Autonomous Drifting Using Reinforcement Learning

Autonomous vehicles or self-driving cars are prevalent nowadays, many vehicle manufacturers, and other tech companies trying to develop autonomous vehicles. One major goal of the algorithms is perform manoeuvres safely, even when some anomaly arises. To solve these kinds complex issues, Artificial Intelligence Machine Learning methods used. motion planning problems tires lose their grip on road, an should handle this situation. Thus paper provides Drifting algorithm using Reinforcement Learning. The based a model-free learning algorithm, Twin Delayed Deep Deterministic Policy Gradients (TD3). model trained six different tracks in simulator, which developed specifically for driving systems; namely CARLA.