具有随机观测滞后和丢失的网络系统鲁棒H1滤波  

Robust H-infinity filtering for network-based systems with delayed and missing measurements

作　　者：李秀英[1] 王金玉[1] 孙书利[1] 

机构地区：[1]黑龙江大学电子工程学院,黑龙江哈尔滨150080

出　　处：《控制理论与应用》2013年第11期1401-1407,共7页Control Theory & Applications

基　　金：国家自然科学基金资助项目(NSFC–60874062,61174139);教育部新世纪人才资助项目(NCET–10–0133);黑龙江大学高层次人才资助项目(Hdtd2010–03);黑龙江省研究生创新科研资助项目(YJSCX2012–271HLJ);黑龙江省重点实验室基金资助项目(DZZD20100023)

摘　　要：研究一类同时具有随机观测滞后和丢失的网络控制系统鲁棒H1滤波问题.网络建模采用新近同时描述随机一步时滞和数据包丢失的模型,基于Lyapunov理论和线性矩阵不等式技术设计系统的满阶滤波器,减小了状态增广带来的计算负担,同时引入松弛变量使Lyapunov矩阵和系统矩阵分离,降低结果的保守性,得到了使滤波误差系统在均方意义下鲁棒指数稳定并具有给定H1性能的一个LMI形式的充分条件.仿真结果验证了该方法有效性.The robust H-infinity filtering problem is studied for a class of networked control systems （NCSs） with random delayed and missing measurements. The recent model of NCSs is adopted to describe the random one-step trans- mission delay and possible packet dropouts, simultaneously. A full-order filter is designed based on Lyapunov theory and the linear matrix inequality （LMI） technique which can reduce the computational burden involved in the state augmenta- tion method. Meanwhile, the slack matrix is introduced to separate the system matrix and the Lyapunov matrix, and a less conservative result is obtained. An LMI-based sufficient condition is established such that the filtering error system is robustly exponentially stable in the sense of mean square and achieves the prescribed H-infinity disturbance attenuation level. Simulation results are given to illustrate the effectiveness of the proposed method.

关 键 词：H∞滤波 网络控制系统 数据包丢失 观测滞后 线性矩阵不等式 

分 类 号：TP13[自动化与计算机技术—控制理论与控制工程]