Research On Lane Change Trajectory Planning And Tracking Control Of Autonomous Vehicles In Highway Scenarios

As the most common driving behavior in high-speed scenes,lane change has a great impact on road traffic efficiency and traffic safety.Autonomous vehicles can replace the driver to complete lane changing behavior,and avoid road congestion and traffic accidents caused by the uncertain factor of “people” to the greatest extent.This paper takes highway lane changing behavior as the research object,based on the minimum Snap and model predictive control theory to study the trajectory planning and tracking control in lane changing behavior,and achieves the following research results and technological innovations.(1)A lane-changing trajectory generation algorithm for autonomous vehicles based on minimum Snap is designed.Firstly,the minimum Snap trajectory under unconstrained conditions is derived as a seventh-degree polynomial through the Hamiltonian function.Then,various constraints during the lane change process are considered,and the lane-changing trajectory sets that meet the constraints in different scenarios are solved by MATLAB.On this basis,a trajectory optimization function that takes into account economy and comfort is designed,and the optimal trajectory of lane changing vehicles in different scenarios is generated by simulation experiments.The results show that the optimal lane changing trajectory has comprehensive performance,which is conducive to trajectory tracking.Finally,in the same lane change scenario,compared with the fifth polynomial lane change trajectory,the analysis shows that the minimum distance between the lane change trajectory based on minimum Snap and the preceding vehicle is about 0.13 m higher than the fifth polynomial trajectory,which has higher safety.(2)Based on the model predictive control theory,a horizontal and longitudinal integrated controller considering vehicle dynamics is designed,which takes the front wheel angle and vehicle acceleration of the vehicle as control quantities,and takes the trajectory tracking accuracy and the smooth change of the control quantity as the purpose.At the same time,the safety of the vehicle and the limitation of the actuator in the high-speed scene are add,and the optimal control quantity of the system is continuously solved to achieve the tracking of the optimal lane changing trajectory.(3)The MATLAB/Simulink-CarSim co-simulation platform is built.Firstly,the feasibility and effectiveness of the controller are verified under the double-shifted condition.Then,tracking simulation tests are carried out in different lane changing scenarios to verify the effectiveness and rationality of the lane changing motion control algorithm.The simulation results show that the controller has small tracking error and high trajectory tracking accuracy in high-speed lane changing scenarios,which meets the requirements of lane change behavior control.