Distributed Computing Filters : Multisensor Marine Applications

Abstract

The distributed multisensor estimation problem is considered in this paper, and both the centralized and the de- centralized approach are discussed. The Kalman filter based approaches are compared with the Lainiotis filter based ap- proach, in terms of their computational efficiency and trans- mission requirements. The comparative evaluation shows that using the Lainiotis filter, both in its sequential and distributed form, orders of magnitude decrease in computational require- ments and processing time can be achieved.