Multi-entity Bayesian networks for situation assessment

Wright, Edward J.; Mahoney, Suzanne M.; Laskey, Kathryn Blackmond; Takikawa, Masami; Levitt, Tod S.

doi:10.1109/icif.2002.1020889

Cited by 30 publications

(13 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A number of uncertainty calculi (e.g., Dempster-Shafer, Bayesian, Certainty Factors) have been shown to be effective in interpretation problems that require knowledge-based solutions. In fact, some recent work directly related to some of the problems being addressed by the present project has used Bayesian Belief Nets (Wright et al 2002;Gonsalves and Cunnigham 2000) to represent and propagate uncertainty. Some intelligence analysts believe Bayesian Belief Nets provide an accurate representation of certain aspects of the way that Army analysts reason about answering PIRs (Schlabach 2002).…”

Section: Levels Of Analysismentioning

confidence: 99%

Fusion-Based Knowledge for the Objective Force

Powell¹,

Broome²

2002

View full text Add to dashboard Cite

Introduction.Army Vision 2010 identifies information superiority as the key enabler for such force characteristics as dominant maneuver and precision engagement. These concepts are also central to the design and implementation of the Army's Future Combat System (FCS) and Objective Force. To establish and maintain information superiority, analysts and decision-makers need to identify, analyze and interpret pertinent information relative to achieving their task requirements. Currently, the sheer volume of information presented to Army intelligence analysts significantly exceeds their capabilities to fully analyze and interpret it in a timely manner. Consequently, the answers to commanders' critical information requirements (CCIRs) and priority intelligence requirements (PIRs) are typically based on a hasty, partial analysis of the information available. This condition of information overload experienced by analysts has the potential to significantly worsen for various reasons. First, our capabilities to collect, communicate and store data/information are steadily rising. Second, faster, more precise, and more lethal battlespace systems of the adversary cause an increase in operational tempo, as well as an increased risk to one's own forces, thereby resulting in more severe time constraints on analysis and decision-making. The nature of the analytical and interpretive tasks required to answer PIRs, and our ability to explain and justify their derivation, have largely been outside the realm of current machine capabilities. In recent years, a number of technologies and approaches have been developed (or matured) that show promise for addressing some of the key sources of difficulty characterizing this set of complex tasks either by emulating human methods or by providing automated support for aspects of these tasks that strain or exceed human cognitive capacities.To address this set of complex military intelligence problems, the U.S. Army Communications-Electronics Command and the U.S. Army Research Laboratory have submitted a collaborative proposal that would be carried out under the Army's Science and Technology Objective Program starting in FY03. One perspective for viewing this project is the Joint Directors of Laboratories (JDL) Data Fusion Model (Steinberg et al., 1998). With respect to this model, the present project will focus on problems associated primarily with data fusion Levels 2 and 3. However, it is our belief that data fusion problems are more likely to be understood and solved if they are approached more holistically by utilizing data fusion at any or all levels, if appropriate, to help solve a problem on a given level. The present paper provides a description of the technical challenges facing this project, and our current views on addressing them. The remainder of this paper begins by sketching the intelligence cycle and the military decision making process. Next, we discuss operational problems this project will address. This is followed by a description of some of the approaches and technologies...

show abstract

Section: Levels Of Analysismentioning

confidence: 99%

Fusion-Based Knowledge for the Objective Force

Powell¹,

Broome²

2002

View full text Add to dashboard Cite

show abstract

“…The Bayesian belief network [9] [10] is one of the most popular frameworks used for situation assessment. In this framework, situations become hypotheses and objects are treated as evidences.…”

Section: Introductionmentioning

confidence: 99%

Information fusion for object & situation assessment in sensor networks

Srivastav

Wen

Hendrick

et al. 2011

Proceedings of the 2011 American Control Conference

View full text Add to dashboard Cite

A semantic framework for information fusion in sensor networks for object and situation assessment is proposed. The overall vision is to construct machine representations that would enable human-like perceptual understanding of observed scenes via fusion of heterogeneous sensor data. In this regard, a hierarchical framework is proposed that is based on the Data Fusion Information Group (DFIG) model. Unlike a simple set-theoretic information fusion methodology that leads to loss of information, relational dependencies are modeled as cross-machines called relational Probabilistic Finite State Automata using the xD-Markov machine construction. This leads to a tractable approach for modeling composite patterns as structured sets for both object and scene representation. An illustrative example demonstrates the superior capability of the proposed methodology for pattern classification in urban scenarios.

show abstract

“…the pedigree metadata) that accompanies it. Situation assessment, called "level-two data fusion," involves object-level aggregation and reasoning about an unknown number of interacting entities at various levels in a hierarchy [2]. It involves the estimation of relationships between entities, such as aggregates and the behavior and intent of these aggregates [1].…”

Section: Introductionmentioning

confidence: 99%

Pedigree Information for Enhanced Situation and Threat Assessment

Ceruti¹,

Das

Ashenfelter³

et al. 2006

2006 9th International Conference on Information Fusion

View full text Add to dashboard Cite

This paper describes how pedigree is used to support and enhance situation and threat assessment. It is based on the findings of the technology group of the Data Fusion Levels Two and Three Workshop sponsored by the Office of Naval Research held in Arlington, VA from 15-18 Nov. 2005. It identifies areas that need improvement in situation assessment and threat assessment, such as interoperability, automation, pedigree management, system usability, reliability, and uncertainty. The concept of pedigree must include "standard" metadata, lineage, plus a computational model of the quality of the information. The system must automatically propagate changes and update to derived products when source information or sourcepedigree information changes. Several other processes must be automated: generate pedigree, identify and auto fill gaps, fuse pedigree, update pedigree, display of information quality and confidence. The paper concludes with suggestions for future research and development.

show abstract

Multi-entity Bayesian networks for situation assessment

Cited by 30 publications

References 2 publications

Fusion-Based Knowledge for the Objective Force

Fusion-Based Knowledge for the Objective Force

Information fusion for object & situation assessment in sensor networks

Pedigree Information for Enhanced Situation and Threat Assessment

Contact Info

Product

Resources

About

Multi-entity Bayesian networks for situation assessment

Cited by 30 publications

References 2 publications

Fusion-Based Knowledge for the Objective Force

Fusion-Based Knowledge for the Objective Force

Information fusion for object &amp; situation assessment in sensor networks

Pedigree Information for Enhanced Situation and Threat Assessment

Contact Info

Product

Resources

About

Information fusion for object & situation assessment in sensor networks