基于联邦滤波器结构的分散化H_∞鲁棒滤波算法研究  

Decentralized H_∞Robust Filtering Algorithm Based on Federated Filter Structure

作　　者：李金梁[1] 郝顺义[1] 张宗麟[1] 

机构地区：[1]空军工程大学工程学院,陕西西安710038

出　　处：《信息与控制》2008年第3期284-290,共7页Information and Control

摘　　要：深入研究了基于联邦滤波器结构的分散化H_∞鲁棒滤波算法;通过研究H_∞滤波器的结构和鲁棒机制,提出一套H_∞鲁棒信息滤波公式.在此基础上,建立了一套形式简单、可行性强的联邦H_∞鲁棒滤波算法.与传统联邦卡尔曼滤波算法相比,该算法仅增加一个鲁棒化环节,实际上是一种Krein空间的联邦卡尔曼滤波算法.最后,全面分析了该算法四种结构及其性能,仿真结果表明了其具有有效性和可行性.The decentralized H∞ robust filtering algorithm based on federated filter structure is deeply investigated. By studying the H∞, filter structure and its robustness mechanism, a series of H∞ robust information filtering formulas are presented, on which a set of simple and practicable federated H∞ robust filtering algorithms are also built. Compared with conventional federated Kalman filter, it only adds a robustness step, which equals to a sort of federated Kalman filtering algorithm in Krein space. Finally, four structures of the algorithm and the corresponding performance are analyzed. The simulation results proves the effectiveness and feasibility of the proposed algorithm.

关 键 词：分散化滤波 联邦滤波器 KREIN空间 H∞滤波 信息滤波 

分 类 号：TP13[自动化与计算机技术—控制理论与控制工程]