Intelligent Kalman filter for tracking a manoeuvring target

B. J. Lee, J. B. Park, Y. H. Joo, S. H. Jin

Research output: Contribution to journalArticlepeer-review

42 Citations (Scopus)

Abstract

The Kalman filter (KF) has been widely used in the state estimation of a target, but in the presence of a manoeuvre, its performance may be seriously degraded because the manoeuvre appears as extensive noise on the target model and the process noise variance cannot cover it. To solve this problem, a new intelligent KF (IKF) is proposed for tracking a manoeuvring target. The unknown target acceleration is regarded as additive process noise, and the time-varying variance of the overall process noise is computed in an intelligent manner using a fuzzy system as universal approximator. To optimise the fuzzy system, a genetic algorithm (GA) or DNA coding method can be utilised, and the filter is then termed a GA-based IKF or DNA coding-based IKF according to the optimisation tool used. The proposed IKF can effectively treat a target manoeuvre with only one filter and can relax the additional requirements of conventional manoeuvring target tracking methods. The performance of the proposed IKFs is compared with that of multiple model methods through computer simulations.

Original languageEnglish
Pages (from-to)344-350
Number of pages7
JournalIEE Proceedings: Radar, Sonar and Navigation
Volume151
Issue number6
DOIs
Publication statusPublished - 2004 Dec

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications

Fingerprint Dive into the research topics of 'Intelligent Kalman filter for tracking a manoeuvring target'. Together they form a unique fingerprint.

Cite this