Time-recursive number-of-tracks estimation for MHT.
In this paper we address the issue of measurement-to-track association within the framework of multiple hypothesistracking (MHT). Speci cally, we generate a maximum a posteriori (MAP) cost as a function of the number of tracksK. This cost is generated, for each K, as a marginalization over the set of hypothesized track-sets. The proposedalgorithm is developed based on a trellis diagram representation of MHT, and a generalized list-Viterbi algorithmfor pruning and merging hypotheses. Compared to methods of pruning hypotheses for either MHT or Bayesianmultitarget tracking, the resulting Viterbi MHT algorithm is less likely to incorrectly drop tracks in high clutterand high missed-detection scenarios. The proposed number-of-tracks estimation algorithm provides a time-recursiveestimate of the number of tracks. It also provides track estimates, allows for the deletion and addition of tracks, andaccounts for false alarms and missed detections.
|Main Author:||Bradley, Jessica.|
|Other Authors:||Buckley, Kevin., Perry, Richard.|