测量数据丢失的随机不确定系统鲁棒滤波递推算法

针对一类具有测量数据丢失的不确定离散随机系统,研究了鲁棒状态估计问题,基于间断观测滤波算法和规则最小二乘优化理论,给出一种Kalman形式的递推滤波算法.对于测量数据丢失的问题,采用已知概率的Bernoulli随机序列,使得对于所有可能的测量数据丢失和所能容许的不确定性,间断观测鲁棒状态估计递推算法是稳定的.最后,通过数值仿真和对比结果验证了所提出算法的可行性.

Robust Filter Recursive Algorithm for Stochastic Uncertain System with Missing Measurement

This paper studies robust state estimation problem in uncertain discrete stochastic system with measurement
missing and proposes a Kalman type recursive algorithm based on intermittent observation filtering algorithm and least square optimizing theory. The missing measurement model adopts Bernoulli random series based on given probability, which enables robust state estimation with intermittent observation stable for all possible missing measurement situation and admissible uncertainty. Finally, simulation and comparison results show the feasibility of the algorithm.