Personalized Automatic Driving System Based on Reinforcement Learning Technology

This paper proposes and designs a personalized automatic driving learning system. First, based on the current driving operation and habits of the driver, machine vision, and vehicle sensing, this paper extracts several index parameters using the driving simulator as a data collecting tool during the manual driving process. Then, this paper identifies the characteristics of the driving style, based on which the driving decision algorithm adopting deep reinforcement learning is trained in order to ensure that the self-driving vehicle can operate in accordance with the driving style and that the riding can be more comfortable and safer.