Integrated control for distributed in-wheel motor drive electric vehicle based on states estimation and nonlinear MPC

Abstract

In order to realize the integration of handling, lateral stability, rollover prevention, and ride comfort of distributed in-wheel motor drive electric vehicles (EVs), this study presents an integrated control strategy for distributed EVs based on states estimation through particle filter (PF) and nonlinear model predictive control (NMPC). To estimate the vehicle states including longitudinal velocity, lateral velocity, and sideslip angle, the random walk model is established to avoid complex tire forces calculation. Then PF completes estimation by measuring longitudinal and lateral acceleration and yaw rate. Considering handling, lateral stability, anti-rollover performance, and ride comfort comprehensively, proposed controller takes full advantages of distributed electric vehicle platform with four-wheel steering (4WS), four-wheel-drive (4WD), and active suspension technology to improve the overall performance in the NMPC structure. State/actuator constraints are also taken into account in the scheme. Finally, the co-simulation through commercial software Carsim and Simulink verifies the effectiveness of the vehicle states estimator and designed control algorithm. Results show that estimation error is controlled within 1% and overall performance of distributed EV is improved.