Translator Disclaimer
22 May 2015 Reliable sources and uncertain decisions in multisensor systems
Author Affiliations +
Conflict among information sources is a feature of fused multisource and multisensor systems. Accordingly, the subject of conflict resolution has a long history in the literature of data fusion algorithms such as that of Dempster-Shafer theory (DS). Most conflict resolution strategies focus on distributing the conflict among the elements of the frame of discernment (the set of hypotheses that describe the possible decisions for which evidence is obtained) through rescaling of the evidence. These “closed-world” strategies imply that conflict is due to the uncertainty in evidence sources stemming from their reliability. An alternative approach is the “open-world” hypothesis, which allows for the presence of "unknown" elements not included in the original frame of discernment. Here, conflict must be considered as a result of uncertainty in the frame of the discernment, rather than solely the province of evidence sources. Uncertainty in the operating environment of a fused system is likely to appear as an open-world scenario. Understanding the origin of conflict (source versus frame of discernment uncertainty) is a challenging area for research in fused systems. Determining the ratio of these uncertainties provides useful insights into the operation of fused systems and confidence in their decisions for a variety of operating environments. Results and discussion for the computation of these uncertainties are presented for several combination rules with simulated data sets.
© (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Christian Minor and Kevin Johnson "Reliable sources and uncertain decisions in multisensor systems", Proc. SPIE 9498, Multisensor, Multisource Information Fusion: Architectures, Algorithms, and Applications 2015, 949803 (22 May 2015);


Bias estimation using targets of opportunity
Proceedings of SPIE (September 21 2007)
On target track covariance consistency
Proceedings of SPIE (May 19 2006)
Remaining useful life prediction based on known usage data
Proceedings of SPIE (August 01 2003)
Metrics for evaluating track covariance consistency
Proceedings of SPIE (September 13 2007)

Back to Top