Maneuvering target tracking based on swarm intelligent unscented particle filtering

  • Authors:

  • Yue-Long Wang;Fu-Chang Ma

  • Affiliations:

  • Control Technology Research Institute, Taiyuan University of Technology, Taiyuan, China and College of Electrical and Power Engineering, Taiyuan University of Technology, Taiyuan, China;Control Technology Research Institute, Taiyuan University of Technology, Taiyuan, China

  • Venue:
  • AICI'11 Proceedings of the Third international conference on Artificial intelligence and computational intelligence - Volume Part I
  • Year:
  • 2011

Quantified Score

Hi-index 0.01

Visualization

Abstract

To improve the performance of maneuvering target tracking, a based on Swarm intelligent unscented particle filtering was proposed. In the new filter, application of the un-scented Kalman filter is used to generate the proposal distribution. Moreover, by introducing the thought of artificial fish school algorithm into particle filtering, the particle distribution and filtering accuracy can be improved. In simulation experiment, "Coordinated Turns" model is taken as dynamic model of maneuvering target. The simulation results show that unscented particle filtering optimized by the artificial fish swarm algorithm (AFSA-UPF) has quite higher tracking precision than the PF and UPF by analyzing the tracking performance and the root-mean-square error.