Trajectory Tracking Control Of Intelligent Commercial Vehicle

Intelligent vehicle is the key content of modern vehicle engineering research field and the most important component of Intelligent Transportation System(ITS).Intelligent vehicle trajectory tracking control,as a core technology of intelligent vehicles,is the premise and foundation to realize intelligent vehicle.At present,intelligent passenger vehicles have been studied to reach a high level of automatic driving,while the research of intelligent commercial vehicles is still in the stage of partial automatic driving.Thus,the rese arch for intelligent commercial vehicle trajectory tracking control is of great significance.Therefore,based on Trucksim/Simulink joint simulation platform,the lateral control of trajectory tracking for intelligent commercial vehicles is studied in this thesis.Regarding the problem of low precision and side slip and rollover facilely in the process of trajectory tracking control,a dynamic model predictive control method considering vehicle mass change and longitudinal speed control is proposed,which can solve the problems of low precision of intelligent commercial vehicle trajectory tracking control and vehicle side slip and rollover.Finally,the control effect of intelligent commercial vehicle trajectory tracking controller is tested by double line shifting,snake-shaped line and emergency obstacle avoidance,and the evaluation index of "Human-Vehicle Closed-Loop Maneuverability Evaluation Index" is introduced.The main work includes:1.Regarding the low lateral control precision issues in the process of the intelligent commercial vehicle trajectory tracking control,the lateral control method of intelligent commercial vehicle trajectory tracking is studied.the commonly used trajectory tracking control algorithms at foreign and domestic are studied to solve the problem of low accuracy of trajectory tracking control for intelligent commercial vehicles,including the geometric model control,optimal curvature model control,kinematics model predictive control and dynamic model predictive control of intelligent commercial vehicles.In addition,the simulation experiments are carried out with different speed under dry and wet pavement by double line-shifting vehicle test method.Then,the comparisons are carried out with above lateral control method.The results show that the dynamic model predictive control performance is the best.Furthermore,the problem of predictive control of dynamic model is deeply studied.In order to improve the accuracy of vehicle lateral trajectory tracking,radial basis function(RBF)neural network is used to compensate uncertainties in the dynamic model and disturbances in the external environment.2.Considering that the mass of commercial vehicle is one of the important parameters of dynamic model,an Adaboost-based SVR improved algorithm is proposed.Vehicle mass is an important parameter in the dynamic model,especially when commercial vehicles are under no-load and full-load conditions,it will have a significant impact on the trajectory tracking accuracy.An Ada_SVR algorithm is proposed,which is an improved SVR algorithm based on Adaboost,to solve the problem of low identification accuracy of three weak learning algorithms: Polynomial Regression(PR),Kernel ridge regression(KRR)and Support Vector Regression(SVR).The vehicle mass identification algorithm is simulated by Trucksim/Simulink joint simulation platform.The experimental results show that the proposed Ada_SVR algorithm has good identification accuracy.3.Regarding the side slip and rollover issues in the process of the intelligent commercial vehicle trajectory tracking,a longitudinal velocity control algorithm is designed for the lateral and longitudinal cooperative control of intelligent commercial vehicle.In terms of the problems that the side slip and rollover accidents may occur with a high speed in intelligent commercial vehicle trajectory tracking,a longitudinal speed is developed to deal with this challenge.The longitudinal control algorithm is divided into high level controller and low level controller.For the high level controller,the RBF-PID control algorithm is adopted.For the lower level controller,the control algorithm is composed of logic switching and inverse longitudinal dynamic model.Finally,the effectiveness of the proposed longitudinal controller is verified by setting different test scenarios,including different acceleration and deceleration,double line shifting,snake line and emergency obstacle avoidance.The experimental results show that the proposed controller has good speed tracking ability and can improve the accuracy of lateral trajectory tracking.At the same time,it can effectively avoid side slip and rollover accidents of commercial vehicles at high speed.4.In order to investigate the effect of intelligent commercial vehicle trajectory tracking control,the "Human-Vehicle C losed-Loop Maneuverability Evaluation Index" is introduced to illustrate the proformance of the tracking control algorithmIn order to evaluate the effect of the proposed intelligent commercial vehicle tracking control algorithm,the "Human-Vehicle C losed-Loop Maneuverability Evaluation Index" is introduced.The simulation experiments are carried out under three standard working conditions: double line shifting,snake line and emergency obstacle avoidance.Finally,a comprehensive evaluation analysises can be obtained from five indicators: trajectory tracking error,handling burden,side slip hazard,rollover hazard and driving road feeling.The evaluation results show that the proposed control algorithm has good control performance,but the control burden index is high.