Joint probabilistic data association filter for partially unresolved target groups

In many surveillance problems the observedobjects are so closely spaced that they cannot always be resolvedby the sensor(s). Typical examples for partially unresolvedmeasurements are the surveillance of aircraft in formation,and convoy tracking for ground surveillance. Ignoringthe limited sensor resolution in a tracking systemmay lead to degraded tracking performance, in particularunwanted track-losses. In this paper, we further discuss arecently presented extension of the resolution model by Kochand van Keuk to the case of arbitrary object numbers, andit is shown how that model can be incorporated into theJoint Probabilistic Data Association Filter (JPDAF). Further,through simulations of a ground target tracking scenario,it is shown how the incorporation of the resolutionmodel improves tracking performance when targets are partiallyunresolved