Verstappen AI

Overview

This project uses the NEAT (NeuroEvolution of Augmenting Topologies) algorithm to evolve artificial intelligence for a car racing game. The AI is trained to control a car on a game map, using a neural network to make decisions based on sensor data. The project also incorporates realistic physics, including oversteer, to create a more immersive and challenging game environment.

Features

• Evolutionary Algorithm: The NEAT algorithm is used to evolve the AI over multiple generations, selecting for cars that perform well on the game map.
• Neural Network Control: The AI uses a neural network to make decisions based on sensor data, such as distance to obstacles and speed.
• Game Map: The game map is rendered using Pygame, and the car's position and sensors are updated in real-time.
• Sensor Data: The car's sensors provide data on its surroundings, which is used as input to the neural network.
• Fitness Function: The fitness function rewards cars that stay on the track, avoid obstacles, and maintain a high speed.
• Realistic Physics: The game incorporates realistic physics, including oversteer, to create a more immersive and challenging game environment.

Usage

To run the project, simply execute the main.py file. This will start the NEAT algorithm, which will evolve the AI over multiple generations.

Configuration

The NEAT algorithm is configured using the config_file parameter, which specifies the settings for the evolution process. The game_map parameter specifies the game map to use, and the window parameter specifies the Pygame window to render the game in.

Code Structure

The code is organized into several modules:

• AIController: This class manages the NEAT algorithm and the game loop.
• Car: This class represents the car, and provides methods for updating its position and sensors.
• globals: This module defines global constants, such as the car's starting position and the game map dimensions.

Physics Implementation

The game incorporates realistic physics, including oversteer, to create a more immersive and challenging game environment. The car's physics are updated in real-time, taking into account factors such as speed, oversteer, acceleration, and friction.

License

This project is licensed under the MIT License. See LICENSE for details.

Contributing

Contributions are welcome! If you'd like to contribute to this project, please fork the repository and submit a pull request.