多传感器分布式融合Kalman预报器

Multisensor Distributed Fusion Kalman Predictor

应用现代时间序列分析方法,基于ARMA新息模型,在线性最小方差最优信息融合准则下,对于输入噪声与观测噪声相关且观测噪声相关的多传感器系统,分别提出了按矩阵加权、按标量加权和按对角阵加权的3种分布式融合稳态Kalman预报器。其中提出了基于Lyapunov方程的局部预报估值误差方差阵和协方差阵计算公式。它们被用于计算最优加权,与单传感器情形相比,可提高估值器的精度。一个跟踪系统的仿真例子说明了其有效性,且说明了3种加权融合预报器的精度无显著差别。但标量加权融合预报器可显著减小计算负担,提供一种快速实时信息融合估计算法。

By modern time series analysis method, based on ARMA innovation model, under the linear minimum variance optimal information fusion criterion, the distributed fusion steady-state optimal Kalman predictors weighted by matrices, scalars, and diagonal matrices are presented for multisensor systems with correlated input and observation noises, and with correlated observation noises, respectively. Based on the Lyapunov equations, the formulas of computing local predicting error variances and covariances are given, which are applied to compute optimal weights. Compared to the single sensor case, the accuracy of the fused predictor is improved. A simulation example for tracking systems shows its effectiveness, and shows that the accuracy distinction of the predictors weighted by three ways is not obvious, but the predictor weighted by scalars can obviously reduce the computational burden, and provides a fast real time information fusion estimation algorithm.