Graph Marginalization for Rapid Assignment in Wide-Area Surveillance

Mark Ebden; Stephen Roberts

Ad Hoc Networks. First International Conference, ADHOCNETS 2009, Niagara Falls, Ontario, Canada, September 22-25, 2009. Revised Selected Papers

Research Article

Graph Marginalization for Rapid Assignment in Wide-Area Surveillance

Download

402 downloads

Cite: BibTeX Plain Text

@INPROCEEDINGS{10.1007/978-3-642-11723-7_47,
    author={Mark Ebden and Stephen Roberts},
    title={Graph Marginalization for Rapid Assignment in Wide-Area Surveillance},
    proceedings={Ad Hoc Networks. First International Conference, ADHOCNETS 2009, Niagara Falls, Ontario, Canada, September 22-25, 2009. Revised Selected Papers},
    proceedings_a={ADHOCNETS},
    year={2012},
    month={7},
    keywords={sensor networks coalition formation agents max-sum algorithm belief propagation},
    doi={10.1007/978-3-642-11723-7_47}
}

Mark Ebden
Stephen Roberts
Year: 2012
Graph Marginalization for Rapid Assignment in Wide-Area Surveillance
ADHOCNETS
Springer
DOI: 10.1007/978-3-642-11723-7_47

Mark Ebden¹^,*, Stephen Roberts¹^,*

1: University of Oxford

*Contact email: mebden@robots.ox.ac.uk, sjrob@robots.ox.ac.uk

Abstract

Decentralizing optimization problems across a network can reduce the time required to achieve a solution. We consider a wide-area surveillance (WAS) sensor network observing an environment by varying the state of each sensor so as to assign it to one or more moving objects. The aim is to maximize an arbitrary utility function related to object tracking or object identification, using graph marginalization in the form of belief propagation. The algorithm performs well in an example application with six heterogeneous sensors. In larger network simulations, the time savings owing to decentralization quickly exceed 90%, with no reduction in optimality.

Keywords: sensor networks coalition formation agents max-sum algorithm belief propagation

Published: 2012-07-05

: http://dx.doi.org/10.1007/978-3-642-11723-7_47