Optimization and Benchmarking of Truncated Dempster-Shafer for Airborne Surveillance
process incomplete information, which implies that ignorance should be defined mathematically
not require a priori information, which usually rules out Bayes reasoning
handle conflicts between contact/track so that formalisms such as Dempster- Shafer (DS), where conflict is precisely defined, are favored
have a real-time method, which means that DS truncation is a must
present the operator with the best ID, i.e. give preference to singletons
present the operator with the next best thing, i.e. doublets, then triplets...
be tested operationally, which translates into building complex scenarios
ensure that ordinary DS explodes, which requires a large complex Platform Data Base (PDB)
resist Counter Measures (CM) such that several contact reports should conflict with information contained in the PDB
resist false associations, i.e. ESM reports associated to wrong track
KeywordsDecision Rule Synthetic Aperture Radar Sensor Fusion Basic Probability Assignment Canadian Contingent
Unable to display preview. Download preview PDF.
- 2.Jouan, A. Valin, P. and Bossé, E., (1999) “Testbed for Fusion of Imaging and Non-Imaging Sensor Attributes in Airborne Surveillance Missions”, FUSION’ 99 conference, Vol. 2, pp. 823–830, Sunnyvale, CA.Google Scholar