Deep Reinforcement Learning for Autonomous Vehicle Control

Deep learning provides a new class of biological inspired methods (artificial neural networks) which can
outperform the earlier state-of-the-art techniques as attested by the Imagenet challenge currently dominated by CNNbased
methods. From another hand, reinforcement learning enables the supremacy of algorithms over Humans in complicated
games (e.g. chess, go, etc.) with DeepBlue and Alphago and more recently Starcraft with Alphastar. Deep reinforcement
learning offers a promising framework allowing to build strong models using a positive loop where the algorithm can
experiment and learn to optimize its decision from its errors. Following this idea, our goal is to develop a deep reinforcement
learning paradigm to control autonomous vehicles based on the information provided by its sensors.

Keywords : deep learning, artificial intelligence, autonomous vehicle, optimal control.