Research On Vehicle Transverse And Longitudinal Coupling Control Based On Adaptive Model Prediction Algorithm

With the rapid development of China’s economy and the growing maturity of the automobile industry,the research on global autonomous vehicles has become an inevitable trend.Among them,vehicle control is the basic problem of the research of automatic driving vehicle.Vehicle control mainly includes horizontal control and vertical control.There is a strong coupling relationship between the longitudinal and lateral nonlinear systems of the vehicle in the process of driving.The design of a secure and stable horizontal and vertical collaborative system has become a difficult and hot issue.At present,most scholars study more about the single lateral or longitudinal Lane assistant driving system,but less about the vehicle lateral and longitudinal coupling control of lane assistant system.Therefore,this paper was studied the vehicle lateral and longitudinal coupling control for lane following,using adaptive model predictive control algorithm,designs the lateral and longitudinal coupling lane following controller system based on the algorithm To improve the matching degree between the controller of the auxiliary driving system and the driver’s expectation,so as to meet the safety and comfort requirements of vehicle driving.Firstly,based on the analysis of the forces on the three degree of freedom vehicle in the driving process,the vehicle model and tire model were established to meet the current adaptive model predictive control algorithm,so that the vehicle model can meet the requirements of the internal discrete model of the current lane following controller.Then the prediction model modeling,model discretization,cost function establishment,rolling optimization and feedback control of the adaptive model prediction algorithm theory were derived,which laied the theoretical foundation for the design of longitudinal adaptive cruise controller,lateral lane keeping controller and transverse longitudinal coupling lane following controller.Secondly,the model predictive control algorithm was used to model the longitudinal adaptive cruise system(ACC),to discretize the state equation,to design the model predictive algorithm cruise controller,to simulate the scene of constant speed cruise,stable car following and emergency braking traffic,so as to verify the tracking effect of the cruise system controller under different working conditions.Then,the controller of lane keeping assist(LKA)system is designed by model predictive control algorithm,and the models of different grades of roads were built by using Carsim software.The intervention time of lane keeping controller was determined according to the deviation of lateral displacement and yaw rate,so as to prevent vehicle deviation and make it drive according to the expected path.The driving process of different grades of roads designed by Carsim software is compared with that of Simulink The simulation results of lane keeping controller based on the adaptive model predictive control was built by ink were compared to verify the simulation results of the designed lane keeping controller.The research results of vertical and horizontal coupling control of vehicles in this paper can provide the corresponding controller design requirements for the driveway assistant driving system of automatic driving vehicles in the future,and provide reference for the improvement of multi-objective performance of vehicles such as road safety,stability and comfort.It provides an effective verification method for improving the collaborative control of intelligent transportation.It has important theoretical significance and reference value for the design of assistant driving automatic driving vehicle.