The Automatic train stop control is one of the key technologies of automatic train operation, and its precision is usually required within ±30 cm. The high-precision stop control can ensure the efficient operation of urban rail transit system. If the train does not stop accurately, it not only affects the passengers’ getting on or off, but also could cause delays in the time table, etc. Therefore, conducting research on the train stop control method is of great significance.This paper focuses on urban rail trains, and aim at achieving the key technology of automatic train operation. After analyzing of the automatic stop process of urban trains, we use the Pade approximation method to deal with the brake system; Combining the train traction calculation model and the model of the braking system, we propose the goals of automatic train stop control method, namely to realize high-precision speed tracking performance on the premise of avoiding frequent switching control input.This paper proposes the automatic train stop control method based on the terminal sliding mode control principle. To enhance the adaptability of the control system, we introduce the parameter adaptive mechanism. Thus, the adaptive terminal sliding mode control algorithm is designed. The symbol function of the sliding mode control is replaced by the saturated function, which could prevent the control system from producing the chattering effect. Therefore, we could ensure that the train operation is comfortable. The control performance caused by different line disturbance is analyzed. Introducing the disturbance observer based on adaptive terminal sliding mode control framework and recognizing the observation value of disturbance as the adaptive perturbation estimation in the terminal sliding mode control, the robustness of the system is enhanced.This paper employs Matlab to build the simulation platform to verify the correctness and effectiveness of control algorithm. The simulation results show that the adaptive terminal sliding mode control based on disturbance observer has good robustness and adaptability. The proposed method could overcome the model parametric uncertainty and external disturbance, ensuring that the train could stop precisely and comfortably. |