K Nearest Neighbor Joint Possibility Data Association Algorithm

For the problem of tracking multiple targets, the Joint Probabilistic Data Association approach has shown to be very effective in handling clutter and missed detections. However, it tends to coalesce neighboring tracks and ignores the coupling between those tracks. To avoid track coalescence, a K Ne...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2010 2nd International Conference on Information Engineering and Computer Science S. 1 - 4
Hauptverfasser: Chen Song-lin, Xu Yi-bing, Zhu Ming
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.12.2010
Schlagworte:
Clutter
Covariance matrix
Joints
Measurement uncertainty
Probabilistic logic
Target tracking
Time measurement
ISBN:1424479398, 9781424479399
ISSN:2156-7379
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:For the problem of tracking multiple targets, the Joint Probabilistic Data Association approach has shown to be very effective in handling clutter and missed detections. However, it tends to coalesce neighboring tracks and ignores the coupling between those tracks. To avoid track coalescence, a K Nearest Neighbor Joint Probabilistic Data Association algorithm is proposed in this paper. Like the Joint Probabilistic Data Association algorithm, the association possibilities of target with every measurement will be computed in the new algorithm, but only the first K measurements whose association probabilities with the target are larger than others' are used to estimate target's state. Finally, through Monte Carlo simulations, it is shown that the new algorithm is able to avoid track coalescence and keeps good tracking performance in heavy clutter and missed detections.
ISBN:1424479398
9781424479399
ISSN:2156-7379
DOI:10.1109/ICIECS.2010.5677877