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...

Full description

Saved in:
Bibliographic Details
Published in2010 2nd International Conference on Information Engineering and Computer Science pp. 1 - 4
Main Authors Chen Song-lin, Xu Yi-bing, Zhu Ming
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.12.2010
Subjects
Clutter
Covariance matrix
Joints
Measurement uncertainty
Probabilistic logic
Target tracking
Time measurement
Online AccessGet full text

Cover

More Information
Summary: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