An AI based High-speed Railway Automatic Train Operation System Analysis and Design

摘要

Recent years, the research and application of High-Speed Railway (HSR) automatic train operation (ATO) system are under fast development, while the safety, energy efficiency and passenger comfort of ATO systems still need improvement. On the other hand, Artificial Intelligence (AI) technology, for example, Deep Learning, has been widely applied in automata industry such as robot control and driverless vehicle. In this paper, we propose a new idea of improving train control system performance with AI technologies such as Deep Reinforcement Learning and Imitation learning, and describe the system objective, structure and development process. The details of key processes such as establishment of Train Running Condition Evaluation Index, acquisition and processing of relevant big data, construction of AI based automatic train operation model and the program of simulation and experiment are presented in this paper, which provides a brand new and practical idea to the development of High-Speed Railway automatic train operation systems.