Research On The Trajectory Tracking Control Of Intelligent Vehicle Based On Extension Theory

With the development of intelligent technology,intelligent vehicle has become an important direction of the development of vehicle industry.Trajectory tracking control is one of the key technologies of intelligent vehicle.At medium speed and high speed,the deterioration of tracking accuracy and stability greatly restricts the improvement of intelligent vehicle.Facing the different requirements about control difficulty of trajectory tracking control under large-scale road driving conditions,the single control method cannot effectively improve the global performance of trajectory tracking control.Therefore,this paper based on the extension control theory with the idea of area division control,firstly researches the extension switching trajectory tracking control strategy of intelligent vehicle.By dividing the trajectory tracking control state of intelligent vehicle into different areas,different control methods are adopted in different control areas to improve the accuracy of trajectory tracking control.Secondly,considering the targets of position error and heading angle error in trajectory tracking control,a parallel extension control structure is adopted to improve the representation ability of correlation functions.In order to solve the problem of jitter in the extension switching control,the game coordinated control method is introduced to design the extension game trajectory tracking coordinated control strategy to further improve the performance of the trajectory tracking controller.The main contents are as follows:(1)The tire dynamics model,vehicle dynamics model and preview error model are established to provide the basis for the design of trajectory tracking controller and the evaluation index of tracking performance.On this basis,the extension switching trajectory tracking controller is designed based on the extension control framework.The controller structure is divided into two layers.The upper layer of measurement modes recognition selects a group of characteristic quantities to describe the real-time state of trajectory tracking control.Adaptive extension set areas division method of error tolerance zone based on curvature is adopted and the upper layer realizes three measurement modes recognition in the classical domain,extension domain and non-domain.The lower layer of switching control algorithm combines PID feedback control,additional correlation function feedforward-feedback control and complete feedforward-feedback control strategies to realize the switching control in three measurement modes.The joint simulation results of MATLAB / Simulink and Carsim show that the extension switching control can improve the accuracy of trajectory tracking control.And compared with single control method,the precision of extension switching control is improved by 21.48% on average.However,the results also show that the correlation function with one group of characteristic quantities weakly represents the tracking control state,and the improvement of tracking accuracy is not obvious in the same measurement mode.And,the stability of the trajectory tracking control is affected with the jitter of the control quantity generated by the extension switching control at the switching points.(2)In order to enhance the representation ability of correlation function to improve the control accuracy in the same measurement mode,and solve the jitter problem of the extension switching control,the extension game trajectory tracking coordinated control strategy is designed.In the upper layer of parallel extension measurement modes recognition strategy,considering that the trajectory tracking control needs to meet the requirements of position error and heading angle error,two groups of errors and their derivatives are selected to form the parallel extension sets.And then the parallel extension measurement recognition strategy is designed.In the lower layer of parallel extension game coordinated control strategy,based on the principle of extension control and the additional correlation function feedforward-feedback control,the parallel control outputs in different measurement modes are obtained.On this basis,the game coordinated control strategy is introduced to establish the differential game model of parallel extension control weights,and the game coordinated strategy optimizes the output weights of parallel extension control to solve the problem of jitter caused by extension switching control.The joint simulation results of MATLAB / Simulink and Carsim show that compared with the extension switching controller,the accuracy of the optimal extension game trajectory tracking coordinated controller is increased by about 22.23% on average,and the jitter of stability is decreased by about 24.2% on average.The extension game coordinated controller improves the accuracy and stability of the controller effectively.(3)Based on intelligent driving platform of Chery ‘ARRIZO 5E' electric vehicle,the lane line is detected by camera and the road information is obtained.The designed trajectory tracking control strategy based on extension theory is compiled into an executable algorithm of vehicle controller,and the trajectory tracking control test of lane center keeping is carried out in the actual road conditions with medium speed and high speed.The test results show that the trajectory tracking control method based on extension theory proposed in this paper can meet the driving requirements of the actual vehicle,and compared with single PID control and feedforward-feedback control,the extension switching controller proposed in this paper improves the accuracy of trajectory tracking control.The tracking error of bend is decreased by about 20%.At the same time,the extension game coordinated controller realizes smooth and coordinated switching of various control strategies.Compared with the extension switching controller,the tracking error of bend is decreased by about 11.1% and the jitter of control output is reduced by about 19.2%.The extension game coordinated controller improves the accuracy and stability of trajectory tracking control.