Study On State Estimation Algorithm For Four-wheel-independent-drive Electric Vehicles

As an important part of green transportation in the future,electric vehicles play an important role in China's economic development and urban construction.The four wheel independent drive electric vehicle takes hub motor as the driving system.It has compact structure,flexible torque distribution and more accurate dynamic control.It is the future development trend of electric vehicle.In order to realize the dynamic control in the course of vehicle driving,the key state parameters,such as the longitudinal speed of the electric vehicle and the side angle of the center of mass,must be estimated.Due to the complexity of the control process and the limitation of the sensor test level,some key parameters can not be obtained directly and accurately.Vehicle characteristic design state parameter estimator is used to estimate and solve real-time state information acquisition in dynamic control.In this paper,based on the wheel motor driven electric vehicle,the state estimation and parameter identification method under the driving condition are studied,which provides a theoretical method for the safety and stability control of the electric vehicle.The main research work of this paper is as follows:(1)In view of the characteristics of electric vehicle driven by hub motor,a Carsim Matlab/Simulink joint simulation model is established.According to the degree of freedom required by the Institute,three degree of freedom and seven degree of freedom nonlinear vehicle dynamics models are established respectively.The model is validated by comparing the established dynamic model response curve.(2)The basic principle of Kalman filtering algorithm is expounded,and the state parameter estimation is carried out based on the extended Kalman filter and the Untraced Kalman filter algorithm for the automotive nonlinear dynamic system.By comparing the estimated value of the key parameters such as the longitudinal speed,the yaw rate,the centroid side angle and the Carsim software,the validity of the extended Kalman filter and the Untraced Kalman filtering algorithm is verified.(3)Considering the unknown noise statistical characteristics and the change of vehicle parameters,the state estimation algorithm based on adaptive filtering and double extended Kalman filter is studied.The results show that the improved adaptive Untraced Kalman filter and double spread Kalman filter algorithm have higher accuracy and robustness,and are more suitable for the change of complex environment in the actual driving process.(4)Using the parameter information obtained by the state estimation,the vehicle road adhesion coefficient is estimated by the adaptive Untraced Kalman filtering algorithm combined with the magic tire formula,and the simulation tests are carried out under different working conditions and different adhesion coefficients respectively.The results show that the design self adaption to the Untraced filter Road adhesion coefficient estimator The estimation results are accurate and the speed of convergence is fast.