Application of Cluster Analysis on Gaussian-Mixture Probability Hypothesis Density Filter for Multiple Extended Target Tracking

DOI

10.2991/msota-16.2016.72How to use a DOI?

Keywords

information fusion; extended target tracking; track initiation; measurement partition

Abstract

Based on the multiple Extended Target Gaussian-Mixture Probability Hypothesis Density (GM-PHD) filter, a new algorithm of extended target track initiation and observation partition in the clutter environment are proposed. Firstly the paper take clustering trend of observation into account when carrying track initiation, which make the clustering results more convincing and increase computational efficiency; Then, the improved partition algorithm introduce the concepts of core distance and reached distance to save the sequence of measurement points and extract the measurement cluster. Simulation experiments show that the proposed initiation algorithm has a better computational cost over traditional algorithm when carrying track initiation. In the partition process, the new algorithm is not sensitive to the parameter selection and extended target measurement density, at the same time, the computational cost decreases.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Zhuo Cao
AU  - Xinxi Feng
AU  - Yinglei Cheng
AU  - Hongyan Li
PY  - 2016/12
DA  - 2016/12
TI  - Application of Cluster Analysis on Gaussian-Mixture Probability Hypothesis Density Filter for Multiple Extended Target Tracking
BT  - Proceedings of 2016 International Conference on Modeling, Simulation and Optimization Technologies and Applications (MSOTA2016)
PB  - Atlantis Press
SP  - 335
EP  - 339
SN  - 2352-538X
UR  - https://doi.org/10.2991/msota-16.2016.72
DO  - 10.2991/msota-16.2016.72
ID  - Cao2016/12
ER  -