reinforced-race

A model car learns driving along a track using reinforcement learning

Unity virtual environment

This is a virtual environment the car can learn to drive in. Built with Unity it enables you to test your new parameters, tracks and algorithms before you take your car out to reality.

Open the Unity project in rl-race-virtual-env and run the game, then start the learning algorithm.

The Deep-Q-Learning reinforcement learning algorithm

The reinforcement learning algorithm is inspired from Mnhi et al. 2013. After starting the virtual environment run

python3 rl-race-learning/run.py --save

The module will connect to the virtual environment using a TCP connection.

See python3 rl-race-learning/run.py --help for more options.

The real model car

Coming soon