Interacting multiple models algorithm with wavelet-based unknown measurement noise estimation

  • Authors:

  • Xiaohua Nie

  • Affiliations:

  • Jiangsu Automatic Research Institute, lianyungang, China

  • Venue:
  • CCDC'09 Proceedings of the 21st annual international conference on Chinese Control and Decision Conference
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

A new maneuvering target interacting multiple models tracking algorithm under unknown measurement noise covariance condition is presented based on the interacting multiple models tracking algorithm of the constant velocity and "current" statistical model (IMM-CVCS). In this paper, the effects of the inaccuracy of the measurement noise covariance on the IMM-CVCS algorithm performance are first analyzed. The feature of the wavelet transform separating a noise signal into the signal and noise parts in real time is combined into IMM-CVCS algorithm. The algorithm adapted the real time change of the measurement noise covariance, at the same time keep tracking availably the constant velocity and the maneuvering target. It is strongly robus, and suit for maneuvering target tracking in the data fusion system of multi-plats and multi-sensors. The simulation results verify the effectiveness of the proposed method.