Research On Model Predictive Control-based Trajectory Tracking For Unmanned Vehicles

Man is the most important factor in the driving environment of man-car-road.Traffic accidents often happen because of driving fatigue,drunk driving,lack of driving experience and bad road conditions.Therefore,people became the weakest part during driving.Therefore,the research of intelligent autonomous vehicle not only complies with the trend of vehicle development,but also helps people to escape from complex driving environment and alleviate traffic safety problems.The trajectory tracking control of autonomous vehicle is studied in this paper.In this paper,the vehicle monorail model and the magic formula tire model are used as the foundation of the model firstly.The traditional trajectory tracking control based on preview has been studied.The lateral control of the vehicle based on the pure tracking theory and longitudinal velocity following of the vehicle based on the professional PID control has been developed.The trajectory tracking effect is good under good condition in the CarSim/Simulink simulation platform simulation.However,the vehicle can not track the reference trajectory well under low ground adhesion,even the sideslip and cornering phenomenon of the vehicle will happen.The trajectory tracking control method based on the model prediction control algorithm has been further studied in this paper.In order to enhance the stability of the vehicle,the dynamic constraints of the sideslip angle of mass center,acceleration and the sideslip angle of tire are added to the tradition model prediction tracking algorithm.This algorithm can avoid instability of the vehicle caused by lacking to provide sufficient side force under the emergency condition.And the trajectory planning layer was built on the trajectory tracking control layer.The obstacle avoidance function in the process of vehicle tracking has been realized,and the adaptability of the vehicle to the road environment is improved.In order to verify and analysis of the algorithms by setting different speed and different road conditions in double lane conditions.The simulation results shown that,The model prediction control algorithm with dynamic constraints has the advantage of tracking in bad conditions.It has good adaptability to external environmental interference and internal factors interference in the vehicle.Finally,the simplified model prediction algorithm is applied to the real vehicle validation and verification.After modifying the test car,we choose the right site to verify and adjust the test car.The test results shown that,compared with the conventional method,the simplified model prediction algorithm is simple and practical and has good real-time performance.The control performance is improved effectively and the trajectory tracking effect is good in the selected site.The simulation results shown that the model prediction control algorithm has advantages to solve the trajectory tracking control of autonomous vehicles under the condition of high speed and low adhesion.And it has good robustness and adaptability to the condition of road friction,the speed alteration and the reference trajectory.