Skip to main navigation Skip to search Skip to main content

An improved probabilistic data association method for multi-target tracking in clutter

  • Yazhao Wang*
  • , Yingmin Jia
  • , Junping Du
  • , Fashan Yu
  • *Corresponding author for this work
  • Beihang University
  • Beijing University of Posts and Telecommunications
  • Henan Polytechnic University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Improved data association methods for multi-target tracking are discussed in this paper. We introduce an exquisite gating technique based on the well-known probabilistic data association (PDA) filter, which calculates the association probabilities to the targets being tracked for each validated measurement. An important feature of the new algorithm is that it can divide the joint validation region into several parts in which the association probabilities can be calculated with more accuracy. The algorithm is tested and compared with both the generic PDA and joint probabilistic data association (JPDA) filter. Simulations are also presented to illustrate the effectiveness of the algorithm.

Original languageEnglish
Title of host publicationElectrical, Information Engineering and Mechatronics 2011 - Proceedings of the 2011 International Conference on Electrical, Information Engineering and Mechatronics, EIEM 2011
Pages1375-1382
Number of pages8
DOIs
StatePublished - 2012
Event2011 International Conference on Electrical, Information Engineering and Mechatronics, EIEM 2011 - Jiaozuo, Henan, China
Duration: 23 Dec 201125 Dec 2011

Publication series

NameLecture Notes in Electrical Engineering
Volume138 LNEE
ISSN (Print)1876-1100
ISSN (Electronic)1876-1119

Conference

Conference2011 International Conference on Electrical, Information Engineering and Mechatronics, EIEM 2011
Country/TerritoryChina
CityJiaozuo, Henan
Period23/12/1125/12/11

Keywords

  • Gating technique
  • Multi-target tracking
  • Probabilistic data association
  • Validation region

Fingerprint

Dive into the research topics of 'An improved probabilistic data association method for multi-target tracking in clutter'. Together they form a unique fingerprint.

Cite this