带有色观测噪声的广义系统Kalman滤波器  

Kalman Filter with Coloured Observation Noises for Descriptor Systems

作　　者：陶贵丽[1] 刘文强[1] 于海英[1] 

机构地区：[1]黑龙江科技学院计算机与信息工程学院,黑龙江哈尔滨150027

出　　处：《计算机仿真》2010年第3期106-110,205,共6页Computer Simulation

基　　金：国家自然科学基金(60374026);黑龙江科技学院引进人才启动(07-47)

摘　　要：对于带自回归滑动平均(ARMA)有色观测噪声的多传感器为广义离散随机线性系统,应用奇异值分解,将其变换为等价的两个降阶多传感器子系统,提出了广义系统多传感器信息融合状态滤波问题。为了提高精度,采用Kalman滤波方法,在线性最小方差按块对角阵最优加权融合准则下,给出了按矩阵加权解耦的分布式Kalman滤波器,可减少计算负担和改善局部滤波精度。为了计算最优加权,提出了局部滤波误差协方差阵的计算公式。一个Monte Carlo仿真例子说明了方法的有效性。For multisensor which is a kind of discrete - time stochastic linear systems with autoregressive moving average（ARMA） colored observation noises, using the singular value decomposition, the two equivalent reduced order muhisensor subsystems are obtained, and the problem of muhisensor information fusion state filtering is presented. Based on the linear minimum variance optimal fusion rules weighted by diagonal matrices, a decoupled distributed Kalman filter weighted by matrices is presented by using the Kalman filtering method. They can reduce the computational burden, and can improve the local filtering accuracy. In order to compute the optimal weights, the formulas of computing the covariance matrices among local filtering errors are presented. A Monte Carlo simulation example shows its effectiveness.

关 键 词：广义系统 多传感器信息融合 最优加权融合 奇异值分解 

分 类 号：O211.64[理学—概率论与数理统计]