Research On The Algorithm Of The Road Friction Coefficient Estimation Based On The Cubature Kalman Filter

Road friction coefficient is one of the most important parameters in vehicle safety control and electronic stability program. Nowadays The road friction coefficient identification by the sensor can not meet the complicated situations during the driving. This method is still in the experimental stage and can not be used in the actual vehicle. The algorithm of road friction coefficient estimation by using Cubature Kalman Filter is designed based on the vehicle dynamic model. The main research work includes:1. Studied the Kalman Filter.The advantages of Cubature Kalman Filter is analyzed relatived to other Kalman Filters theoretically. Studied the vehicle dynamics. The 3 DOF and 4 DOF vehicle dynamic model is chose according to the demands of algorithm.2. Designed the vehicle state estimation based on the Cubature Kalman Filter.3 DOF vehicle dynamic model is established. The state is estimated by using Cubature Kalman Filter based on the model which is side slip angle of CoG, yaw rate, longitudinal velocity that are very important to the road friction coefficient estimation. And the algorithm is validated through simulation.3. Designed the road friction coefficient estimation based on the Cubature Kalman Filter.4 DOF vehicle dynamic model is established. The vehicle state is used as observation. The road friction coefficients of four wheels are the state parameters. The algorithm of road friction coefficient based on the Cubature Kalman Filter is designed to estimate the road friction coefficient.4. Validated the algorithm on the co-simulation platform which units Carsim and Matlab/Simulink. Three situations is used to validated the algorithm on this platform which is high friction coefficient surface road, low friction coefficient surface and split friction coefficient surface road. And the estimation based on the Extended Kalman Filter and Unsented Kalman Filter is brought in. The results of the three estimation is compared with the data provided by Carsim.The simuliation result shows that the algorithm of road friction coefficient estimation based on the Cubature Kalman Filter can estimate the road friction coefficient estimation well. And it has advantages over the estimation based on the Extended Kalman Filter and Unsented Kalman Filter on dynamic and steady performance.