多传感器时滞系统信息融合最优Kalman滤波器

Multi-sensor information fusion optimal Kalman filter for time-delay systems

基于线性最小方差最优加权融合估计算法,对多传感器的离散线性状态时滞随机系统.给出了一种非增广分布式加权融合最优Kalman柚滤波器.推导了状态时滞系统任两个传感器子系统之间的滤波误差互协方差阵的计算公式.它与状态增广加权融合滤波器具有相同的精度.与每个传感器的局部滤波器相比,分布式融合滤波器具有更高的精度.与状态和观测增广最优滤波器相比,具有较小的精度,但避免了增广所带来的高维计算和大的空间存储。可减小计算负担.仿真例子验证了其有效性.

Based on the optimal weighted fusion estimation algorithm with minimum variance, a non-augmentation distributed weighted fusion optimal Kalman filter is given for discrete linear state time-delay stochastic systems with multiple sensors. The cross-covariance matrix of filtering errors between any two-sensor subsystems is derived for state time-delay systems. It has the same accuracy with weighted fusion filter with state augmentation. Compared with local filter based on each sensor, the distributed fusion filter has higher accuracy. Compared with the optimal filter with state and measurement augmentation, it has lower accuracy, but avoids the high-dimension computation and the large memory by augmentation, and has the reduced computational burden. A simulation example also shows its effectiveness.