整车质量与道路坡度同步估计算法对比

Comparison of synchronous estimation algorithms for vehicle mass and road slope

针对现代车辆控制系统难以实时测得整车质量和道路坡度这一问题,文章基于车辆纵向动力学模型给出无迹卡尔曼滤波(unscented Kalman filter,UKF)算法的流程图,利用CarSim与MALTAB/Simulink联合仿真比较双遗忘因子递归最小二乘法(recursive least square with multiple forgetting factors,RLS-MFF)、扩展卡尔曼滤波(extended Kalman filter,EKF)和UKF 3种算法对整车质量和道路坡度估计的效果,并分别从计算精度和实时性2个方面对3种算法进行比较分析。分析结果验证了UKF算法在整车质量估计和道路坡度估计中表现良好。

In view of the fact that the vehicle mass and road slope are difficult to be measured in real time for the modern vehicle control system,the basic steps of unscented Kalman filter(UKF)algorithm were given based on the vehicle longitudinal dynamic model.The co-simulation with CarSim and MALTAB/Simulink was carried out to compare the estimation effects of recursive least square with multiple forgetting factors(RLS-MFF),extended Kalman filter(EKF)and UKF algorithms on vehicle mass and road slope estimation.The three algorithms were compared and analyzed in terms of estimation accuracy and real-time performance,and the results show that the UKF algorithm performs well in vehicle mass and road slope estimation.