随机系数矩阵卡尔曼滤波  

Random parameter matrices Kalman filtering

作　　者：罗丹丹[1] 朱允民[1] 

机构地区：[1]四川大学数学学院,成都610064

出　　处：《四川大学学报（自然科学版）》2008年第6期1309-1312,共4页Journal of Sichuan University(Natural Science Edition)

基　　金："863"项目(2006AA12A104)

摘　　要：作者考虑了状态转移矩阵和量测矩阵是随机阵的线性离散时间动态系统的状态的线性最小方差递推估计问题,即随机系数矩阵卡尔曼滤波,说明了该系统可化为过程噪声和量测噪声均依赖于状态,而转移矩阵和量测矩阵是非随机阵的线性动态系统,从而证明了新系统的状态的最小方差估计问题仍有卡尔曼滤波形式.The Linear Minimum Variance recursive state estimation is studied in the linear discrete time dynamic system with random state transition and observation matrices, i. e., random parameter matrices Kalman filtering. It is shown that such system can be reduced to a linear dynamic system with deterministic parameter matrices and state-depending process and measurement noises, and the Linear Minimum Variance recursive state estimation of the new system is still in the form of Kalman filtering.

关 键 词：随机系数矩阵 随机卡尔曼滤波 线性最小方差递推估计 

分 类 号：O211.9[理学—概率论与数理统计]