Izvorni znanstveni članak
Tracking gate algorithm for general nonlinear systems with target class information
Hongwei Quan
; Xijing University
Dongliang Peng
; Institute of Information and Control, Hangzhou Dianzi University, Hangzhou
Junhua Li
; College of Control Engineering, Xijing University, Xi'an
Sažetak
Multitarget tracking in clutter usually involves data association. The traditional method to handle this problem is to construct a tracking gate for predicting the position of the target being tracked, which leads to great uncertainties of measurements-to-tracks association with the unknown class of targets. This paper proposes a new tracking gate algorithm for general nonlinear systems, where the target class information is integrated into our algorithm. Firstly, a joint probability density description of the target state and target class is given, by which the tracking gates for each target class in general nonlinear system are developed. Then, a simulation with ground formation target tracking is carried out to examine our algorithm. Compared with the traditional tracking gate, the results demonstrate that our algorithm has significantly improved the probabilities of the measurements-to-tracks association.
Ključne riječi
target tracking; data association; tracking gate; nonlinear system
Hrčak ID:
151834
URI
Datum izdavanja:
27.1.2016.
Posjeta: 1.280 *