Deep Reinforcement Learning for Decision-making and Control of Autonomous Driving: A Survey

In recent years, academic research in the field of Autonomous Driving (AD) has been favored. Intelligent driving technology is a multi-disciplinary integration technology, which the application of Deep Reinforcement Learning (DRL) to intelligent driving strategy, control and other fields has witnessed significant research efforts. This paper is a comprehensive survey of this body of work, which is conducted at three levels: First, the overview, mathematical modeling and algorithm classification of reinforcement learning are introduced. Second, the application of DRL in automatic driving of decision and control is reviewed comprehensively. Finally, an in-depth discussion is conducted on how the critical issues of AD applications regarding driving safety, interaction with other traffic participants and sample efficiency are addressed by the DRL models, and provides some prospects for the research in this field.