Intelligent Control Strategy of Urban Rail Train Based on Sarsa(λ) Algorithm

This paper proposes a train energy-saving control strategy based on Sarsa( ) reinforcement learning algorithm for urban rail transit. The strategy aims to achieve energy-saving driving while taking into account both timeliness and comfortability when the urban rail train is in autonomous driving mode. Finally, a simulation example is given for Beijing Railway Yizhuang Line Xiaohongmen to Xiao Village Station. Experimental results show that compared with traditional dynamic programming methods, Sarsa( ) algorithm can save 3.70% energy under the premise of meeting comfortability and timeliness requirements. The simulation results prove that Sarsa( ) algorithm has better energy-saving effect.