基于指数加权衰减记忆无迹卡尔曼滤波的r路面附着系数估计

摘要

A Fading Memory Unscented Kalman Filtering(FM-UKF) algorithm, which was based on exponential weighting for estimating tire-road adhesion coefficient, was proposed by combining the key information of vehicle active safety control. On the basis of traditional Unscented Kalman Filter (UKF), the fading memory filter was used to solve problems that the filtering error was too large or even divergent caused by model error. The co-simulation directed at proposed algorithm was conducted with CarSim and MATLAB/Simulink, vehicle road test was carried out under different road conditions, and the results were compared with those of the traditional UKF algorithm. The results show that the algorithm not only improves the stability of the filtering and improves the estimation accuracy of the algorithm, but also has a certain degree of adaptability.%针对车辆主动安全控制中路面附着系数这一关键信息,提出一种指数加权衰减记忆无迹卡尔曼滤波(FM-UKF)估计算法.该算法在传统无迹卡尔曼滤波(UKF)的基础上,利用衰减记忆滤波来解决由于模型不准确造成的滤波误差过大甚至发散等问题.利用CarSim和MATLAB/Simulink对算法进行了联合仿真和实车道路试验,并与传统UKF算法的估计结果进行对比分析.结果表明,该算法增强了滤波的稳定性、提高了算法的估计精度,且具有一定的自适应性.