Abstract: This paper focuses on the problem for a class of networked control system with filtering gain uncertainties, sensor- to-filter stochastic packet dropouts and time delays are considered simultaneously, an approach is presented to design the non-fragile dissipative filter with additive gain uncertainties for the estimation task, where the filter parameters are allowed to have norm-bounded uncertainty. By using Lyapunov stability theory and linear matrix inequality methods, a sufficient condition for the existence of the dissipative filter is derived. The problem of non-fragile dissipative filter is turned into the solution of linear matrix inequalities in terms of filtering parameter uncertainties, and then the non-fragile filtering parameters are obtained by solving the LMI. Finally, a numerical simulation is exploited to demonstrate the effectiveness of the proposed design method.

Key words: stochastic time delay, packet dropout, non-fragile dissipative filter, norm-bounded uncertainty

摘要： 针对网络化控制系统的滤波器参数存在摄动的问题，考虑到传感器-滤波器存在随机时延和丢包，设计对系统待估信号进行估计的加性非脆弱耗散滤波器，其中滤波器参数具有范数有界不确定性。首先使用Lyapunov稳定性理论和线性矩阵不等式分析方法，推导出耗散滤波器存在的充分条件，在此基础上考虑到滤波器参数不确定性，从而将非脆弱耗散滤波问题转化为求解线性矩阵不等式的问题，然后通过求解线性矩阵不等式得到非脆弱耗散滤波器的参数表达式，最后数值仿真验证了设计方法的有效性。

关键词: 随机时延, 丢包, 非脆弱耗散滤波器, 范数有界不确性