Title: Fusion and decision making in the presence of uncertainty and imprecision with a focus on surveillance

Authors: Edwin El-Mahassni

Addresses: ISRD 200 Labs, Defence Science and Technology Organisation, P.O. Box 1500, Edinburgh SA, 5111, Australia

Abstract: Sensor fusion is the combination of data from two or more sensors in order to obtain more complete or accurate information than that available from a single source. Most methods for achieving this are Bayesian in nature. However, Dempster-Shafer theory (DST) has also attracted attention for its ability to account for ignorance. Further, after the information is fused, often a decision will need to be made regarding as to what course of action should be employed. Indeed, for certain situations, it is not unreasonable to assume that some operators, given their performance history, might be better suited for some jobs over others. The choice of the optimal operator from a set of potential candidates might itself be coloured by uncertainty. In this paper, we aim to model and combine the uncertainties present in fused sensors with the uncertainty of how an operator might then decide given the fused sensors' output. We rely on bringing together previous work that employs the transferable belief model, a variant of DST.

Keywords: transferable belief model; TBM; mathematical theories; uncertainty; sensor fusion; expected decision value formula; decision making; imprecision; surveillance; data; fused sensors; accurate information; single source; Bayesian methods; Thomas Bayes; Dempster-Shafer theory; DST; evidence; Arthur Dempster; Glenn Shafer; ignorance; fused information; performance history; optimal operators; potential candidates; sensor output; applied decision sciences.

DOI: 10.1504/IJADS.2012.050020

International Journal of Applied Decision Sciences, 2012 Vol.5 No.4, pp.318 - 328

Published online: 09 Aug 2014 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article