Research on Automatic Train Operation Performance Optimization of High Speed Railway Based on Asynchronous Advantage Actor-Critic

As a complex control problem, a high-speed train operation process has been widely studied, integrating safety, punctuality, energy saving, comfort, and other multi-objective requirements. In recent years, artificial intelligence (AI) algorithms and technologies have developed rapidly and have been widely used in the traditional automatic control industry. With the demand for the intellectual development of high-speed railways, AI technology provides new research directions and effective methods for developing automated train control technology. In this paper, based on the asynchronous advantage actor-critic algorithm (A3C) to study the optimization problem of high-speed railway automatic train operation (ATO) performance, and design a reasonable multi-objective performance evaluation index. Also, because of the problem that the A3C algorithm is prone to convergence oscillation or deterioration, an improved parameter updating method based on a weighted average of advantage value is proposed. The simulation results show that the enhanced A3C algorithm can improve the convergence stability of the A3C algorithm and improve the overall performance of the train operation.