Research On Control Problems Of Active Rear Wheel Steering With Domain Control Architecture

With the continuous rise of intelligent driving technology,the number of on-board controllers is increasing.In order to integrate the redundant bus and controller,the domain control architecture came into being.Active rear wheel steering technology can effectively improve the handling stability and active safety of vehicles.This paper aims to study the implementation and control strategy of active rear wheel steering under domain control architecture.Firstly,aiming at the realization of ARS function under domain control architecture,the ARS domain control architecture for engineering application is introduced.Then,according to the requirements of domain control architecture for the following effect of rear wheel steering actuator,the motor three closed-loop control theory is proposed,and the position following experiment of rear wheel steering actuator is carried out on the bench.In addition,aiming at the influence of the communication cycle between the domain controller and the rear wheel steering controller on the actuator position following control effect,the actuator position following experiment is carried out and analyzed with different communication cycles.Compared with the reference cycle provided by the vehicle manufacturer,the appropriate communication cycle is selected.In practical application,the ARS software protection and communication fault response strategy is designed and updated.Secondly,in order to analyze the ARS control strategy,the linear two degree of freedom dynamic models of front wheel steering vehicle and active rear wheel steering vehicle are established.Then,three ARS control strategies based on classical control theory are theoretically analyzed,and then based on modern control theory,a variable weight coefficient LQR controller is designed.Then,taking the sideslip angle of mass center and yaw rate as the two control objectives,the simulation experiments of the above strategies are carried out.Through the analysis of the experimental results,each strategy can compress the vehicle steady-state centroid sideslip angle,so as to reduce the phase difference between yaw rate and lateral acceleration.Through the angle step simulation,it is found that when the vehicle turns at high speed,the same direction rotation of the rear wheel relative to the front wheel improves the yaw stability of the vehicle.Through the angle step simulation,it is found that the rotation of the rear wheel in the same direction suppresses the yaw and brings the side effect of reducing the response speed of yaw angular velocity.At this time,the ARS strategy outside the proportional feedforward compensates the side effect and compresses the dynamic centroid sideslip angle while improving the yaw stability by reasonably designing the first reversal and then co rotation of the rear wheels.By analyzing the phase trajectory of centroid sideslip angle,it can be proved that ARS has a good improvement on vehicle stability.Among them,the control effect of LQR strategy is more significant,and it is insensitive to parameter accuracy and has strong robustness.However,as the ARS strategy is single input and single output,it can only make one control target tend to be ideal,resulting in the reduction of steering sensitivity.Thirdly,as an important evaluation index and state information in ARS strategy,it is very important to extract the centroid sideslip angle.Due to the high price of equipment for measuring centroid sideslip angle,a nonlinear full order state observer is designed based on Dugoff and Uni Tire tire models to estimate centroid sideslip angle.Then a simulation experiment is carried out to verify the accuracy of the observer,which provides a reference for real vehicle experiment and application.Finally,in order to verify the above strategy,the target vehicle provided by the enterprise is modified,and the real vehicle test environment is established.In the experiment,it is found that the closed-loop feedback strategy will produce oscillation,which is analyzed.Then,two ARS strategies without closed-loop feedback are used for real vehicle experiments,and the experimental results are evaluated.The results prove the accuracy of the theoretical analysis and verify the feasibility of its application in engineering practice.Finally,through the data recording of the real vehicle experiment,the data is reinjected in Simulink,and the centroid sideslip angle estimation experiment of the real vehicle data is carried out to verify the accuracy of the nonlinear full order state observer.This paper analyzes and studies the implementation and requirements of domain control architecture,active rear wheel steering control strategy,practical application of centroid sideslip angle estimation and the problems found in real vehicle experiments,which provides a reference for the engineering application of ARS technology.