Approximate Multi-Hypothesis Multi-Bernoulli Multi-Object Filtering Made Multi-Easy
Journal article, 2016

In multiple target tracking (MTT) it becomes necessary to use a multihypothesis approach if the trajectories of two or more targets cross. However, multihypothesis approaches, e.g., the multiple hypothesis tracker (MHT) or the emerging generalized labelled multi-Bernoulli (GLMB) filter, are computationally demanding. In this paper, we propose a simple multi-Bernoulli (MB) filter and a post processing method, which together deliver a multihypothesis tracking estimate at a computational cost that is only slightly larger than the cost of a single-hypothesis tracking filter even for many targets. The proposed MB filter is shown to be similar to the labeled MB filter, itself an approximation of the multihypothesis GLMB filter. In a simulation study with multiple targets and several trajectory crossings, the proposed filter is shown to be capable of correctly estimating the multihypothesis output. The filter is also tasked with presenting to an operator a principled perspective on a scene with many feasible track switches.

RFS

finite sets

PHD

Target tracking

probability hypothesis density

Bernoulli filter

Author

Karl Granström

Chalmers, Signals and Systems, Signalbehandling och medicinsk teknik, Signal Processing

P. Willett

University of Connecticut

Y. Bar-Shalom

University of Connecticut

IEEE Transactions on Signal Processing

1053-587X (ISSN)

Vol. 64 7 1784-1797 7330014

Subject Categories

Signal Processing

DOI

10.1109/tsp.2015.2500884

More information

Created

10/7/2017