一类不确定系统分布式鲁棒融合滤波器设计*

作者：孙 航,文成林 日期：2008-04-23/span> 浏览：3969 查看PDF文档

一类不确定系统分布式鲁棒融合滤波器设计*
孙 航,文成林
（杭州电子科技大学 信息与控制研究所,浙江 杭州 310018）

摘 要：结合鲁棒滤波理论与数据融合技术，基于参数依赖Lyapunov函数,研究了一类多传感器不确定连续系统的鲁棒融合滤波器设计问题。在集中式融合的基础上，探讨了分布式加权融合滤波器的设计方法，通过仿真实验比较了鲁棒融合滤波器的性能。仿真结果表明，利用该分布式加权融合算法，不仅可解决系统模型中存在参数不确定性时的滤波问题，且鲁棒性好和计算量低，可以满足不同精度鲁棒融合滤波器的设计需要，具有较大的灵活性。
关键词：多传感器连续不确定系统；分布式；集中式；H2滤波

Design of distributed robust fusion filters for a class of discrete uncertain systems
SUN Hang, WEN Chenglin
(Institute of Information and Control, Hangzhou Dianzi University, Hangzhou 310018, China)

Abstract: In terms of parameterdependent Lyapunov function, the design problems of robust fusion filters for a class of multisensor uncertain systems were dealt with, by means of combining robust filter theory and data fusion technology. On the basis of the design method of the centralized fusion, distributed robust fusion filter was discussed, moreover, the performance of the new filter was illustrated by the simulation. The results show that by using the distributed weight fusion method, it does not only have a better robust performance and lower computational complexity for resolving the filter problems when parameter uncertainties exist in the system model, but also have more flexibility to satisfy the design requirements of the robust fusion filters for multisensor systems with different precisions.
Key words: multisensor discrete uncertain systems; distributed; centralized; H2 filter
参考文献(Reference):
［1］LIN Y, XUE A, QIAN J. Discretetime Fusion Filter in Data Fusion System［C］. Proceedings of 2003 IEEE Conference on Control Applications, Istanbul: Bogazici University,2003:1420-1424.
［2］金学波,孙优贤.基于LMI方法的参数不确定多传感器系统的集中式融合估计［J］.电路与系统学报,2006,11(4):70-77.
［3］BROWN C, DURRANTWHYTE H, LEONARD J. Centralized and Decentralized Kalman Filter Techniques for Tracking, Navigation and Control［C］. In Proceedings of the Defense Advanced Research Projects Agency IU Workshop, Palo Alto, CA: 1989:651-675.
［4］金学波,孙优贤.集中式与分布式鲁棒状态融合估计［J］.控制理论与应用,2005,21(6):865-869.
［5］高会军.基于参数依赖Lyapunov函数的不确定动态系统的分析与综合［D］.哈尔滨:哈尔滨工业大学计算机科学与技术学院,2005:59-79.
［6］BOYD S, GHAOUI L, FERON E, et al. Linear Matrix Inequalities in System and Control Theory［M］. Philadelphia: SIAM,1994.
［7］TUAN H, APKARIAN P, NGUYEN T. Robust and reducedorder filtering,new LMIbased characterizations and methods［J］.IEEE Transactions on Signal Processing,2001,49(12):2975-2984.
［8］GEROMEL J. Optimal linear filtering under parameter uncertainty［J］.IEEE Transactions on Signal Processing，1999,47(1):168-175.