Artificial intelligence techniques applied to the non-cooperative identification (NCID) problem

Barrette, Anne; Jamerson, Lawrence; Rogers, Richard G.

doi:10.2514/6.1989-3005

Cited by 3 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first rule defined for this network is that all evidence updates at each node uses an evidential combination rule, of which there are multiple available [ 1 , 13 , 14 , 15 , 28 , 29 ] as shown in Figure 5 . This rule aligns with previous work [ 3 , 27 ]. The inherent drawback to using combination methods to update each node is immediately realized in that the network values are no longer consistent with each other once evidence has been propagated through a transition and combined at a node—that is, the node’s marginal probabilities no longer equal the previous node’s marginal probabilities multiplied by the transitional values, which can be clearly seen in Figure 5 .…”

Section: Transition Potentials Updates Based On Evidencesupporting

confidence: 92%

“…In addition, information about causality between hypotheses is obfuscated. Within a network, however, new evidence provided at each node often still includes uncertainty and should be combined with previous evidence at that node [ 3 , 27 ]. This is especially true for evidence inferred through a transition from another node, which can be considered to be less reliable than directly observed evidence.…”

Section: Transition Potentials Updates Based On Evidencementioning

confidence: 99%

See 1 more Smart Citation

Automatic Updates of Transition Potential Matrices in Dempster-Shafer Networks Based on Evidence Inputs

Dunham

Johnson

Féron

et al. 2020

Sensors

View full text Add to dashboard Cite

Sensor fusion is a topic central to aerospace engineering and is particularly applicable to unmanned aerial systems (UAS). Evidential Reasoning, also known as Dempster-Shafer theory, is used heavily in sensor fusion for detection classification. High computing requirements typically limit use on small UAS platforms. Valuation networks, the general name given to evidential reasoning networks by Shenoy, provides a means to reduce computing requirements through knowledge structure. However, these networks use conditional probabilities or transition potential matrices to describe the relationships between nodes, which typically require expert information to define and update. This paper proposes and tests a novel method to learn these transition potential matrices based on evidence injected at nodes. Novel refinements to the method are also introduced, demonstrating improvements in capturing the relationships between the node belief distributions. Finally, novel rules are introduced and tested for evidence weighting at nodes during simultaneous evidence injections, correctly balancing the injected evidenced used to learn the transition potential matrices. Together, these methods enable updating a Dempster-Shafer network with significantly less user input, thereby making these networks more useful for scenarios in which sufficient information concerning relationships between nodes is not known a priori.

show abstract

Section: Transition Potentials Updates Based On Evidencesupporting

confidence: 92%

Section: Transition Potentials Updates Based On Evidencementioning

confidence: 99%

Automatic Updates of Transition Potential Matrices in Dempster-Shafer Networks Based on Evidence Inputs

Dunham

Johnson

Féron

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…Target intent is one of the main discriminators for classifying whether a target is friend or foe since a particular type of aircraft may be in service in both forces. For example, identification of an aircraft as a MiG-21 or F-5 in the Middle East does not automatically indicate whether the target is friend or foe [9]. Intent assessment is a critical step in a process of determining a target's lethality.…”

Section: Meanings Of Target Intentmentioning

confidence: 99%

“…A modelling concept based on problem characteristics allows synergistic matching of the Artificial Intelligence uncertainty management technique. The concept should define: (1) how to organise the knowledge for solving the intent estimation problem, (2) how to represent a state of belief in any hypotheses drawn from the current scenario, (3) how to update the state of belief given the evidence, and (4) how to make decisions given the current state of belief [9] [18].…”

Section: Rationalementioning

confidence: 99%