Modified Joint Probability Data Association Algorithm Controlling Track Coalescence

Joint Probabilistic Data Association has been proven to be effective in tracking multiple targets from measurements amidst clutter and missed detections. But the traditional Joint Probabilistic Data Association algorithm will cause track coalescence when the targets are parallel neighboring or small-angle crossing. To avoid track coalescence, a modified Joint Probabilistic Data Association algorithm is proposed in this paper. An exclusive measurement is defined for every target in the new algorithm, and an arbitrary positive scaling factor will be employed to multiply the maximum probabilities of every target associated with measurements. At last, the Entropy Value Method will be used twice to give weights to the association probabilities of every measurement. The simulation results show that the new algorithm can effectively avoid track coalescence in all kinds of scenarios and its performance is better than the track performance when the Entropy Value Method is used only one time.