| The handling stability and roll stability of a vehicle are related to the driving safety of the vehicle.Improving the handling stability and anti-rollover capability has become an important research content of active safety control.In recent years,assisted driving with lateral and longitudinal dynamics such as active front steering(AFS)and differential braking system(DBS)have become important ways of research on vehicle handling stability and anti-rollover control.DBS and AFS adjust the vehicle yaw moment through the tire longitudinal force and tire lateral force,respectively,to achieve the above stability control.However,they have different yaw moment adjustment capability and sensitivity.In emergency conditions,it is necessary to select the actuator for vehicle control based on the actuator’s yaw moment adjustment capability(controllable region)and sensitivity to quickly achieve vehicle control goals.Firstly,an 8-degree of freedom nonlinear dynamic model and a Simulink simulation model including lateral motion,longitudinal motion,yaw motion,roll motion and four wheel rotation motions is built.The longitudinal and lateral motion correlation is reflected through a "vertical-lateral-longitudinal" coupled tire model.Compared with the Car Sim vehicle model,the accuracy of the vehicle model is verified.Further,considering that in practical applications,dynamic modeling will face the problem of unknown tire models and road adhesion coefficients.An off-line identification method of tire model and road adhesion coefficient based on BP neural networks and vehicle dynamics response is proposed.This method utilizes the advantage that a BP neural network that does not require a physical model can approximate arbitrary nonlinear mapping relationships.The method achieves offline tire model identification and online identification of road adhesion coefficient based on Inertial Measurement Unit(IMU),wheel speed sensors,steering wheel angle torque sensors,and other on-board sensors,as well as vehicle dynamics responses.Simulation experiments show that the identification method has high accuracy,practicality,and generalization.Then,considering the tire coupling effect,the controllable regions of yaw moment for AFS,DBS with single wheel intervened,and DBS with dual wheel intervened are studied,respectively.A method is proposed to calculate the controllable region of AFS and DBS yaw moment based on the current state(current front wheel steering angle,current slip ratio of the front wheels,current wheel side slip angle of the front wheels,longitudinal acceleration,lateral acceleration,and road adhesion coefficient)of the vehicle,and the influence of different vehicle states on the controllable region of the actuators is analyzed.Based on this,a method is proposed to determine the AFS and DBS sensitivity by the current state of the vehicle and the required additional yaw moment.Finally,according to the control region and sensitivity analysis of different actuators,a coordinated control strategy is developed.Simulation experiments of vehicle yaw stability control and anti-rollover control were carried out by MATLAB,which verified the practicability of controllable region analysis and sensitivity analysis and the effectiveness of coordinated control strategy,indicating that the research in this paper has important significance for vehicle yaw stability control and anti rollover control.In this paper,the research on nonlinear tire model identification based on on-board sensor and vehicle dynamics response has a guiding role in accurate modeling of nonlinear dynamic systems;The research on the yaw moment controllable domain is expected to provide a new decision-making strategy for the execution of system tasks in the chassis integrated control systems,and is of great significance for vehicle yaw stability control and anti-roll control. |
PDF Full Text Request