrian 7 walks ahead of the vehicle 5 by about half the road width as compared with the specific example 1.
In FIG. 34A, the pedestrian 7 is on the pedestrian crossing 90 and passes through the center of the road, but the point where the distance to reach the sidewalk 92 ahead in the traveling direction is longer than 2 m. Assume that you are walking.
Other conditions are the same as in the first embodiment.

In (a) of FIG. 34, the pedestrian 7 and the vehicle 5 do not collide in the collision prediction based on the position information, the azimuth information, and the speed information because the traveling directions (arrows in the figure) are parallel to each other. The edge server 3 (the calculation unit 31a thereof) does not mutually specify the vehicle 5 and the pedestrian 7 as collision prediction targets. Therefore, the edge server 3 sets the evaluation value of the vehicle 5 of the pedestrian 7 as the target moving object to “0” in the individual calculation processing. Similarly, the edge server 3 sets the evaluation value of the pedestrian 7 in the vehicle 5 as the target moving object to “0” in the individual calculation processing.

Further, the pedestrian 7 is walking at a point on the pedestrian crossing 90, passing through the center of the road but reaching a sidewalk 92 with a distance longer than 2 m. Therefore, in the addition processing based on the relative distance, the edge server 3 adds the added value “100” to the evaluation value of the pedestrian 7 in the vehicle 5 as the target moving body (Step S148 in FIG. 30).
Therefore, the evaluation value of the pedestrian 7 in the vehicle 5 as the target moving body is “100”.

In the present embodiment, the edge server 3 notifies the collision prediction result when the evaluation value is “200” or more. Therefore, when the evaluation value of the pedestrian 7 in the vehicle 5 as the target moving object is “100”, the edge server 3 does not notify the vehicle 5 of the collision prediction result.
However, if for some reason the evaluation value obtained by the individual calculation processing becomes “100” or more and exceeds the notification determination threshold (“200”), the edge server 3 notifies the collision prediction result.

In other words, if an addition value smaller than the notification determination threshold is added by the addition processing based on the relative distance, the addition value is added even when the evaluation value obtained only by the individual calculation processing does not reach the notification determination threshold. Thus, the result of the collision prediction can be notified. Thereby, the situation between the vehicle 5 and the pedestrian 7 when the edge server 3 determines to notify the collision prediction result can be further diversified.

Next, in (b) of FIG. 34, the pedestrian 7 has crossed the pedestrian crossing 90 immediately before the sidewalk 92, and the vehicle 5 is approaching the intersection and is about to start a right turn.
In this case, it is the same as (b) in FIG. 33, and it is a case-by-case whether or not the edge server 3 specifies the vehicle 5 and the pedestrian 7 as collision prediction targets in the individual calculation processing.
Here, it is assumed that the edge server 3 does not mutually identify the vehicle 5 and the pedestrian 7 as collision prediction targets. Therefore, the edge server 3 sets the evaluation value of the vehicle 5 of the pedestrian 7 as the target moving object to “0” in the individual calculation processing. Similarly, the edge server 3 sets the evaluation value of the pedestrian 7 in the vehicle 5 as the target moving object to “0” in the individual calculation processing.

On the other hand, it is assumed that the pedestrian 7 has crossed the pedestrian crossing 90 immediately before the sidewalk 92 and is located within 2 m before reaching the sidewalk 92, for example.
In this case, the edge server 3 does not add to the evaluation value of the pedestrian 7 in the vehicle 5 as the target moving body in the addition processing based on the relative distance (Step S146 in FIG. 30).

In the case of (b) in FIG. 34, thereafter, as shown in (c) and (d) in FIG. 34, the pedestrian 7 quickly finishes crossing the pedestrian crossing 90 and reaches the sidewalk 92. Therefore, the possibility that the pedestrian 7 and the vehicle 5 collide is low.
In the vehicle 5 and the pedestrian 7 in such a relationship, the need to notify the collision prediction result is low.

For this reason, the edge server 3 does not add to the evaluation value of the pedestrian 7 in the vehicle 5 as the target moving body in the addition processing based on the relative distance. Therefore, the evaluation value of the pedestrian 7 in the vehicle 5 as the target moving body is “0”. Therefore, the edge server 3 does not notify the vehicle 5 of the collision prediction result.
Also, the edge server 3 performs the same processing in the case of the pedestrian 7 as the target moving body. Therefore, the edge server 3 does not notify the pedestrian 7 of the collision prediction result.

As described above, in the present embodiment, even if the edge server 3 specifies the approaching moving body approaching the target moving body, if the necessity of notifying the collision prediction result is low, the edge server 3 Do not notify the collision prediction result. Thereby, it is possible to suppress unnecessary notification of the collision prediction result.

[Specific example 3]
FIG. 35 is a diagram for describing another specific example of the addition process based on the relative distance, and is a diagram illustrating a pedestrian 7 crossing a crosswalk 90 at an intersection and a vehicle 5 turning left at the intersection.
In FIG. 35, (a), (b), and (c) show, in chronological order, a case where the vehicle 5 enters an intersection and turns left when the pedestrian 7 starts crossing the pedestrian crossing 90.
In the specific example 3, it is assumed that the conditions regarding the vehicle 5, the pedestrian 7, and the like are the same as those in the specific example 1 except that the vehicle 5 turns left at the intersection and passes the pedestrian crossing 90 where the pedestrian 7 crosses. Therefore, in (a), (b), and (c) of FIG. 35, when the vehicle 5 is the target moving object, the pedestrian 7 is specified as the approaching moving object, and when the pedestrian 7 is the target moving object. The vehicle 5 is specified as an approaching moving body.

35A, the pedestrian 7 and the vehicle 5 have traveling directions (arrows in the figure) parallel to each other, and the pedestrian 7 is on the sidewalk 91.
Therefore, the edge server 3 does not notify the vehicle 5 of the collision prediction result regarding the pedestrian 7 as in the case of (a) in FIG. You will not be notified.

In (b) of FIG. 35, the pedestrian 7 walks on the pedestrian crossing from the sidewalk 91 to the sidewalk 92, and the vehicle 5 enters the intersection and starts to turn left.
Here, the pedestrian 7 is on the pedestrian crossing 90 and has not yet passed the center of the road.
Therefore, even if the pedestrian 7 is not specified as the collision prediction target by the individual calculation process, the edge server 3 adds the addition value “200” to the evaluation value of the pedestrian 7 in the vehicle 5 by the addition process based on the relative distance. It is added (step S147 in FIG. 30).
Therefore, the evaluation value of the pedestrian 7 in the vehicle 5 is equal to or greater than the notification determination threshold, and the edge server 3 notifies the vehicle 5 of the collision prediction result regarding the pedestrian 7 as in the case of (b) in FIG. Then, the pedestrian 7 is notified of the collision prediction result regarding the vehicle 5.

In the case of (b) in FIG. 35, thereafter, as shown in (c) in FIG. 35, when the pedestrian 7 continues to cross the pedestrian crossing 90 and the vehicle 5 continues to turn left, the pedestrian 7 There is a high possibility that the pedestrian 7 and the vehicle 5 collide.
On the other hand, the edge server 3 notifies both the pedestrian 7 and the vehicle 5 of the collision prediction result.
As a result, the vehicle 5 and the pedestrian 7 can be made to recognize each other before the situation as shown in FIG. 35 (c), and a collision can be avoided.

[Specific Example 4]
FIG. 36A is a diagram for explaining another specific example of the addition process based on the relative distance, in which two

vehicles

5A and 5B head on a two-lane road R10 in the same direction (the direction of the arrow in the figure). FIG.
Vehicle 5A is traveling in lane R10a, and vehicle 5B is traveling in lane R10b adjacent to lane R10a.
In FIG. 36A, it is assumed that the road R10 is curved within a range of 100 m ahead of the vehicle 5A.

It is assumed that the vehicle 5A and the vehicle 5B are registered as moving objects in the dynamic information map M1 and the moving object database 34a.
Also, it is assumed that the vehicle 5A has the in-vehicle device 50 mounted thereon, but the vehicle 5B does not have the in-vehicle device 50 mounted thereon. Therefore, in the following description, a case will be described in which the vehicle 5A is a target moving body and the vehicle 5B is an approaching moving body.
In the following description, an addition process based on a relative distance will be mainly described.

For example, consider a case in which the relative distance between the vehicle 5A and the vehicle 5B is within 100 m, the vehicle runs at about 50 km / h, and the vehicle 5A is gradually approaching the vehicle 5B.
In this case, the edge server 3 specifies the vehicle 5B as the approaching mobile when the vehicle 5A is the target mobile.

Here, if the road R10 does not curve within a range of 100 m ahead of the vehicle 5A, the edge server 3 does not perform the addition in the addition processing based on the relative distance (step S154 in FIG. 31).
As described above, when both vehicles are running side by side or traveling in the same lane, even if both vehicles are gradually approaching, if both vehicles are traveling normally on a straight road, It is not necessary to actively notify the vehicle in such a state of the collision prediction result.
Therefore, when the vehicle 5A and the vehicle 5B are running in parallel, if it is determined that the road is not curved within 100 m ahead, the edge server 3 does not perform the addition processing in the addition processing based on the relative distance. .

個別 In addition, individual calculation processing between the vehicle 5A and the vehicle 5B is performed. The edge server 3 obtains an evaluation value of the vehicle 5A with respect to the vehicle 5B by the individual calculation processing. When the evaluation value of the vehicle 5A with respect to the vehicle 5B by the individual calculation processing exceeds the notification determination threshold, the edge server 3 notifies the vehicle 5A of the collision prediction result regarding the vehicle 5B.

On the other hand, as illustrated in FIG. 36A, when the road R10 is curved within a range of 100 m in front of the vehicle 5A, the edge server 3 performs the addition processing based on the relative distance to the vehicle 5B in the vehicle 5A that is the target moving body. The added value “200” is added to the evaluation value of (step S155 in FIG. 31).
When the vehicle 5A and the vehicle 5B enter the curve, the vehicle 5A and the vehicle 5B need to turn the steering wheel according to the curve. At this time, for example, even if the following vehicle 5A turns the steering wheel according to the curve, if the vehicle 5B running ahead does not turn the steering wheel, the possibility of contact with each other increases. Therefore, the edge server 3 performs an addition process.

If the edge server 3 adds the addition value “200” to the evaluation value of the vehicle 5A for the vehicle 5B, the evaluation value becomes equal to or larger than the notification determination threshold.
Therefore, even if the vehicle 5B is not specified as the collision prediction target of the vehicle 5A in the individual calculation processing, the edge server 3 notifies the vehicle 5A of the collision prediction result by the addition processing based on the relative distance.
As a result, the vehicle 5A can be notified of the possibility of contact with the vehicle 5B before entering the curve, and the vehicle 5A can be alerted.

[Specific Example 5]
FIG. 36B is a diagram for describing another specific example of the addition processing based on the relative distance, in which the vehicle 5A is traveling in the lane R11 of the road R10 in the direction of the arrow in the figure, and the vehicle 5A is traveling. FIG. 5 is a diagram showing a state in which a vehicle 5B is traveling in a lane R12 facing a lane R11 in a direction of an arrow in the figure.

For example, consider a case where the relative distance between the vehicle 5A and the vehicle 5B is within 100 m and both vehicles are traveling at about 50 km / h.
In FIG. 36B, there is a curve ahead of vehicle 5A, and vehicle 5B is traveling along the curve. Therefore, the traveling direction of the vehicle 5A and the traveling direction of the vehicle 5B cross each other. The vehicle 5A and the vehicle 5B approach each other. Therefore, the edge server 3 specifies the vehicle 5B as the approaching mobile when the vehicle 5A is the target mobile.

Here, as shown in FIG. 36B, when the median strip C exists between the lane R11 and the lane R12, the edge server 3 does not perform the addition in the addition processing based on the relative distance (see FIG. 31). Step S152).

As described above, when the two vehicles are not approaching each other on the same road in the same direction and are not traveling in the same lane when the two vehicles are approaching each other, the two vehicles face each other. You are driving in the lane. At this time, if there is a median strip, it is considered that the possibility of collision between the two is extremely low.
For this reason, as shown in FIG. 36B, when the vehicle 5A and the vehicle 5B are traveling on opposite lanes (lane R11 and lane R12) and there is a median strip C, the edge server 3 In the addition process based on the above, the addition value is not added to the evaluation value.
In addition, the individual calculation process between the vehicle 5A and the vehicle 5B is performed. The edge server 3 obtains an evaluation value of the vehicle 5A with respect to the vehicle 5B by the individual calculation processing. Therefore, the edge server 3 determines whether or not to notify the collision prediction result according to the evaluation value of the vehicle 5A with respect to the vehicle 5B by the individual calculation processing.

On the other hand, when there is no median strip between the lanes R11 and R12, the edge server 3 adds “50” to the evaluation value of the vehicle 5B in the vehicle 5A as the target moving object in the addition processing based on the relative distance. Is added (step S153 in FIG. 31).
In this case, the edge server 3 does not notify the vehicle 5A of the collision prediction result regarding the vehicle 5B.

Both vehicles are traveling on opposite lanes, and it is considered that they do not collide in principle.
However, there is no central divider between the lanes R11 and R12, and a combination of other factors may possibly cause a collision between the two vehicles. Therefore, the edge server 3 adds the additional value “50” to the evaluation value of the vehicle 5B in the vehicle 5A. Thereby, the edge server 3 notifies the collision prediction result to the vehicle 5A when the evaluation value by the individual calculation processing is “150” or more.
With this, when the vehicle 5 traveling in the oncoming lane exists, if for some reason the evaluation value obtained by the individual calculation processing becomes “150” or more, the evaluation value obtained by adding the addition value of the addition processing based on the relative distance becomes Exceeding “200”, the edge server 3 notifies the collision prediction result.

As described above, by adding the addition value smaller than the notification determination threshold value by the addition processing based on the relative distance, even when the evaluation value obtained only by the individual calculation processing does not reach the notification determination threshold value, the addition value becomes smaller. By being added, a collision prediction result can be notified.

[Others]
It should be noted that the embodiments disclosed this time are illustrative in all aspects and not restrictive.
For example, the types of attributes in the attribute information on the vehicle 5 and the pedestrian 7 are examples, and the present invention is not limited to this.
Further, the addition value added in the addition processing is also an example, and is not limited to this.

Further, in the above embodiment, a moving object other than the evaluation target is specified, and a moving object having the shortest collision prediction time is specified as the collision prediction target, and an addition process is performed between the evaluation target and the collision prediction target. Is performed, and the evaluation value is obtained. However, for example, an addition process may be performed on a moving object other than the evaluation target to obtain an evaluation value, and the evaluation value may be obtained by the determination unit 31b.

Further, in the above embodiment, the case where the determination unit 31b determines whether or not to notify the collision prediction result based on the notification determination threshold value is illustrated. For example, the evaluation values of the respective evaluation targets are compared with each other. Alternatively, it may be configured to determine whether to notify the collision prediction result based on the comparison result.
For example, the evaluation values of the respective evaluation objects may be compared with each other, and a determination may be made to notify the collision prediction result for a predetermined number of evaluation objects in the descending order of the evaluation values.
In this case, since the number of evaluation targets for notifying the collision prediction result does not exceed the predetermined number, it is possible to suppress an unnecessary load on the wireless communication system.
This makes it possible to appropriately provide the mobile terminal with a collision prediction result having a high need for notification while suppressing the load on the wireless communication system.

Further, in the above-described embodiment, the case where the notification determination threshold is set in advance is illustrated, but the determination unit 31b may be configured to adaptively change the notification determination threshold.
For example, the determination unit 31b may be configured to count the number of notifications of notification of the collision prediction result to each evaluation target in a past predetermined period, and change the notification determination threshold according to the number of notifications.
In this case, if the number of notifications is a value that causes an unnecessary load on the wireless communication system, the determination unit 31b adjusts the notification determination threshold to a larger value so as to reduce the number of notifications, and If the number of times is a value that can be determined to be a load having sufficient capacity for the wireless communication system, the determination unit 31b adjusts the notification determination threshold to a larger value so that the number of times of notification increases.
This makes it possible to appropriately provide the mobile terminal with a collision prediction result having a high need for notification while suppressing the load on the wireless communication system.

The scope of the present disclosure is defined by the terms of the claims, rather than the meanings described above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

1A to 1D Communication terminal 2 Base station 3 Edge server 4

Core server

5, 5A, 5B, 5C,

5D Vehicle

7, 7A, 7B Pedestrian 8 Roadside sensor 9 Traffic signal controller 31 Control unit

31a Operation unit

31b Judgment unit

31c Notification Unit

31d detection unit 32 ROM
33 RAM
34 storage unit 34a mobile object database 34b evaluation value database 34c relative distance database 35 communication unit 41 control unit 42 ROM
43 RAM
44 storage unit 45 communication unit 50 in-vehicle device 51 control unit 52 GPS receiver 53 vehicle speed sensor 54 gyro sensor 55 storage unit 56 display 57 speaker 58 input device 59 in-vehicle camera 60 radar sensor 61 communication unit 70 pedestrian terminal 71 control unit 72 storage Unit 73 display unit 74 operation unit 75 communication unit 81 control unit 82 storage unit 83 roadside camera 84 radar sensor 85 communication unit 90 crosswalk 91 sidewalk 92 sidewalk C median strip d distance D1 display D2 arrow D3 display D4 arrow D5 display D6 arrow G1 Obstacle G2 Building M1 Dynamic information map M2 Dynamic information map N1 to N4 Node R1, R2 Route R10 Road R10a Lane R10b Lane R11 Lane R12 Lane T Intersection S1-S4 Network slice V1, V2, V3 Output Surface

Claims

Acquisition for acquiring movement information indicating a moving state of the plurality of moving objects and appearance information of the plurality of moving objects based on dynamic map information in which dynamic information of the plurality of moving objects is superimposed on map information. Department and
A target mobile having a mobile terminal among the plurality of mobiles, and performing a collision prediction of whether or not a mobile other than the target mobile collides based on the movement information; And an arithmetic unit for calculating an evaluation value indicating a possibility of collision with the other moving object;
A determining unit that determines whether to notify the mobile terminal of a collision prediction result indicating a result of the collision prediction by the arithmetic unit,
The arithmetic unit includes:
Included in the collision prediction result, a collision prediction time until the target mobile body and the other mobile body collide,
An appearance attribute value according to the appearance information of each of the target moving object and the other moving object;
Obtained from the movement information, the target mobile object, and an action attribute value according to action information indicating the action content of each of the other mobile objects,
The evaluation value is obtained based on
The information providing system, wherein the determination unit determines whether to notify the collision prediction result to the mobile terminal based on the evaluation value.
The information providing system according to claim 1, further comprising: a notifying unit that notifies the collision prediction result to the mobile terminal that the determining unit determines to notify.
The calculation unit specifies, based on the collision prediction result, a moving body predicted to collide with the target moving body among the other moving bodies, and the collision of the moving body predicted to collide is performed. The information providing system according to claim 1, wherein the evaluation value is obtained using a predicted time, the appearance attribute value, and the behavior attribute value.
The mobile terminal further includes a control unit that causes the mobile terminal to execute a process of outputting the collision prediction result toward a user of the mobile terminal,
4. The mobile terminal according to claim 1, wherein the control unit controls the mobile terminal such that an output mode of the collision prediction result differs according to the appearance information and the behavior information of the moving body predicted to collide. 5. An information providing system according to any one of the preceding claims.
When the other moving object is a vehicle, the appearance attribute value includes an attribute value corresponding to a size of the vehicle, an attribute value corresponding to a shape of the vehicle, an attribute value corresponding to a color of the vehicle, and The information providing system according to any one of claims 1 to 4, wherein at least one of attribute values corresponding to information described on a license plate attached to the vehicle is included.
When the other moving object is a pedestrian, the appearance attribute value includes an attribute value corresponding to the height of the pedestrian, an attribute value corresponding to the clothes of the pedestrian, and an accessory value of the pedestrian. The information providing system according to any one of claims 1 to 4, wherein at least one of the attribute values is included.
2. The method according to claim 1, wherein when the other moving object is a vehicle, the action attribute value includes at least one of an attribute value according to a current traveling state and an attribute value according to a past traveling history. 3. Item 5. The information providing system according to any one of Items 4.
When the other moving object is a pedestrian, the action attribute value includes an attribute value corresponding to a walking place of the pedestrian, an attribute value corresponding to a trajectory of the pedestrian, and an operation with respect to a light color of a traffic light. The information providing system according to claim 1, wherein at least one of the attribute values according to the information is included.
The calculation unit obtains a relative distance between the target mobile unit and each of the other mobile units based on the position information of the plurality of mobile units included in the movement information, and a displacement of the relative distance. The information providing system according to any one of claims 1 to 8, wherein an addition value based on a displacement of the distance and the relative distance is added to the evaluation value.
An acquisition unit configured to acquire movement information indicating a movement state of the plurality of moving objects based on dynamic map information in which dynamic information of the plurality of moving objects is superimposed on map information;
A target mobile having a mobile terminal among the plurality of mobiles, and performing a collision prediction of whether or not a mobile other than the target mobile collides based on the movement information; And an arithmetic unit for calculating an evaluation value indicating a possibility of collision with the other moving object;
A determining unit that determines whether to notify the mobile terminal of a collision prediction result indicating a result of the collision prediction by the arithmetic unit,
The arithmetic unit includes:
Based on the position information of the plurality of moving objects included in the movement information, a relative distance between the target moving object and each of the other moving objects, and a displacement of the relative distance are determined, and the relative distance and the relative distance are calculated. Calculating the evaluation value based on the displacement of
The information providing system, wherein the determination unit determines whether to notify the collision prediction result to the mobile terminal based on the evaluation value.
A mobile terminal that receives the collision prediction result from the information providing system according to any one of claims 1 to 10, and outputs the collision prediction result to a user.
An information providing method for providing information to a mobile terminal,
Acquisition for acquiring movement information indicating a moving state of the plurality of moving objects and appearance information of the plurality of moving objects based on dynamic map information in which dynamic information of the plurality of moving objects is superimposed on map information. Steps and
The target mobile unit having the mobile terminal among the plurality of mobile units and a collision prediction of whether or not a mobile unit other than the target mobile unit collides is performed based on the movement information, and the target mobile unit A calculating step of obtaining an evaluation value indicating a possibility that the body and the other moving body collide with each other;
A determination step of determining whether to notify the mobile terminal of a collision prediction result indicating a result of the collision prediction by the calculation step,
In the calculation step,
Included in the collision prediction result, a collision prediction time until the target mobile body and the other mobile body collide,
An appearance attribute value according to the appearance information of each of the target moving object and the other moving object;
Obtained from the movement information, the target mobile object, and an action attribute value according to action information indicating the action content of each of the other mobile objects,
The evaluation value is obtained based on
In the determining step, an information providing method for determining whether to notify the collision prediction result to the mobile terminal based on the evaluation value.
A computer program for causing a computer to execute an information providing process of providing information to a mobile terminal,
Based on the dynamic map information obtained by superimposing the dynamic information of the plurality of moving objects on the map information, the computer may include moving information indicating a moving state of the plurality of moving objects and appearance information of the plurality of moving objects. An acquiring step for acquiring;
The target mobile unit having the mobile terminal among the plurality of mobile units and a collision prediction of whether or not a mobile unit other than the target mobile unit collides is performed based on the movement information, and the target mobile unit A calculating step of obtaining an evaluation value indicating a possibility that the body and the other moving body collide with each other;
A determination step of determining whether to notify the mobile terminal of a collision prediction result indicating a result of the collision prediction by the calculation step,
In the calculation step,
Included in the collision prediction result, a collision prediction time until the target mobile body and the other mobile body collide,
An appearance attribute value according to the appearance information of each of the target moving object and the other moving object;
Obtained from the movement information, the target mobile object, and an action attribute value according to action information indicating the action content of each of the other mobile objects,
The evaluation value is obtained based on
In the determining step, a computer program for determining whether to notify the collision prediction result to the mobile terminal based on the evaluation value.