racing-evolutionary-agent

Winner of RythmHack's Most Creative Use of AI and ML Prize Track.

Train your own self-driving race car!
This game uses the NEAT (NeuroEvolution of Augmenting Topologies) algorithm to evolve neural networks capable of navigating a pre-designed race track. Tweak parameters, experiment with evolution, and watch each AI generation get better at cornering, avoiding walls, and maintaining speed.

Features

• AI driver learns automatically through evolution
• Neural networks dynamically grow and mutate (NEAT)
• Interactive gameplay with menu-driven configuration
• Real-time stats and performance display

Installation

Requirements

• Python 3.10 — 3.12 recommended
• pip installed

1. Clone the repo

git clone https://github.com/jmagali/racing-evolutionary-agent.git

2. In the terminal, navigate to the directory where the repository was cloned, e.g.,

C:\Users\User\pygame-projects\racing-evolutionary-agent

3. Install the required Python libraries

pip install -r requirements.txt # This installs the required libraries

4. Run the game

python menu.py # This is the main file

How It Works

Each car is controlled by a small neural network. NEAT evolves:

• network weights
• network connections
• hidden nodes (topology)

Cars that drive farther score higher and pass their “genes” to the next generation through:

• crossover
• mutation
• speciation to preserve innovation

Over time, the population learns to race aggressively and efficiently.