A basic complete ground-up implementation of the NEAT algorithm to enable cars to learn how to drive a user-created track. Built with inbuilt Python libraries and Pygame for visualization.
Currently, there is no functionality to draw the track.
- Draw the track using any image editing software ( Black colour => Path )
- The starting position of the car must be a path.
- The dimensions of the image should be 280x225 for performance
To run the script, you'll need the following Python packages:
- pygame==2.3.0
To get started, follow these steps:
- Clone this repository to your local machine.
- Create a virtual environment (optional but recommended).
- Install the required packages using pip.
- Run main.py
- Credits to The Coding Train for the inspiration. Nice Video Series on the topic: Introduction to Neuroevolution
- Also check out this cool video where I borrowed some ideas for the car Self-Driving Car with JavaScript
Have fun :)