针对一类具有测量数据丢失的不确定离散随机系统,讨论鲁棒H2/H∞滤波器的问题.文章的目的是应用一种新的方法设计滤波器,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,滤波误差系统都是均方意义下指数稳定,且同时具有H2/H∞性能特性.利用线性矩阵不等式(LM I)的方法给出满足给定性能的滤波器存在的充分条件.并通过数值仿真和对比结果验证了所提出算法的可行性.
The robust H2/H∞ filtering problem is studied for a class of uncertain discrete-time stochastic system with missing measurement. A filter, for all possible missing measurement and admissible uncertainty, is designed to make sure that the estimation error system is exponentially mean-square quadratically stable, and guarantee a prescribed H2 and H∞ performance simultaneously. The sufficient condition for existence of the filter is given by using the technique of the linear matrix inequalities(LMI) , a new approach is established to solve the robust H2/H∞ filtering problem. At last, a numerical example and the result of contrast have been presented to demonstrate the feasibility of the new proposed design approach.