DocumentCode :
1363649
Title :
New interacting multiple model algorithms for the tracking of the manoeuvring target [Brief Paper]
Author :
Fu, Xiao ; Jia, Yunde ; Du, Jinyang ; Yu, F.
Author_Institution :
Dept. of Syst. & Control, Beihang Univ. (BUAA), Beijing, China
Volume :
4
Issue :
10
fYear :
2010
fDate :
10/1/2010 12:00:00 AM
Firstpage :
2184
Lastpage :
2194
Abstract :
This study is devoted to the problem of state estimation of discrete-time stochastic systems with Markov switching parameters. Three improved interacting multiple model (IMM) algorithms for manoeuvring target tracking are presented, in which the filter outputs are combined based on three optimal multi-model fusion criterions weighted by scalars, diagonal matrices and general matrices, respectively. The proposed algorithms can receive the optimal state estimations of target in the linear minimum variance sense. It is proved that the traces of variance matrices of tracking errors in three proposed algorithms are less than the trace in the classical IMM algorithm. Extensive Monte Carlo simulations verify that the proposed algorithms are effective and have an absolute advantage in the velocity estimation. In particular, one of the proposed algorithms is obviously better than the IMM algorithm in accuracy and elapsed time and, therefore, can be a competitive alternative to the classical IMM algorithm for the tracking of manoeuvring target in real time.
Keywords :
Markov processes; Monte Carlo methods; discrete time systems; parameter estimation; state estimation; stochastic systems; target tracking; Markov switching parameters; Monte Carlo simulations; diagonal matrices; discrete-time stochastic systems; interacting multiple model; linear minimum variance sense; manoeuvring target; optimal multimodel fusion; state estimation; tracking errors; variance matrices; velocity estimation;
fLanguage :
English
Journal_Title :
Control Theory & Applications, IET
Publisher :
iet
ISSN :
1751-8644
Type :
jour
DOI :
10.1049/iet-cta.2009.0583
Filename :
5611738
Link To Document :
بازگشت