Hierarchical Optimal Velocity Planning and Torque Distribution Method for Dual-Motor-Drive EV in Urban Driving Scenarios

Dual-motor-drive electric vehicles (DMDEVs) have been explored for the superiority in dynamic response, energy saving and safety function. In this study, a hierarchical optimal speed planning algorithm and torque distribution method based on bi-level model predictive control (MPC) is proposed to enhance traffic mobility and energy efficiency in urban signalized intersections vehicle-following scenario. In the upper layer, the traffic information and the pre-vehicle states are integrated into velocity planning by Gaussian process (GP) prediction and solved by sequential quadratic programming (SQP). In the lower torque distribution layer, a real-time distribution algorithm combining rules and particle swarm algorithm (Rules-PSO) is employed to instantaneously optimize the torque distribution between the front and rear drive motors based on the planned velocity profiles. The simulation results demonstrate that the proposed method not only contributes to preferable driving comfort and vehicle-following behaviors, but effectively improves the fuel economy by 3.24% and 4.17% under standard condition and specific urban condition method when compare with the rule-based torque distribution strategy.