Previous [ 1] [ 2] [ 3] [ 4] [ 5] [ 6] [ 7] [ 8] [ 9] [ 10] [ 11] [ 12] [ 13] [ 14] [ 15] [ 16] [ 17] [ 18] [ 19] [ 20] [ 21] [ 22] [ 23] [ 24]

@

Journal of Information Science and Engineering, Vol. 26 No. 4, pp. 1479-1490 (July 2010)

The Dual-Kalman Filtering and Neural Solutions to Maneuvering Estimation Problems*

YI-NUNG CHUNG1, DEND-JYI JUANG2, KUO-CHANG HU2, MING-LIANG LI2,3 AND KAI-CHIH CHUANG2
1Department of Electrical Engineering
National Changhua University of Education
Changhua, 500 Taiwan
E-mail: ynchung@cc.ncue.edu.tw
2Department of Electrical Engineering
Da-Yeh University
Changhua, 515 Taiwan
3Department of Electronic Engineering
Nan Kai University of Technology
Nantou County, 542 Taiwan

Tracking maneuvering targets in a radar system is more complicated because the target accelerations cannot be directly measured. It may occur severe tracking error even diverge the estimates when the maneuvering situations are happened. In this paper, we develop a Dual-Kalman filtering algorithm to handle the maneuvering targets tracking problems. In this approach, two collaborative Kalman filters are devised which one for pursuing the track estimation and the other for estimating the status of maneuver. Based on this approach, the most approximate targets acceleration can be detected and estimated in real time. Moreover, it is also shown that one Competitive Hopfield Neural Network-based data association combined with a multiple-target tracking system demonstrates target tracking capability.

Keywords: maneuvering targets, Dual-Kalman filtering algorithm, competitive hopfield neural network, data association

Full Text () Retrieve PDF document (201007_20.pdf)

Received April 21, 2008; revised October 25, 2008; accepted December 24, 2008.
Communicated by Pau-Choo Chung.
* This paper was partially supported by the National Science Council under Grant NSC 95-2221-E-212-021.