智能汽车换道轨迹跟踪横纵向控制方法研究OA

In allusion to the problems of insufficient trajectory tracking accuracy or reduced vehicle stability caused by a single control algorithm when intelligent vehicles actively change lanes in urban multi-lane environments,a horizontal and longitudinal decoupled trajectory tracking control strategy is proposed.In this strategy,the trajectory tracking is divided into horizontal path tracking and longitudinal path control.In the horizontal control,the feedforward predictor and linear quadratic regulator(LQR)feedback controller are combined,and the path curvature feedforward compensation is introduced to improve response speed and reduce steady-state errors.In the longitudinal control,the position-speed dual-layer PID architecture is adopted,the upper layer can calculate the target acceleration according to the trajectory error,and the lower layer can generate execution instructions by means of the throttle and brake calibration table,so as to achieve accurate speed adjustment.Based on the joint simulation platform of Matlab/Simulink,PreScan and CarSim,the results show that the proposed horizontal and longitudinal decoupling control strategy can maintain high trajectory tracking accuracy under different lane changing conditions,and the control error can meet the constraint requirements.In comparison with model predictive control(MPC),this strategy can effectively reduce the vehicle's centroid lateral deflection angle and yaw angular velocity,significantly improving the stability and comfort of the lane change process.