Research on Optimal Shift Schedule for Multi-speed Electric Vehicle Based on Trajectory Prediction

Based on dynamic programming in the frame of moving horizon control, an optimal control strategy for multi-speed electric vehicle shifting schedule considering the curvature of the road ahead is proposed. Based on the on-board visual perception capacity with map information, the obtained information of the lane line in front of the vehicle is obtained as the future trajectory. With the multi-point preview longitudinal driver model, the vehicle speed prediction is performed on multiple preview points on the future remote trajectory, acting as predicted information for following optimization. With the obtained predicted vehicle speed sequence, the dynamic programming method is used to perform optimal predictive control in the prediction domain, and the optimization objective function including the current driver’s dynamic intention, battery energy consumption, shift frequency and other factors is investigated. A simulation model is built to compare the proposed strategy with classic dual-parameter economy and dynamic shifting schedule. The results show that the optimal shift schedule strategy based on preview trajectory can save the power consumption of multi-gear electric vehicles, ensure the dynamic performance and reduce the shifting frequency.