具有不同观测丢失率多传感器随机不确定系统的加权观测融合估计

本文研究了具有丢失观测的多传感器线性离散随机不确定系统的最优线性估计问题,其中不同的传感器具有不同的丢失率.首先将乘性噪声转化为加性噪声,然后基于矩阵满秩分解和加权最小二乘理论,提出了具有较小计算负担的加权观测融合估计算法.分析了加权观测融合估计算法的稳态特性,给出了稳态存在的一个充分条件.所提出的加权观测融合估值器与集中式融合估值器具有相同的精度,即具有全局最优性.仿真研究验证了算法的有效性.

Weighted measurement fusion estimation for stochastic uncertain systems with multiple sensors of different missing measurement rates

This paper is concerned with the optimal linear estimation problem for a multisensor linear discrete-time stochastic uncertain system with missing measurements. Different sensors have different missing measurement rates. Firstly, multiplicative noises are transferred to additive noises. Then, based on full-rank decomposition of a matrix and weighted least-squares theory, the weighted measurement fusion estimation algorithms with small computational burden are developed. The steady-state property of the weighted measurement fusion estimation algorithms is analyzed. A sufficient condition for the existence of the steady state is given. The weighted measurement fusion estimators proposed here have the same accuracy as the centralized fusion estimators, i.e., they have the global optimality. A simulation example shows the effectiveness of the algorithms.