Research On Vehicle State Parameters Estimation Based Model

The research of vehicle state parameters estimation as the key technologies of vehicle active safety control system has drawn growing attention,espetially for the esitmation of vehicle longitudinal speed,sideslip angle and yaw rate.In this paper,the research concentrates on the vehicle state parameters estimation based on dynamic model and kinematic model to ensure the real-time performance and accuracy of the estimation.In the research of vehicle state parameters estimation based on dynamic model,the nonlinear observation technique which can bring into play the nonlinear simulation characteristcs of dynamic model is adopted to estimate vehicle longitudinal speed and sideslip angle.The parameters of magic formula tire model based on the experimental data are identified,then the vehicle dynamic model of 3-DOF including the movement of longitudinal,lateral and yaw is built and the united simulation analysis with vehicle dynamic software Carsim is conducted.Based on the vehicle dynamic model a nonlinear state observer is designed and it includes a sub nonlinear observer and an amending module.The sub nonlinear observer has fixed gain parameters to ensure the real-time performance of the estimation,the amending module adjusts the gain parameters according to the simulation data under typical conditions to ensure the accuracy of the estimation.In the research of vehicle state parameters estimation based on kinematic model,the kalman filter technique which can bring into play the good real-time characteristics of kinematic model is adopted to estimate vehicle longitudinal speed.The signals of wheel speed sensor and gyroscope are filtered to reduce the measurement noise.Then the relationships between wheel speeds and longitudinal speed are inferred and the vehicle kinematic model for estimation considering the influence of longitudinal slip is established for the estimation.The kalman filter bank is built based on the knematic model and kalman filter techque,then the final longitudinal speed estimation value is got by merging the longitudinal speed from the longitudinal acceleration integration with the kalman filter eatimation values through weighted mean method.According to the united simulation results under typical operating conditions,the proposed two methodes for vehicle state parameters estimation have good accuracy,and can satisfy the requirment of the active safety control system.