End-to-End Driving in a Realistic Racing Game with Deep Reinforcement Learning

Etienne Perot, Maximilian Jaritz, Marin Toromanoff, Raoul de Charette

Abstract: We address the problem of autonomous race car driving. Using a recent rally game (WRC6) with realistic physics and graphics we train an Asynchronous Actor Critic (A3C) in an end-to-end fashion and propose an improved reward function to learn faster. The network is trained simultaneously on three very different tracks (snow, mountain, and coast) with various road structures, graphics and physics. Despite the more complex environments the trained agent learns significant features and exhibits good performance while driving in a more stable way than existing end-to-end approaches.