Information Fusion Algorithm for Vehicle State Estimation Based on Extended Kalman Filtering

Abstract

Some state variables of a vehicle in running are not easy to measure accurately or cheaply, however these variables are of great significance to chassis control. A nonlinear 3 degree-of-freedom vehicle model including yaw motion, longitudinal motion and side motion is set up, and an information fusion algorithm based on extended Kalman filtering (EKF) theory is established, which gives out a fusion result of vehicle state variables at minimum square error. Fusing a few state variables of vehicles (steering wheel angle, longitudinal acceleration and lateral acceleration), the needed variables of the vehicle (yaw rate and sideslip angle) are procured. Off line simulation is carried out in Matlab/simulink environment by using real vehicle field test data. The algorithm is accurate in estimating longitudinal rate, side slip angle, and especially yaw rate, and shows good performance even in the nonlinear zone of the vehicle. The algorithm is simple and stable, and needs less fusion input, so it is possible to apply it to actual vehicle control.