关键词: 主动四轮转向, 驾驶员模型, 模型预测控制, 操纵稳定性控制

Abstract: This paper proposes a linear time-varying model predictive control (LTV-MPC) method based on a nonlinear tire model for the stability control problem of active four-wheel steering with ground adhesion constraints and vehicle steering actuator peak constraints. In this method, the nonlinear tire model is locally linearized. The problem is converted into a Quadratic programming through the ground adhesion coefficient and wheel angle input constraints. The optimal front and rear wheel angles are obtained. The nonlinear seven degree of freedom vehicle dynamics model built based on Matlab/Simulink is simulated under the front wheel angle step input and double lane shift input conditions respectively. The results show under two working conditions, LTV-MPC helps reduce the peak the centroid sideslip angle of the vehicle by 29% and 19.62% respectively when compared with the LQR active four-wheel steering controller, improving the vehicle’s handling stability.