Research On Path Tracking Control Based On MPC Intelligent Driving

With the rapid development of the intelligent city construction process,the traditional control method to control the vehicle tracking curve path often has large position deviation and poor tracking stability due to the lack of consideration of path curvature information,which is difficult to meet the needs of urban intelligent traffic safety,it is an urgent problem that how to carry out efficient and real-time path tracking control for intelligent vehicles in curve path condition.Based on the model predictive control theory,a path tracking control method based on curvilinear path curvature and preview model is proposed to realize fast,stable and accurate path tracking control in curvilinear path scene,and solve a series of intelligent driving problems such as low turning accuracy and poor stability of intelligent driving in this paper,and complete the construction of related algorithms and framework.The main contents of this paper are as follows:(1)Building a vehicle system model for path following control.The kinematics and dynamics models of intelligent vehicle are analyzed in detail,the load on the wheel in all directions in the mathematical model is considered,the transverse and longitudinal dynamic model of vehicle based on Brush tire model is established.Through reasonable simplification,a vehicle dynamic model considering wheel ground interaction and tire slip ratio is proposed,which provides support for considering the change of road adhesion coefficient in tracking control process;furthermore,the wheel ground forces and parameters needed to be considered in the curved path tracking problem are analyzed,and a vehicle model suitable for both path tracking control and vehicle stability control is derived by synthesizing the accuracy and computational complexity of the model,which lays a foundation for the follow-up research of intelligent driving path tracking control.(2)The improved model predictive control algorithm is studied,and a path tracking control algorithm is proposed to adaptively adjust the weight coefficient of cost function according to tracking deviation and road curvature,which improves the path tracking accuracy and vehicle driving stability.Firstly,the dynamic model of the intelligent vehicle is established.According to the characteristics of the mechanical structure of the vehicle,the constraint conditions and optimization functions that need to be satisfied in the design process of the control method are analyzed.According to the intelligent driving operation rules,the fuzzy logic controller is designed.The weight coefficient of optimization function in MPC control can be adjusted with the change of tracking position deviation and path curvature.So that the vehicle has a higher tracking accuracy in the process of driving,and can ensure that the vehicle has better stability.The design of the controller is completed,and the simulation and experimental results show that the method can further reduce the position deviation,and the yaw rate is smaller and the whole car body runs more smoothly.(3)Aiming at the tracking control of curvilinear path,a path tracking control method is proposed,which adaptively adjusts the control strategy according to the parameters of curvilinear path.Firstly,the vehicle longitudinal speed controller is designed according to the path curvature.Considering the influence of road curvature and vehicle speed on preview distance,a preview model is established to adaptively adjust vehicle speed and road curvature.According to the mathematical model and the empirical formula,the preview distance of different lengths can be obtained adaptively.The model predictive controller is designed based on the dynamic model.Finally,combined with dynamic preview distance and switching rules,the hybrid switching control method is designed,and the experimental results show that the method has better tracking performance and stability than other control methods.(4)A mobile robot platform is designed,and the feasibility of the proposed path tracking control algorithm is verified by experiments.The control platform of autonomous navigation mobile robot is designed and built,and the path tracking control algorithm designed in this paper is transplanted to the robot system.The mobile robot platform is developed.The feasibility of the path tracking control algorithm is verified by experiments.The mobile robot platform is designed and built,and the path tracking control algorithm designed above is transplanted into the robot system.The feasibility of the intelligent vehicle path following control method proposed in this paper is verified by the relevant experimental analysis.The tracking error is within the allowable range,which can meet the accuracy requirements of the actual work.The feasibility of the intelligent vehicle path tracking control method proposed in this paper is verified by the relevant experimental analysis,and the tracking error is within the allowable range,which can meet the accuracy requirements of the actual work.To sum up,aiming at the requirements and problems of curve path tracking control,this paper proposes a path tracking control method based on curve path parameters to adapt to the control strategy.The relationship between preview distance,vehicle speed and road curvature is established in the preview model,then,the weight coefficient of the cost function in the MPC algorithm is adaptively adjusted according to the tracking deviation and road curvature,which can track the curve path quickly,stably and accurately.The control method proposed in this paper is suitable for path tracking control in complex curvature changing conditions.