Efficient Approaches for Report/Cluster Correlation in Multitarget TrackingSystems

摘要

Gating is an important component of most multi-object tracking systems. Itsfunction is to identify sensor reports, e.g., radar or infrared (IR) returns from missiles, planes, etc., that correlate highly with current state estimates (i.e., tracks). For small numbers of objects, it is feasible to calculate a probability of correlation for every track/report pair and reject those whose probabilities fall below some threshold. For large numbers of objects, however, the quadratic growth in the number of pairs for which correlation probabilities are computed by this 'brute force' approach represents an enormous bottleneck. This combination problem is of particular concern in Strategic Defense Initiative (SDI) tracking and correlation for which numbers of objects on the order of 100,000 must be processed in real time. This report discusses an approach that significantly reduces the computational complexity of the correlation process in the TRC tracking and correlation system developed at the Naval Research Laboratory. TRC is a multihypothesis tracker/correlator that was developed to conduct experiments in multiple-target tracking. Unfortunately, early tests of the TRC revealed that combinatorial problems severely limited the size of the scenarios that could be examined. Subsequent analysis demonstrated that these limitations were the result of a correlation (gating) algorithm that scaled in time quadratically in the size of the scenarios that could be examined. Subsequent analysis demonstrated that these limitations were the result of a correlation (gating) algorithm that scaled in time quadratically in the size of the scenario.