Artificial Intelligence

   

Target Type Tracking with a new Probabilistic Belief Transformation

Authors: Jean Dezert, Florentin Smarandache, Albena Tchamova, Pavlina Konstantinova

Abstract-In this paper we analyze the performances of a new probabilistic belief transformation, denoted DSmP, for the sequential estimation of target ID from classifier outputs in the Target Type Tracking problem (TTT). We complicate here a bit the TTT problem by considering three types of targets (Interceptor, Fighter and Cargo) and show through Monte-Carlo simulations the advantages of DSmP over the classical pignistic transformation which is classically used for decision-making under uncertainty when dealing with belief assignments. Based on our previous works for the justification of rules of combination for TTT problem, only the Proportional Conflict Redistribution rule and the hybrid fusion rules are considered in this work for their ability to deal consistently with high conflicting sources of evidence with three different belief assignment modelings.

Comments: 8 pages

Download: PDF

Submission history

[v1] 13 Mar 2010

Unique-IP document downloads: 83 times

Add your own feedback and questions here:
You are equally welcome to be positive or negative about any paper but please be polite. If you are being critical you must mention at least one specific error, otherwise your comment will be deleted as unhelpful.

comments powered by Disqus