一种新型非线性滤波理论及其应用  

A new nonlinear filter theory and its application

作　　者：吴江飞[1] 雷辉[2] 王世忠[3] 

机构地区：[1]信息工程大学测绘学院 [2]中国科学院国家授时中心 [3]81804部队

出　　处：《测绘工程》2008年第6期16-19,共4页Engineering of Surveying and Mapping

基　　金：中国科学院知识创新工程重要方向资助项目(KJCX2-SW-T1);国家863计划资助项目(2002AA731171)

摘　　要：研究一种新型的非线性滤波理论,即Unscented卡尔曼滤波(UKF),同时为了获得更高的计算效率和确保协方差阵的非负定性,研究了平方根UKF。将UKF和平方根UKF应用到星载GPS卫星定轨中,实际算例表明UKF和平方根UKF的性能要优于常用的推广卡尔曼滤波的性能。A new nonlinear filter theory （the Unscented Kalman filter, UKF） is studied in this paper. At the same time, in order to obtain higher calculation efficiency and guarantee positive semi-definiteness of the state covariances, the square root Unscented Kalman filter is studied, and its algorithm is presented. Lastly, UKF and SR-UKF are used to determine the orbit of satellites equipped with GPS. And a practical example shows that the performance of the UKF and SR-UKF are superior to that of the EKF.

关 键 词：非线性滤波 UNSCENTED卡尔曼滤波 平方根Unscented卡尔曼滤波 定轨 

分 类 号：P228[天文地球—大地测量学与测量工程]