对于带不同局部动态模型(多模型)的多传感器线性定常随机控制系统,应用现代时间序列分析方法,在按标量加权最优融合准则下,提出了最优信息融合稳态白噪声反卷积估值器。可统一处理白噪声反卷积融合滤波、平滑和预报问题。它的精度高于每个局部估值器的精度。为了计算最优加权,提出了局部估计误差互协方差计算公式。一个Bernoulli-Gussian白噪声反卷积融合器的仿真例子证明其有效性。
For the multisensor linear discrete time-invariant stochastic control systems with the different local models (multi-model), using modem time series analysis method, under the optimal fusion criterion weighted by scalars, a class of the optimal information fusion white noise deconvolution estimators is presented. It can handle the white noise deconvolution fused filtering, smoothing and prediction problems in a unified framework. Its accuracy is higher than that of each local estimator. In order to compute the optimal weights, the formula of computing cross-covariances among local estimation errors is presented. A simulation example for the Bemoulli-Gussian white noise deconvalution fuser shows its effectiveness.