Estimation Of Vehicle States And Tire-road Friction Coefficient For Vehicle Stability Control

In recent years, more and more active safety systems were used for automotive control systems to improve vehicle handling stability and security. More sensors are required to obtain the vehicle status and parameters, which increase the cost of vehicle development and make vehicle control system more complex,. Therefore, this article will research through model-based estimation method, without changing the configuration of the sensor in the existing production car, to obtain the vehicle states and parameters. Then the integrated chassis system stability controller is designed based on the estimated states and parameters.First, based on the two degrees of freedom vehicle model and linear tire model, the Kalman filter algorithm is utilized to estimate the lateral velocity, yaw rate and the front and rear tire slip angle of vehicle with the help of the measurement signals of ABS wheel speed sensors. The simulation results show, the Kalman filter algorithm can accurately estimate the lateral velocity and yaw rate, however, the front and rear tire slip angle estimation will become unsatisfactory with increase of tire slip angle.Secondly, the tire slip angle and tire-road friction coefficient play an important role to reflect the mechanism of interaction between tire and road, so in order to obtain more accurate estimations than the Kalman filter algorithm above, two methods of estimating the friction coefficient are designed : a direct algebraic method and a nonlinear observer. Because the aligning torque is not at the peak value all the time, the friction coefficient estimated by direct algebraic method receives a good accuracy at large tire slip angle. However, when tire is at small slip angle, the accuracy of this method becomes lower. In order to acquire better accuracy of vehicle parameters, the nonlinear observer based on vehicle lateral dynamics and brush tire model can accurately estimate the tire slip angles and road friction coefficient. The validity of the observer is verified via simulation test.Finally, according to the vehicle driving states and parameters estimated, the hybrid coordinated main-servo-actuators are utilized to realize the vehicle longitudinal, lateral and vertical integrated control. In the main loop, because of modelling uncertainty, disturbance and the parameters of state coupling, the sliding mode control algorithm is used to trace the reference vehicle model for calculating the desirable forces, which are required of the vehicle stability. And in the servo loop, in consideration of the tire nonlinear coupling, the quadratic programming algorithm is applied to optimally allocate the target generalized forces to sub-systems respectively. In the last part of this paper, the double lane change simulation test is carried out to verify the effectiveness of the controller designed in the MATLAB/Simulink environment. The simulation results demonstrate that the integrated controller has enhanced the potential of sub-systems, followed the desirable motion effectively, suppressed the side slip and improved the vehicle active safety.