Title :
On the optimality of decentralized filtering systems
Author :
Wen-Hua Chen ; Xin Yuan
Author_Institution :
Dept. of Autom. Control, Nanjing Aeronaut. Inst., China
Abstract :
For the multisensor systems, the decentralized information structure can improve the fault-tolerant ability and increase the processing speed. This paper presents the necessary and sufficient condition for the optimality of decentralized linear dynamic filters, i.e., the decentralized filter is equal to the centralized filters. It is pointed out that this condition is satisfied only in rare case. However, by analyzing federated filter presented by Carlson (1990), we show that it is possible to obtain optimal decentralized filters with the modified information structure.<>
Keywords :
distributed processing; filtering theory; sensor fusion; decentralized information structure; decentralized linear dynamic filters; fault-tolerant ability; federated filter; multisensor systems; necessary and sufficient condition; optimal decentralized filtering systems; Aerodynamics; Aircraft navigation; Equations; Global Positioning System; Information filtering; Information filters; Nonlinear filters; Sufficient conditions; Vectors; Vehicle dynamics;
Conference_Titel :
TENCON '93. Proceedings. Computer, Communication, Control and Power Engineering.1993 IEEE Region 10 Conference on
Conference_Location :
Beijing, China
Print_ISBN :
0-7803-1233-3
DOI :
10.1109/TENCON.1993.320455
