交互式多模型卡尔曼滤波的车辆悬架系统状态估计  

Vehicle Suspension System State Estimation Combining with Interacting Multiple Model Kalman Filter

作　　者：顾亮[1] 王振宇 王振峰[1] GU Liang;WANG Zhen-yu;WANG Zhen-feng(School of Mechanical and Vehicle Engineering,Beijing Institute of Technology,Beijing 100081,China)

机构地区：[1]北京理工大学机械与车辆学院,北京100081

出　　处：《东北大学学报（自然科学版）》2018年第11期1642-1647,共6页Journal of Northeastern University(Natural Science)

基　　金：国家自然科学基金资助项目(U1564210);中国博士后科学基金资助项目(2016M600934;BX201600017);国家留学基金委项目(CSC201706030029)

摘　　要：针对车辆悬架状态无法准确估计的问题,设计了自适应交互式多模型卡尔曼滤波(IMMKF)状态观测器.首先,建立了标准路面激励模型与四分之一线性化悬架模型;然后,利用递归最小二乘方法与IMMKF理论,设计了不同工况下广义悬架模型自适应IMMKF状态观测器;最后,分析了在标准C级路面激励工况下簧载质量变化对悬架系统状态估计的影响.仿真与台架试验结果表明,在簧载质量变化工况下,所设计的自适应IMMKF状态观测器与传统卡尔曼滤波状态观测器相比其估计精度至少可以提高20%.In order to estimate accurately the vehicle suspension state,an interactive multiple model adaptive Kalman filter(IMMKF)state observer was proposed.Firstly,a standard road excitation model and a quarter vehicle linear model were established.Secondly,by combining recursive least square algorithm with IMMKF theory,an IMMKF state observer was designed based on the augmented suspension model in various working conditions.Finally,the influence on the state estimation of the suspension system with the change of sprung mass under the ISO level C road input excitation was analyzed.The results of simulation and experiment on a quarter of vehicle test rig showed that compared with the tradition Kalman filter(KF)state observer,the estimation accuracy of the proposed IMMKF state observer could be improved beyond 20%with the change of sprung mass.

关 键 词：状态估计 交互式多模型卡尔曼滤波 递归最小二乘算法 悬架系统 簧载质量 

分 类 号：U463.3[机械工程—车辆工程]