Title :
Fixed-order ℌ∞ filtering for discrete-time Markovian jump linear systems with unobservable jump modes
Author :
Shu, Zhan ; Lam, James ; Hu, Yuebing
Author_Institution :
Dept. of Mech. Eng., Univ. of Hong Kong, Hong Kong, China
Abstract :
In practical applications, it is often encountered that the jump modes of a Markovian jump linear system may not be fully accessible to the filter, and thus designing a filter which partially or totally independent of the jump modes becomes significant. In this paper, by virtue of a new stability and Hinfin performance characterization, a novel necessary and sufficient condition for the existence of mode-independent Hinfin filters is established in terms of a set of nonlinear matrix inequalities that possess special properties for computation. Then, two computational approaches are developed to solve the condition. One is based on the solution of a set of linear matrix inequalities (LMIs), and the other is based on the sequential LMI optimization with more computational effort but less conservatism. In addition, a specific property of the feasible solutions enables one to further improve the solvability of these two computational approaches.
Keywords :
Markov processes; discrete time filters; discrete time systems; filtering theory; iterative methods; linear matrix inequalities; linear systems; optimisation; stability; LMI optimization; computational approach; discrete-time Markovian jump linear system; fixed-order Hinfin filtering theory; iterative calculation; linear matrix inequality; mode-independent Hinfin filter; nonlinear matrix inequality; stability; unobservable jump mode; Control systems; Filtering; Kalman filters; Linear matrix inequalities; Linear systems; Noise robustness; Nonlinear filters; Stability; Statistics; Sufficient conditions; ℌ∞ filtering; Fixed-order filter; Markovian jump linear systems; iterative calculation; linear matrix inequality (LMI); mode-independence;
Conference_Titel :
Asian Control Conference, 2009. ASCC 2009. 7th
Conference_Location :
Hong Kong
Print_ISBN :
978-89-956056-2-2
Electronic_ISBN :
978-89-956056-9-1