DE102019216147B3 - Method for determining the drivability of an area of a site for a vehicle - Google Patents

Abstract

In a method (V) for determining the drivability of at least one area (50, 50 *) of a site for a vehicle (1), wherein the vehicle has at least one environment sensor (2) and at least one control device (3) which communicate with each other are connected, an environment (4) of the vehicle (1) is recorded by means of the at least one environment sensor (2), whereby image data (B) are generated. A 3D point cloud (P) is generated based on the image data (B). A surface is fitted by the 3D point cloud (P). Vehicle-specific accessibility criteria (Z) are applied to the fitted 3D point cloud, with the accessibility criteria (Z) being shaped by means of vehicle-specific width and angle limits. On the basis of this, the at least one area (50, 50 *) of the surroundings (4) is determined which corresponds to the accessibility criteria (Z). This area (50) is output as passable (50 *).

Description

The present invention relates to a method for determining a drivability of an area of a terrain for a vehicle with the features according to claim 1, a control device with the features according to claim 10, a computer program product with the features according to claim 11, and a vehicle with the features according to claim 12.

Vehicles in the off-road sector, for example construction or agricultural machinery, have only a limited ability to negotiate inclines and declines in the terrain without getting stuck. In the case of vehicles of this type, this is particularly problematic when they perform work functions in the field that are interrupted by getting stuck. This leads to failures in operation and loss of time. Freeing the vehicles is time-consuming and often also cost-intensive. Furthermore, getting stuck can pose a risk to a driver or another driver of the vehicle.

Out DE 10 2017 206 245 A1 a method and a device for detecting a terrain course, in particular for detecting elevations and / or depressions in the area around a vehicle, are known. Here, image data in the vicinity of the vehicle are recorded and the course of a horizon in the image data is compared with a predetermined reference course for the horizon. As a function of the deviation between the actual horizon and the reference course, elevations or depressions can then be determined in the terrain course.

The pamphlet DE 10 2016 110 461 A1 refers to the determination of road gradients. In order to determine the inclination of a lane area, the first lane area is divided into a predetermined number of strips. Surrounding points detected by a vehicle-side detection device are assigned to these in order to determine the respective height of the strips on the basis of the height coordinates of the surrounding points. A road incline results from the difference in the heights of two strips.

The pamphlet DE 10 2017 115 810 A1 discloses a method for autonomous parking of a vehicle based on a trajectory. In the method, the area around the vehicle is captured using image information and divided into areas that are classified with regard to their drivability. Vehicle-specific data is incorporated into the determination of a drivable area.

The pamphlet DE 10 2017 203 276 A1 suggests generating a trajectory in off-road scenarios using 2D sensors to record a 2D image of an environment. Another sensor system is used to determine a height profile of the environment. The pixels of the image are classified and projected into the height profile.

Based on the prior art, the present invention is based on the object of proposing a method by means of which drivable areas for off-road vehicles can be reliably identified.

Based on the aforementioned object, the present invention proposes a method for determining the drivability of an area of a terrain for a vehicle according to claim 1, a control device according to claim 10, a computer program product according to claim 11, and a vehicle according to claim 12. Further advantageous configurations and developments emerge from the subclaims.

In a method for determining the drivability of at least one area of a site for a vehicle, the vehicle having at least one environment sensor and at least one control device which are connected to one another, an environment of the vehicle is detected by means of the at least one environment sensor, whereby Image data are generated. A 3D point cloud is generated based on the image data. A surface is fitted by the 3D point cloud. Vehicle-specific accessibility criteria are applied to the fitted 3D point cloud, with the accessibility criteria being shaped by means of vehicle-specific width and angle limits. On the basis of this, the at least one area of the surroundings that corresponds to the accessibility criteria is determined. This area is issued as passable.

The vehicle is a land vehicle which is designed for an off-road area, and can be a commercial vehicle (commercial vehicle), e.g. B. a truck, an agricultural or forestry vehicle, namely an agricultural machine, or a construction machine, a snow groomer, a mining vehicle, a cleaning machine, or another utility vehicle, but also an off-road vehicle. The vehicle can be designed in such a way that it can perform automated driving functions and work functions from level 3 of the SAE J3016 autonomy levels. The vehicle has an external shape with certain unchangeable dimensions. In addition, the vehicle has a predetermined maximum weight which cannot be exceeded.

The vehicle has the at least one environment sensor. This is used to record the surroundings of the vehicle. Of course, the vehicle can have more than one environment sensor. The environment sensor is designed in such a way that it can generate image data from which a 3D point cloud can be generated. For example, the at least one environment sensor can be designed as a camera, for example as a mono or stereo camera, or as a LIDAR. A shape as an imaging radar or 3D ultrasound sensor would also be possible, in which case a range would be reduced compared to the shape using a camera or a LIDAR. The at least one environment sensor is arranged on the vehicle in such a way that it can detect the environment in the direction of travel in front of the vehicle. If the vehicle has several environmental sensors, these are arranged, for example, in such a way that a surround view can be implemented.

Furthermore, the vehicle has the control device which is connected to the at least one environment sensor, so that data and signals can be exchanged. This connection can be wired or wireless. The control device can, for. B. be designed as an ECU or domain ECU. In addition, the control device has a memory device in which data can be stored. The control device is set up to evaluate the image data that the at least one environment sensor supplies. Furthermore, the control device is set up to control vehicle systems, e.g. B. a drive system, a steering system, a braking system, o. Ä., Or to forward information to a display device, whereby this can be displayed. The display device can for example be designed as a display with or without an input function. For this purpose, the control device is connected to the vehicle systems and to the display device in such a way that data and signals can be exchanged. This connection can be wired or wireless.

In addition, the control device is set up to evaluate and utilize position data of the vehicle that can be determined by means of a position determination system. For this purpose, the control device is connected to the position determination system in such a way that data and signals can be exchanged. This connection can be wired or wireless. These position data preferably reflect the global position of the vehicle, for example these can be present as GPS coordinates or other suitable coordinates that have been determined by means of a global navigation satellite system (GNSS). Alternatively, the position data can also be specified without GPS coordinates or the like, for example as local position data, which z. B. can be determined on the basis of a vehicle odometry.

The environment of the vehicle is recorded by means of the at least one environment sensor, whereby image data are generated. The surroundings of the vehicle refer to the surroundings of the vehicle in which the vehicle is moving. The area around the vehicle is not flat, but has depressions and elevations. The surroundings of the vehicle thus represent an off-road area. The generated image data are based on the shape of the at least one surroundings sensor. For example, the area in front of the vehicle in the direction of travel is recorded by means of a stereo camera, whereby camera data are generated as image data. Alternatively, the area in the direction of travel in front of the vehicle z. B. captured by means of a LIDAR, whereby lidar data are generated as image data. In other words, the image data represent the recordings that are generated by the at least one environment sensor.

The 3D point cloud is generated based on the image data. The 3D point cloud is generated directly from the image data. This is done by means of the control device. The 3D point cloud thus depicts the vehicle's surroundings. The 3D point cloud is z. B. generated from the image data using the “Structure from Motion” process or using the “SLAM” process. These methods are known from the prior art.

A surface is fitted by the 3D point cloud. The area can either be used directly, or a grid can be generated by the fitting, which evaluates the area. This grid can be designed as a uniform or an adaptive grid. The surface is not to be understood as a flat surface. The fitting takes place, for example, by means of a regression method, but can also take place by means of alternative methods. In other words, an area is mapped in the point cloud.

Vehicle-specific accessibility criteria are applied to the fitted 3D point cloud, with the accessibility criteria being shaped by means of vehicle-specific width and angle limits. The accessibility criteria are set at the factory and stored in the storage device of the control device. In other words, the accessibility criteria define criteria which indicate which restrictions are specified for passable areas due to the shape of the vehicle. The accessibility criteria are specific to each vehicle. The width limitation is determined by the shape of the vehicle. This indicates the minimum width an area must have so that the vehicle can drive on it safely. The angle limitation is determined by the shape of the vehicle. This indicates the maximum uphill or downhill gradient an area may have so that the vehicle can drive on it safely. Driving safely means that the vehicle can pass the area without getting stuck, tipping, sinking or the like. Changing these accessibility criteria is only possible at the factory, for example. For example, a driver can have the accessibility criteria adjusted in such a way that the angle limitation is exceeded. If this is the case, this will be documented so that any damage to the vehicle is the responsibility of the respective driver.

Based on the application of the vehicle-specific accessibility criteria to the fitted 3D point cloud, the at least one area of the environment which corresponds to the accessibility criteria is determined. In other words, this at least one area has a sufficient width so that the vehicle can drive on it. Furthermore, the at least one area has an angle of incline or decline which is small enough that the vehicle can cope with it. Of course, more than one area can meet the accessibility criteria. Other areas in the vicinity of the vehicle that do not meet the accessibility criteria cannot be safely driven into by the vehicle. These are not considered further.

The at least one area is then output as drivable. That is to say, the vehicle can safely drive into this at least one area. Output means here on the one hand that the at least one area is identified as drivable by the control device and can be used for possible further method steps. In addition, the at least one area can be output to a driver of the vehicle by means of the display device, that is to say displayed to him.

The advantage of this method is that only that area or areas are output which the vehicle can drive. The experience of a driver of the vehicle who can safely identify such drivable areas himself is thus automated. This can increase safety, especially in vehicles that perform autonomous driving functions.

According to a further developing embodiment, based on the 3D point cloud, the area is additionally reduced to an occupancy card, which is designed as a 2D occupancy card or a 2.5D occupancy card, the occupancy card being displayable. The 2D occupancy map shows two room dimensions. The 2.5D occupancy map, on the other hand, depicts two spatial dimensions, with height information being stored in addition to each point. In other words, the third dimension, namely the height of the terrain, is modeled as an attribute of the location.

The occupancy map can be used to show which areas can be driven by the vehicle. This shows which areas are "occupied", i. H. are not passable. The occupancy card can be issued to a driver of the vehicle, for example by means of the display device of the vehicle. For example, a map of the surroundings or a recording of the surroundings can be overlaid with the 2D occupancy map or with the 2.5D occupancy map. The map of the surroundings can be generated, for example, using the image data and represents the immediate surroundings of the vehicle. Alternatively, the map of the surroundings can already be present and made available, for example, by a map provider. This can be done, for example, using commercially available navigation software.

In a further-developing embodiment, the 3D point cloud or the grid is combined with the environment map. In other words, the 3D point cloud or the grid is overlaid on this map of the surroundings. This means that the information from the 3D point cloud or the information from the grid is displayed over the map of the surroundings. Alternatively, the environment map can be added with the 3D point cloud or with the grid. The environment map has a coarser resolution than the 3D point cloud or the grid, but is complete. With this addition, the details from the 3D point cloud or from the grid flow into the map of the surroundings in order to obtain an improved map. This is preferably output to the driver of the vehicle, for example by means of the display device of the vehicle.

According to a further-developing embodiment, it is additionally checked whether the at least one area is connected to a position of the vehicle, the area being output as reachable if there is a connection. In this case, connected means that, based on the position of the vehicle, the vehicle can travel a distance in order to reach the at least one area. For example, the at least one area can directly adjoin the position of the vehicle. As an alternative to this, the at least one area can be arranged at a distance from the position of the vehicle, but can be reached via a road or other areas classified as drivable.

The check takes place on the basis of the image data by means of the control device of the vehicle. The position of the vehicle is determined by means of the position determination system based on the position data.

If the at least one area is classified as reachable, it is output to the driver of the vehicle, for example by means of the display device of the vehicle. For example, this area can be highlighted in the map of the area and / or the occupancy map, e.g. B. by means of a colored highlighting or by means of a lighting highlighting in which the corresponding area is illuminated more brightly than the other areas.

According to a further-developing embodiment, a position to be approached for the vehicle is additionally determined, with at least one trajectory being determined, which can be traveled by the vehicle, based on the at least one drivable area. The position to be approached is preferably determined by the driver of the vehicle. For example, the driver can enter and define the position to be approached using the input function of the display device or using an additional input device that the vehicle has.

Can be driven off means that the vehicle can drive along the trajectory. Starting from the current position of the vehicle, the control device determines the at least one trajectory that can be followed by the vehicle. This is the case when the position to be approached lies within the at least one drivable area, and when the at least one trajectory runs through the at least one drivable area or through several drivable areas. In addition, the at least one trajectory should only traverse reachable areas. For example, the at least one trajectory can be determined in such a way that the widest path is selected. Alternatively, the at least one trajectory can be determined such that the smoothest path is selected. Alternatively, the at least one trajectory can be determined in such a way that the path with the smallest angles in the terrain is selected.

If the driver specifies a position to be approached that is not located within the at least one drivable area, a warning message can be output to the driver by means of the display device.

Of course, more than one trajectory can be determined in order to get to the position to be approached. These different trajectories can be displayed to the driver by means of the display device. This can then select one of the trajectories that he would like to travel.

According to a further-developing embodiment, the driving dynamics of the vehicle are intervened on the basis of the at least one trajectory, so that the vehicle follows the trajectory. The driving dynamics of the vehicle are composed of a longitudinal movement, a lateral movement and a lifting movement, as well as yaw, pitch and roll movements.

The driving dynamics of the vehicle are influenced, for example, by means of the vehicle systems such as B. the drive system, the steering system, the brake system, o. Ä. The intervention in the driving dynamics takes place by means of the control device, which can control the vehicle systems. For example, the control device can control the actuators of the steering system in such a way that a steering angle is set on the wheels of the vehicle. For example, the control device can control the actuators of the drive system in such a way that the vehicle moves, accelerates or decelerates at a certain speed. For example, the control device can control the actuators of the brake system in such a way that the vehicle is decelerated. This following the trajectory can take place partially or completely autonomously.

According to a further developing embodiment, the at least one trajectory is displayed in a view of the surroundings of the vehicle. In other words, the at least one trajectory can be displayed in a map of the surroundings or in a camera image which is displayed by means of the display device. In addition, the trajectory can also be displayed on the occupancy map.

According to a further developing embodiment, the at least one drivable area is displayed in the view of the surroundings of the vehicle. In other words, the at least one drivable area can be displayed in a map of the surroundings or in a camera image which is displayed by means of the display device. This is done, for example, by means of color highlighting or by means of lighting highlighting. This has already been described in the previous description

According to a further-developing embodiment, a trained artificial neural network is used on the basis of the image data to additionally determine the nature of the soil, which is checked using the accessibility criteria. In this case, at least one driving area of the surroundings is determined which corresponds to the accessibility criteria, this driving area being compared with the at least one area in order to check the plausibility of the at least one area.

Soil condition refers to the condition of the ground to be driven on. For example, the ground to be driven on can have rock, rubble, vegetation, a body of water, mud, sand, gravel or the like. In other words, the nature of the ground is an alternative designation for the shape of the ground to be driven on.

The trained artificial neural network was trained at the factory using image data from various surfaces to be driven on. The training data include series of images in which the drivable areas are already marked. As a result, by means of a trained artificial neural network which is used by an artificial intelligence, it can be determined on the basis of the image data which soil conditions are present around the vehicle in the environment around the vehicle detected by the at least one environment sensor. The environment sensor is here preferably designed as a camera.

In addition, the accessibility criteria have been expanded to include ground conditions that are accessible to the vehicle. For example, it can be set at the factory that the vehicle cannot move on mud or sand.

The at least one driving area in the surroundings is determined which corresponds to the accessibility criteria with regard to the nature of the ground. That is, the image data are evaluated by means of the control device in such a way that the at least one driving area is selected which has the nature of the ground on which the vehicle is to be driven.

This driving area is compared with the at least one area that has already been classified as drivable in order to check the plausibility of this at least one drivable area. This is done by means of the control device. The determination of the drivable area by means of the accessibility criteria takes place before the determination of the driving area. This means that the driving area is less weighted in the plausibility check than the drivable area.

The plausibility check is advantageous since it can be determined with a higher degree of certainty whether the at least one area classified as drivable is actually drivable. This increases the safety for the vehicle.

The control device for the vehicle can be connected to at least one environment sensor of the vehicle, the control device having means that carry out the method which has already been described in the previous description. Connectable means that the control device, when it is used in the vehicle, is connected to the at least one environment sensor. The shape of the control device has already been described. The control device can use a computer program product, for example.

The computer program product comprises commands which, when the program is executed by the control device already described, execute the method which has already been described in the previous description. The computer program product can be present, for example, as a downloadable data stream or embodied on a data carrier.

The vehicle has the control device already described and the at least one environment sensor. The control device is connected to the at least one environment sensor. The vehicle and its vehicle systems have already been described in the previous description.

• 1 eine schematische Darstellung eines Fahrzeugs nach einem Ausführungsbeispiel,
• 2 eine schematische Darstellung einer Umgebung des Fahrzeugs,
• 3 eine schematische Darstellung einer Belegungskarte für das Fahrzeug, welche aus dem Umfeld aus 2 generiert wurde,
• 4 eine schematische Darstellung der Belegungskarte für das Fahrzeug aus 3, wobei die Belegungskarte nachbearbeitet ist,
• 5 eine schematische Darstellung der Belegungskarte für das Fahrzeug aus 4, wobei die Belegungskarte nochmals nachbearbeitet ist,
• 6 eine schematische Darstellung des Umfelds des Fahrzeugs, mit einer überlagerten Belegungskarte,
• 7 eine schematische Darstellung eines Verfahrens zur Bestimmung einer Befahrbarkeit eines Bereichs für das Fahrzeug.

Various exemplary embodiments and details of the invention are described in more detail with the aid of the figures explained below. Show it:

• 1 a schematic representation of a vehicle according to an embodiment,
• 2 a schematic representation of the surroundings of the vehicle,
• 3 a schematic representation of an occupancy card for the vehicle, which from the environment 2 was generated,
• 4th a schematic representation of the occupancy card for the vehicle 3 , whereby the occupancy card is post-processed,
• 5 a schematic representation of the occupancy card for the vehicle 4th , whereby the occupancy card has been processed again,
• 6 a schematic representation of the surroundings of the vehicle, with an overlaid occupancy map,
• 7th a schematic representation of a method for determining a drivability of an area for the vehicle.

1 shows a schematic representation of a vehicle 1 according to an embodiment. The vehicle 1 has a control device 3 on. This control device 3 is with an environment sensor 2 connected. This environment sensor 2 is shaped as a camera. The connection between the control device 3 and the environment sensor 2 is such that data and signals can be exchanged. In particular, image data from the environment sensor 2 to the control device 3 are forwarded, the control device 3 can evaluate this image data.

The vehicle also has 1 a display device 5 on. This display device 5 is thus within the passenger compartment of the vehicle 1 arranged that a driver of the vehicle the display device 5 can perceive and operate. The display device 5 can for example have a display. The display device 5 is with the control device 3 connected so that data and signals can be exchanged.

The vehicle 1 also has a drive system 6 on. This drive system 6 is with the control device 3 connected. The connection is such that the control device 3 the drive system 6 can drive. Between the control device 3 and the drive system 6 data and signals can be exchanged. For example, the control device 3 the drive system 6 control in such a way that it provides more drive energy, so that the vehicle 1 is accelerated. The control device 3 in other words, the drive system can 6 control in such a way that a driving dynamics of the vehicle 1 can be customized.

The vehicle 1 also has a positioning system 7th on. This position determination system 7th serves to get a global position of the vehicle 1 ascertain. This global position can be in GPS coordinates, for example. The position determination system 7th is with the control device 3 connected so that data and signals can be exchanged. For example, the position determination system 7th the determined global position of the vehicle 1 to the control device 3 hand off.

The vehicle 1 is designed as an off-road vehicle, here as an agricultural vehicle. The vehicle 1 moves in an environment 4th . This environment 4th is in 2 shown in more detail.

2 shows a schematic representation of an environment 4th of the vehicle. You can see an off-road environment, which has a rocky subsoil and vegetation. This environment is recorded by the vehicle's environment sensor. The image data thus generated can be output to the driver of the vehicle by means of the display device. For example, the recordings of the environment sensor can be transmitted directly to the display device.

One area is shown 50 , which can be driven over by a vehicle and a non-passable area 60 . This inaccessible area 60 shows, for example, rocks and excessive vegetation that the vehicle cannot drive over. This is determined by the vehicle's accessibility criteria. The accessibility criteria represent vehicle-specific width and angle limits for the vehicle. Only areas that have or meet these accessibility criteria can be accessed by the vehicle.

3 shows a schematic representation of an occupancy card M. for the vehicle, which from the environment 2 generated. The creation of the occupancy card M. is in 7th explained in more detail. The occupancy card M. is generated from a 3D point cloud. This 3D point cloud is generated using the image data generated by the vehicle's environment sensor. In the occupancy card M. is the non-passable area 60 and the presumably passable area 50 shown. The presumably passable area 50 represents the area that is not blocked by debris or vegetation.

4th shows a schematic representation of the occupancy card M. for the vehicle 1 out 3 , with the occupancy card M. is post-processed. Post-processing of the occupancy card M. takes place with the help of the accessibility criteria. Areas that do not meet these accessibility criteria because they are too narrow or have too steep an angle in the terrain are no longer displayed. This changes the area on the right edge of the screen. Now only the area is shown 50 that is passable 50 * is. This is almost congruent with the area 50 out 3 . There is still the non-passable area 60 shown.

5 shows a schematic representation of the occupancy card M. for the vehicle 4th , with the occupancy card M. has been reworked again. In 5 is a trajectory T drawn by means of the method described in 7th is shown, was calculated. In the occupancy card M. is a position X of the vehicle. At this position X the vehicle is at an initial point in time. There is also a position to be approached Y shown. To this position to be approached Y should the vehicle move along the trajectory T move. The trajectory T is only through the passable area 50 * guided and does not cross the non-passable area 60 . This prevents the vehicle from entering an inaccessible area 60 device and is thereby endangered.

6 shows a schematic representation of the environment 4th of the vehicle with an overlaid occupancy card M. . On the one hand, the same image is shown as in 2 as well as another shot. Both recordings are side by side arranged and thus represent a surround view of the vehicle. The joining of the two recordings is also referred to as image stitching.

The occupancy card M. has been placed over the actual camera images. This clearly shows in the images which areas are passable 50 * and which areas are not passable 60 are. This environment view U can be output to the driver of the vehicle by means of the display device of the vehicle. Thus, while driving the vehicle, the driver can clearly determine whether he is moving towards a drivable area or a non-drivable area 60 .

7th shows a schematic representation of a method V to determine the passability of an area 50 for the vehicle 1 . In a first step 101 an environment is created by means of the environment sensor of the vehicle 4th of the vehicle. This creates image data B. generated.

In a second step 102 is made using this image data B. a 3D point cloud P is generated. This is done using a Structure from Motion or a Visual SLAM process.

In a third step 103 an area is fitted by the 3D point cloud P. A grid can thereby be obtained.

In a fourth step 104a are vehicle-specific accessibility criteria Z applied to the fitted 3D point cloud P. The accessibility criteria Z are shaped by means of vehicle-specific width and angle limits. In an optional additional fourth step 104b is based on the 3D point cloud P the area in addition to an occupancy card M. reduced. This occupancy card M. can be, for example, a 2D occupancy card. 2D occupancy cards are in the 3 , 4th and 5 shown.

This occupancy card M. can be output to the driver by means of the display device of the vehicle.

In a fifth step 105 is based on the accessibility criteria Z and the 3D point cloud P that area 50 in the area determined which of the accessibility criteria Z corresponds.

In a sixth step 106 becomes that area 50 as passable 50 * issued, which precisely these accessibility criteria Z corresponds. The following steps are optional. These can either be implemented in full or only some of these steps can be implemented.

In a seventh step 107 it is also checked whether the passable area 50 * with one position X of the vehicle is connected. Connected means that the vehicle is starting from its current position X the area 50 * can reach without a non-navigable area 60 having to cross.

In an optional eighth step 108 also becomes a position to be approached Y set for the vehicle. Starting from the current position X of the vehicle and from the specified position to be approached Y and from the passable area 50 * becomes a trajectory T determines which the vehicle can travel to from its current position X to the position to be approached Y to get. In a subsequent further optional step 109 is based on the trajectory T intervened in the driving dynamics of the vehicle, so that the vehicle's trajectory T departs. The intervention in the driving dynamics takes place by means of the control device which accesses the drive system of the vehicle and controls it. The control takes place in such a way that the vehicle follows the trajectory T can leave.

In another optional step 110 becomes the passable area 50 * and / or the trajectory T in the environment view U of the vehicle. This is output to the driver of the vehicle by means of the display device.

This in 7th The process shown here runs continuously. That is, as long as the vehicle is in a driving mode and is moving in an off-road environment, the method V is applied.

List of reference symbols

1

vehicle

2

Environment sensor

3

Control device

4th

Surroundings

5

Display device

6th

Drive system

7th

Positioning system

50

Area

50 *

passable area

55

Driving area

60

not passable area

101

first step

102

second step

103

Third step

104a

fourth step

104b

fourth complementary step

105

fifth step

106

sixth step

107

seventh step

108

eighth step

109

ninth step

110

tenth step

B.

Image data

M.

2D occupancy map

P

3D point cloud

T

Trajectory

U

Environment view

V

Procedure

X

position

Y

position to be approached

Z

Accessibility criteria

Claims (12)

Method (V) for determining the drivability of at least one area (50; 50 *) of a site for a vehicle (1), the vehicle having at least one environment sensor (2) and at least one control device (3) which are connected to one another , in which - An environment (4) of the vehicle (1) is recorded by means of the at least one environment sensor (2), whereby image data (B) are generated, - A 3D point cloud (P) is generated based on the image data (B), - an area is fitted by the 3D point cloud (P), - Vehicle-specific accessibility criteria (Z) are applied to the fitted 3D point cloud (P), with the accessibility criteria (Z) being shaped by means of vehicle-specific width and angle limits, - Based on this, the at least one area (50, 50 *) of the surroundings (4) is determined which corresponds to the accessibility criteria (Z), - this area (50, 50 *) is output as passable (50 *). Procedure (V) according to Claim 1 , based on the 3D point cloud (P) the area is additionally reduced to an occupancy card (M), which is designed as a 2D occupancy card (M) or a 2.5D occupancy card, the occupancy card (M) being displayable. Method (V) according to one of the preceding claims, wherein the 3D point cloud (P) or the grid is combined with an environment map. Method (V) according to one of the preceding claims, wherein it is additionally checked whether the at least one area (50) is connected to a position (X) of the vehicle (1), and when there is a connection, the area (50, 50 * ) is output as reachable. Method (V) according to one of the preceding claims, wherein a position (Y) to be approached is also determined for the vehicle (1), at least one trajectory (T) being determined based on the at least one drivable area (50, 50 *) which can be driven off by the vehicle (1). Procedure (V) according to Claim 5 , the driving dynamics of the vehicle (1) being intervened on the basis of the at least one trajectory (T) so that the vehicle (1) travels along the trajectory (T). Procedure (V) according to Claim 5 or 6 , wherein the at least one trajectory (T) is displayed in a view of the surroundings (U) of the vehicle (1). Method (V) according to one of the preceding claims, wherein the at least one drivable area (50, 50 *) is displayed in the view of the surroundings (U) of the vehicle (1). Method (V) according to one of the preceding claims, wherein a trained artificial neural network based on the image data (B) is used to additionally determine a soil condition, which is checked using the accessibility criteria (Z), with at least one driving area (55 ) the environment (4) is determined which corresponds to the accessibility criteria (Z), this driving area (55) being compared with the at least one area (50, 50 *) in order to identify the at least one area (50) check plausibility. Control device (3) for a vehicle (1), wherein the control device (3) can be connected to at least one environment sensor (2) of the vehicle (1), the control device (3) having means which the method (V) according to perform one of the preceding claims. A computer program product comprising instructions which, when the program is executed by a control device (3), are generated according to Claim 10 , the procedure according to one of the Claims 1 to 9 To run. Vehicle (1) having a control device (3) according to Claim 10 and at least one environment sensor (2), the control device (3) being connected to the at least one environment sensor (2).

Images