Generate random tracks of different shapes and sizes that adhere to the Formula Student Rules for the Formula Student Driverless competitions in terms of track width, cone spacing and corner radius. This script was developed by Formula Student Team Delft, used for an FSSIM based simulator or the Formula Student Driverless Simulator.
The tracks are generated using a bounded Voronoi diagram, which is created from a uniformly sampled set of points. From this Voronoi diagram, regions are selected based on one of three selection modes. By interpolating the selected region's vertices using a cubic spline, a track is generated.
n_points: Number of points to generate a Voronoi diagram from.n_regions: Number of Voronoi regions (cells) to select.min_bound: Minimum x and y bound of Voronoi diagram.max_bound: Maximum x and y bound of Voronoi diagram.mode: Method used to select Voronoi regions.Expand: Find the n closest neighbours to the first randomly selected region.Extend: Select a region randomly. Draw a line with random orientation and find the closest n regions to this line.Random: Select n regions randomly.
Please note that not all settings result in stable behaviour, and a track might not be found for your selected settings.
track_width: Track width in meters.cone_spacing: Cone spacing in meters.length_start_area: Length of the start area in meters.curvature_threshold: Maximum curvature (minimum radius of curvature) threshold in 1/meters.straight_threshold: Minimum curvature threshold for the starting area in 1/meters.
plot_track: Plot the resulting track.visualise_voronoi: Visualise the Voronoi diagram and selected regions.create_output_file: Create yaml output track file.output_location: Location to output track file.sim_type: Selection between output format for different simulators
- python >= 3.7.8
- numpy >= 1.18.5
- scipy >= 1.4.1
- matplotlib >= 3.1.3
- shapely >= 1.7.1
- yaml >= 0.2.5
pip3 install -r requirements.txtpython3 main.pyThis method is based on Ian Hudson's Race-Track-Generator.