By Topic

Gaussian mixture modeling of rule base to track maneuvering targets, using fuzzy EKF

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

4 Author(s)
Pradeepa, R. ; Naval Phys. & Oceanogr. Lab., Kochi, India ; Unnikrishnan, A. ; Deepa, V. ; Mija, S.J.

Various techniques have been developed for the tracking of maneuvering targets. When the target maneuvers, the quality of the state estimates provided by the constant velocity filter can degrade significantly. Unknown target acceleration during the maneuver appears as excessive process noise on the target model and the noise variance changes drastically. In this paper, the authors propose a new fuzzy logic based algorithm for bearing only tracking (BOT) for a maneuvering target tracking. The unknown target acceleration is regarded as additive process noise, and the time-varying variance of the overall process noise is computed using a fuzzy system as a universal approximator. The bearing error distribution is represented as a mixture of Gaussians by Gaussian mixture modeling (GMM), and the fuzzy rule base system is designed using the parameters of the mixture components. The expectation maximization algorithm is utilised to learn the GMM parameters. We have demonstrated a fast tracking of maneuvering target with only one filter, using this proposed method. The performance of the proposed method viz. fuzzy-GMM based EKF, is compared with the method based on the popular Chi-square test and interacting multiple model filter (IMM), through computer simulations.

Published in:

TENCON 2009 - 2009 IEEE Region 10 Conference

Date of Conference:

23-26 Jan. 2009