Research on DQN Algorithm with Step Control Strategy Applied in ATO
Abstract— Achieving good speed control is the key to the operation of automatic train operation system. In this paper, the image processing method in deep learning is introduced. Combined with classical DQN algorithm and step control strategy, a speed controller with good tracking effect is designed. This method enhances the control accuracy and introduces the step control strategy into the DQN algorithm. With the combination of neural network and reinforcement learning, a good train speed control effect is achieved.
Index Terms— ATO control, reinforcement learning, DQN algorithm, step control.
Zhejiang Train Intelligent Engineering Technology Research Center Co., Ltd, CHINA
Yanmei Guo, Ziheng Wu, Xiangxian Chen
Department of Instrumentation Science and Engineering, Zhejiang University, CHINA
Cite: Yanmei Guo, Ziheng Wu, Xiangxian Chen, Zhujun Ling, "Research on DQN Algorithm with Step Control Strategy Applied in ATO," Proceedings of 2018 the 8th International Workshop on Computer Science and Engineering, pp. 48-53, Bangkok, 28-30 June, 2018.