An interacting multipattern joint probabilistic data association (IMP-JPDA) algorithm for multitarget tracking

This paper presents a newly developed algorithm for multitarget tracking us ing multipattern data association. Multipattern data association is based o n a novel approach for tracking multiple targets using multiple patterns ex tracted from measurement sequences. A Markov chain controls the switching b ehavior of multipatterns, which allows an interacting operation of multipat terns. The introduction of patterns leads to a new paradigm of developing h igh-performance algorithms. A powerful tracking algorithm - interacting mul tipattern joint probabilistic data association (IMP-JPDA) is developed. The IMP-JPDA algorithm can be considered as a generalized joint probabilistic data association (JPDA) algorithm; when only one scan of measurements is us ed, IMP-JPDA reduces to JPDA. (C) 2000 Elsevier Science B.V. All rights res erved.