关键词: 自动驾驶, 轨迹跟踪控制, 模型预测控制, 多参数优化, 模糊控制

Abstract: A multi-parameter optimized model predictive control（MPC）trajectory tracking control strategy is proposed to address the problem of large tracking position error at large curvature paths for autonomous vehicle lateral control.The trajectory tracking MPC controller is built according to the vehicle dynamics model and objective function，and the vehicle speed，lateral position error and yaw angle error are taken as fuzzy inputs，and the output front wheel angle acts on the vehicle.The prediction time domain，control time domain and weight matrix of the MPC controller are optimized in real time through fuzzy control，and the Carsim／Simulink joint simulation is completed under different speeds of the double-shifted line trajectory and different road adhesion coefficients to validate the effectiveness of the control strategy.Our simulation results show the MPC multi-parameter optimization algorithm is superior to the MPC traditional algorithm and the MPC single-parameter optimization algorithm.Meanwhile，the average trajectory tracking accuracy is improved by 27.4% in the high adhesion road；the maximum yaw angle error is reduced by 27.3% in the low-adhesion road，demonstrating it better balances the tracking accuracy and the stability of the maneuver.