Research On Robust Control Algorithm Of Intelligent Vehicle Trajectory Tracking Based On Output Feedback

As the development trend of automobiles,automotive intelligence and electrification have received wide attention from domestic and foreign research scholars and government agencies.Trajectory tracking,as the main task of the control layer of intelligent vehicles,directly affects the safety and comfort of vehicle driving,and is the highlight of current vehicle control field.In order to meet the needs of the industry development,this thesis takes the four-wheel independent drive electric vehicles as the research object and studies the intelligent vehicle trajectory tracking controller design problem.In this thesis,the smart car trajectory tracking task is decomposed into three parts:lateral motion control,longitudinal motion control and torque allocation.A model-based robust control approach is used to design the controllers,so a car dynamics model is first established and verified in CarSim.For the vehicle longitudinal motion control,a feedforward+PID feedback control scheme is proposed.Assuming the car is driving on a float road with constant car mass and no environmental disturbance,the feedforward controller is designed.And then the environmental disturbance,turning resistance and total mass change during the car driving process are regarded as disturbances,the PID feedback controller is designed to stabilize the tracking error.In order to find the PID controller with robustness,the PID controller design problem is converted into a static output feedback problem by theoretical derivation.By introducing the coordinate transformation matrix into the pole configuration and H_∞performance constraints of the resultant system,the linear matrix inequality sufficient conditions for the bilinear matrix inequality constraints to hold are obtained.Finally,the multi-objective PID controller design algorithm is developed by combining the cone complement linearization and coordinate transformation matrix optimization methods.And the feedback controller is designed by this algorithm.A robust gain-scheduled PD control method is proposed to address the influence of tire cornering stiffness uncertainty,speed continuous change,and environmental disturbances on lateral motion during vehicle driving.The preview control model is divided into two segments according to the vehicle speed,and each segment is established separately with a triangle polytope linear parameter varying(LPV)system with parameter uncertainty.In order to design the controller,the robust gain-scheduled PD controller design problem is transformed into the SOF controller design problem,and then the bounded real lemma and pole configuration conditions with uncertain parameters are dimensionally expanded to obtain the equivalence conditions without uncertain parameters.The LMI sufficient conditions for the above equivalence conditions to hold are obtained by introducing the coordinate transformation matrix.In solving the initial coordinate transformation matrix,the subsystem of the LPV system at the polytope vertices is rewritten into a new polytope model according to the variation range of the uncertain parameters,and the NM-HS algorithm is used to solve the initial SOF controllers and then find the initial coordinate transformation matrix.Finally,a multi-objective multivariable gain-scheduled PD controller design algorithm is developed by combining the coordinate transformation matrix optimization method and NM-HS algorithm,and the lateral motion controller is designed based on this algorithm.In order to allocate the desired yaw moment and longitudinal torque to each wheel,the torque allocation problem is described as a quadratic programming problem with the objective of minimum allocation error and maximum stability margin,and the torque allocation algorithm is obtained.Finally,in the joint MATLAB/Simulink and CarSim simulation platform,the control effect of longitudinal controller and lateral controller is verified under complex operating conditions.The multi-objective PID controller design algorithm and the robust gain-scheduled PD controller design algorithm in this paper can handle the PID/PD controller design problem for other similar systems and are equally effective for the multi-input multi-output controller design problem.The intelligent vehicle trajectory tracking controller designed based on these two algorithms can accurately track the reference trajectory and has the advantages of simple structure and good robustness performance.