Efficient network-constrained trajectory queries

The large search companies have very clearly shown that full-text search on very large datasets can be executed efficiently. In this paper, we show how querying spatio-temporal trajectory data can be converted to a full-text search problem. This allows for the reuse of efficient data and index structures from the full-text domain. The core idea is to convert a trajectory into a document consisting of spatial and temporal terms. For example, spatial terms are municipality names, zip codes, or road-network segment numbers. Temporal terms are, for example, morning, weekday, spring, and 2020. Using a dataset consisting of +62 million trajectories (24.9 billion GPS points) we show how to query this dataset efficiently. These queries cover spatial, temporal, and spatio-temporal queries.