Racing Line Optimization with PSO
This repository contains a racing line optimization algorithm in python that uses Particle Swarm Optimization.
This version was developed and tested with
python==3.8.12. The following modules are required:
How it works
First of all, we need to define the structure of the input data and which parameters the algorithm is optimizing.
The input data (i.e. the track layout) is stored in the
tracks.json file: it consists of an array of points defining the central line of the track and the track width (which is considered constant along the track).
Then, it will find the inner and outer track borders and define the search space of each sector (defined as a segment that goes from one border to the other of the track). The points through which the racing line passes, will move along these segments:
Run the algorithm
main.py script to see the optimizer work. Inside the main function you will have the possibility to change the hyper-parameters of the PSO algorithm.
To find the racing line, the algorithm will fit a cubic spline to the sector points and compute the vehicle’s speed at each point of the racing line with a simple formula using the coefficient of friction and the radius of the corner (optionally it is possible to add the down-force and the vehicle’s mass).
This project is under the MIT license. See LICENSE for more information.