关键词: 模型预测控制；轨迹跟踪；换道控制；二自由度车辆模型；联合仿真

Abstract: Aiming at the lane-changing trajectory tracking control problem of intelligent vehicles, this paper proposes a trajectory tracking control strategy based on model predictive control (MPC). Firstly, a 2-degree-of-freedom vehicle dynamic model is established, and the state-space expression is adopted to describe the lateral motion characteristics of the vehicle. Secondly, a fifth-order polynomial lane-changing trajectory is designed as the desired path, which features continuous and smooth curvature and heading angle. Then, a trajectory tracking controller is designed based on the MPC algorithm, and the 2-degree-of-freedom vehicle model is discretized. Taking the lateral position deviation and heading angle deviation as state variables and the front-wheel steering angle as the control variable, a cost function containing tracking error and control increment is constructed.Constraint conditions including front-wheel steering angle, steering-angle change rate and lateral acceleration are introduced, and the optimal control problem is transformed into a quadratic programming problem for solution. Finally, the co-simulation of MATLAB/Simulink and CarSim is used to verify the feasibility and effectiveness of the proposed control strategy. Simulation results show that the designed controller achieves favorable tracking performance and high tracking accuracy under different vehicle-speed working conditions. The maximum lateral displacement error is kept within 0.045 m, and the front-wheel steering angle varies smoothly.

Key words: model predictive control; trajectory tracking; lane changing control; two-degree-offreedom vehicle model; co-simulation