Self-driving car in 2D environment, using NEAT genetic algorithm

2D simulation to train an autonomous car that must navigate a series of increasingly challenging tracks. It uses neural networks that evolve thanks to the NEAT genetic algorithm. For a complete overview of the simulation's implementation and the achieved results, check out the following report.

Modes

1. Simulation: in the main file, you can choose a range of values for key algorithm parameters, then simulate the behavior for a set number of generations. A simulation is run for each variation. At the end, graphs are generated to show which variations yields the best results. The simulation can run on multiple tracks consecutively.
2. Training: train the model on a set of parameters for an unlimited number of generations, with automatic saving of the best model.
3. Validation: validate the previously trained model on one or more selected tracks.
4. Change tracks: change the training and/or validation tracks.

Six tracks are used, which can be selected or deselected via the "Tracks" menu option.

How to run the project

1. Clone the repository

First, clone the repository to your local system:

git clone https://github.com/tuo-utente/tuo-progetto.git
cd AI-Autonomous-Car

2. Create a virtual environment

python3 -m venv venv
source venv/bin/activate

3. Install dependencies and run the program

pip install -r requirements.txt
python3 main.py

Report

For more details on the simulation and the achieved results, refer to the following: full report.