Multitarget list Viterbi tracking algorithm

We present an approach to multitarget tracking algorithm development. The approach is based on a trellis diagram which depicts the possible progressions of sequences of location measurements, over time. Resulting algorithms are sequential and very flexible in that the approach can handle multiple tracks, track initiation, missed detections, false alarms, and various performance cost functions, while managing computation cost by pruning based on track feasibility. The output of a resulting algorithm can be either a single best set of K tracks, or a list of L best best sets of K tracks. The latter is useful, for example, in data fusion where information from other platforms can be used to select one set from the list.