Research On Lateral Control Strategy Of Autonomous Vehicle Based On Model Predictive Control

With the increasingly serious problems of traffic congestion,safety accidents and energy depletion caused by the year-by-year increase in car ownership,as well as the development of sensor technology,the improvement of computer data processing capabilities and the breakthrough of 5G communication technology,autonomous driving has become a research hotspot in academia and industry.As one of the core key technologies of automatic driving,lateral tracking control according to the planned path is an important prerequisite for its application in various complex traffic scenarios.This article has launched an in-depth study on the lateral tracking control of autonomous vehicles.The main contents are as follows:(1)Vehicle system dynamics modeling.Analyze the relationship between the tire lateral force and the cornering angle,establish a linear tire model,and use the integrated navigation data to identify the tire cornering stiffness to obtain the vehicle parameters of the test vehicle;establish a two-degree-of-freedom dynamic model of the vehicle,introduction The experimental platform and the design of real vehicle experiments,collect state parameters that can characterize the lateral dynamics of the vehicle and steering wheel angle,compare the gap between the model output and the actual vehicle state,and verify the accuracy of the model.(2)Research on lateral control strategy under normal operating conditions.Use the lateral tracking error as the objective function to establish the constraint bar and complete the vehicle lateral control strategy design.Use Car Sim and Simulink to perform path tracking co-simulation to verify the tracking performance of the control strategy;the design of the LQR controller is compared with the model predictive controller,and simulation tests are carried out under different vehicle speed conditions to prove that the model predictive controller can maintain good control effects at higher vehicle speeds.(3)Research on lateral control strategy under the extreme conditions of road adhesion.Considering the non-linear characteristics of the tires,the stable state of the vehicle is analyzed,and the boundary of the stable area of the vehicle that changes with the speed of the vehicle is derived;The lateral force of the front wheel is used as the control input to design the lateral controller to maintain the non-linearity of the tire while making the dynamic input and the state variable have a linear relationship,and the stability area boundary is used as the constraint condition of the model predictive controller;Completing the path tracking comparative simulation test under medium and high speed conditions to verify the control performance of the controller under the extreme conditions of tire-road adhesion.(4)Autonomous vehicle lateral control real vehicle experiment.Taking the "Xing Yuan" autonomous driving vehicle platform as the experimental object,the largecurvature turning,circular driving,double-line shifting working condition and high-speed lane change path tracking real vehicle experiments were completed on the autonomous driving test site in the school.Tracking error and stable vehicle state changes indicate that the designed controller has good tracking control effects in practical applications.