A self-driving car simulation using NeuroEvolution of Augmenting Topologies (NEAT) in Python. This project demonstrates the use of evolutionary algorithms to train neural networks for autonomous vehicle control. The car is trained to navigate through a simulated environment, learning to make driving decisions based on its sensors.
NeuroEvolution: Uses NEAT to evolve neural networks that control the self-driving car. Simulation Environment: A simulated driving environment where the car can interact with obstacles and road features. Sensor Inputs: The car uses simulated sensors (e.g., distance sensors) to make driving decisions. Training Visualization: Visual representation of the car's performance and evolution over time.
Python 3.6 or higher neat-python library pygame for simulation visualization
After running the simulation, you will see a visual representation of the car navigating through the environment, with the NEAT algorithm evolving neural networks to improve driving performance over time.