Federated Filter for Distributed Navigation and Tracking Applications

Neal A. Carlson

Abstract:	The federated filter comprises an extension of Kalman filtering to distributed estimation problems characterized by a collection of local estimators whose solutions are combined by a master or global estimator. The federated filter employs a rigorous information-sharing methodol- ogy to gather, propagate, and combine information from local sensors in a near globally optimal manner. This methodology offers several implementation modes in two basic classes. In the first class, local filters collectively retain the cumulative system information, and the master filter performs periodic fusion (combining) operations. In the second class, the master filter retains the cumulative system information, and local filters perform short-term information-gathering operations. The paper reviews the information-sharing principles and implementation modes of the federated filter. It then describes several federated filter examples. The first two examples involve a distrib- uted navigation system with a GPS receiver containing its own local filter, an inertial navigation unit, and other navigation sensors. The third example is a multiplatform target track fusion filter, characterized by local trackers processing own-ship sensor measurements on each platform, and a global tracker that combines local tracks across platforms. The fourth example is an integrated GPS signal-tracking/navigation filter, with local filters repre-senting signal tracking loops, and a master GPS/inertial navigation filter. The last example involves a federated filter/smoother that combines the outputs of two post- processors designed for separate reference instrumentation systems, to reconstruct flight test trajectories in a fully optimal manner. Illustrative simulation results are presented for the first three application examples.
Published in:	Proceedings of the 58th Annual Meeting of The Institute of Navigation and CIGTF 21st Guidance Test Symposium (2002) June 24 - 26, 2002 Hyatt Regency Hotel Albuquerque, NM
Pages:	340 - 353
Cite this article:	Carlson, Neal A., "Federated Filter for Distributed Navigation and Tracking Applications," Proceedings of the 58th Annual Meeting of The Institute of Navigation and CIGTF 21st Guidance Test Symposium (2002), Albuquerque, NM, June 2002, pp. 340-353.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In