Please use this identifier to cite or link to this item:
Scopus Web of Science® Altmetric
Type: Journal article
Title: An efficient, variational approximation of the best fitting multi-Bernoulli filter
Author: Williams, J.
Citation: IEEE Transactions on Signal Processing, 2015; 63(1):258-273
Publisher: IEEE
Issue Date: 2015
ISSN: 1053-587X
Statement of
Jason L. Williams
Abstract: The joint probabilistic data association (JPDA) filter is a popular tracking methodology for problems involving well-spaced targets, but it is rarely applied in problems with closely spaced targets due to its complexity in these cases, and due to the well-known phenomenon of coalescence. This paper addresses these difficulties using random finite sets (RFSs) and variational inference, deriving a highly tractable, approximate method for obtaining the multi-Bernoulli distribution that minimizes the set Kullback–Leibler (KL) divergence from the true posterior, working within the RFS framework to incorporate uncertainty in target existence. The derivation is interpreted as an application of expectation-maximization (EM), where the missing data is the correspondence of Bernoulli components (i.e., tracks) under each data association hypothesis. The missing data is shown to play an identical role to the selection of an ordered distribution in the same ordered family in the set JPDA algorithm. Subsequently, a special case of the proposed method is utilized to provide an efficient approximation of the minimum mean optimal subpattern assignment estimator. The performance of the proposed methods is demonstrated in challenging scenarios in which up to twenty targets come into close proximity.
Keywords: Bayesian estimation; coalescence; expectation maximization; Kullback-Leibler divergence; optimum sub-pattern assignment; random finite sets; target tracking; variational inference
Rights: © 2014 British Crown Copyright
RMID: 0030021158
DOI: 10.1109/TSP.2014.2370946
Appears in Collections:Electrical and Electronic Engineering publications

Files in This Item:
There are no files associated with this item.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.