Naturalistic Trajectories of 110 000 Vehicles Recorded at German Highways
The highD dataset is a new dataset of naturalistic vehicle trajectories recorded on German highways. Using a drone, typical limitations of established traffic data collection methods such as occlusions are overcome by the aerial perspective. Traffic was recorded at six different locations and includes more than 110 000 vehicles. Each vehicle's trajectory, including vehicle type, size and manoeuvres, is automatically extracted. Using state-of-the-art computer vision algorithms, the positioning error is typically less than ten centimeters. Although the dataset was created for the safety validation of highly automated vehicles, it is also suitable for many other tasks such as the analysis of traffic patterns or the parameterization of driver models.Large Scale 110 000 vehicles 45 000 driven kilometers 147 driven hoursHigh Quality and Variety Six different recording locations Different traffic states (e.g. traffic jams) Typical positioning error <10 cmEnriched DataPre-extracted information include: Surrounding vehicles Metrics like THW or TTC Driven maneuvers (e.g. lane changes)Easy StartProvided scripts for Matlab and Python: Visualization of recorded trajectories Maneuver classification (soon) Maneuver statistics (soon)