Autonomous Racing using a Hybrid Imitation-Reinforcement Learning Architecture

Abstract--In this work, we present a rigorous end-to-end control strategy for autonomous vehicles aimed at minimizing lap times in a time attack racing event. We also introduce AutoRACE Simulator developed as a part of this research project, which was employed to simulate accurate vehicular and environmental dynamics along with realistic audio-visual effects. We adopted a hybrid imitation-reinforcement learning architecture and crafted a novel reward function to train a deep neural network policy to drive (using imitation learning) and race (using reinforcement learning) a car autonomously in less than 20 hours. Deployment results were reported as a direct comparison of 10 autonomous laps against 100 manual laps by 10 different human players. The autonomous agent not only exhibited superior performance by gaining 0.96 seconds over the best manual lap, but it also dominated the human players by 1.46 seconds with regard to the mean lap time. AutoRACE Simulator depicting a "Need for Speed" 2013 Ford Mustang GT A The task of autonomous driving has been researched for typical racing line is defined following an out-in-out maneuver, several years now, and although engineers have been able wherein the vehicle drives form outer side of the track up to demonstrate successful deployment of self-driving cars in to the turn-in point, then turns in while aiming for the the real-world, their reliability is still questionable.