The Open Automation and Control Systems Journal

2015, 7 : 991-995
Published online 2015 September 10. DOI: 10.2174/1874444301507010991
Publisher ID: TOAUTOCJ-7-991

Target Tracking Algorithm Based on Improved Unscented Kalman Filter

Wang Yingyan and Zeng Rui
School of Electro-mechanical and Information Technology, Yi Wu Industrial &Commercial College, Yi Wu, Zhejiang, 322000, P.R. China.

ABSTRACT

In order to improve the performance of target tracking and solve the defects of unscented Kalman filter, a target tracking algorithm based on improved unscented Kalman filter is proposed in this paper. Firstly, the fading factor is introduced into the filter based on strong tracking filter to avoid the filter divergence, and then wavelet transform is used to estimate the statistical characteristics of measurement noise to improve unscented Kalman filter tracking ability, finally the simulation experiment is used to test the performance of algorithm. The results show that the proposed algorithm increases adaptive ability of target tracking, and obtain good performance for weak maneuvering and non maneuvering target tracking, and fastens the tracking speed.

Keywords:

Strong tracking filter, target tracking, unscented kalman filter, wavelet transform.