Informatics and Applications
2019, Volume 13, Issue 2, pp 7-15
ON THE CONDITIONALLY MINIMAX NONLINEAR FILTERING CONCEPT DEVELOPMENT: FILTER MODIFICATION AND ANALYSIS
Abstract
The main result of the research is a new suboptimal filter developed from the conditionally minimax nonlinear filtering (CMNF) method for nonlinear stochastic systems in discrete time. The main idea of the proposed modification is to omit the time and resource consuming phase ofa priori CMNF parameter calculation in favor of their online approximation together with the current state estimation. In the original CMNF filter, the simulation study is used in order to approximate dynamic system parameters' unconditional expectation and covariances, while the modified version deals with the conditional moments which are also calculated by means of the Monte-Carlo method. The proposed filter modification is provided with the minimax justification, similar to the underlying CMNF concept. Simulation examples show the proposed algorithm effectiveness and performance gain in comparison with the original conditionally minimax nonlinear filter.
[+] References (10)
- Liptser, R., and A. N. Shiryaev. 2001. Statistics of random processes. Berlin - Heidelberg: Springer-Verlag. 427 p.
- Kallianpur, G. 1980. Stochastic filtering theory. New York, NY: Springer-Verlag. 318 p.
- Pugacev, V. S. 1978. Recursive estimation of variables and parameters in stochastic systems described by autoregression equations. Sov. Math. Dokl. 19:991-995.
- Pugachev, V. S. 1979. Estimation of variables and parameters in discrete-time nonlinear systems. Automat. Rem. Contr. 40(4):39-50.
- Pankov, A. R. 1993. Recurrent conditionally minimax fil-tering ofprocesses in nonlinear difference stochastic sys-tems. J. Comput. Sys. Sc. Int. 31(4):54-60.
- Pankov, A. R., and A. V. Bosov. 1994. Conditionally mini-max algorithm for nonlinear system state estimation. IEEE T. Automat. Contr. 39(8):1617-1620.
- Sinitsyn, I.N., and E. R. Korepanov. 2015. Ustoychivye lineynye uslovno-optimal'nye fil'try i ekstrapolyatory dlya stokhasticheskikh sistem s mul'tiplikativnymi shumami [Stable linear conditionally optimal filters and extrapo- lators for stochastic systems with multiplicative noises]. Informatika i ee Primeneniya - Inform. Appl. 9(1):70-75.
- Sinitsyn, I.N., and E. R. Korepanov. 2016. Normal'nye uslovno-optimal'nye fil'try i ekstrapolyatory Pugacheva dlya stokhasticheskikh sistem, lineynykh otnositel'no so- stoyaniya [Normal Pugachev conditionally-optimal filters and extrapolators for state linear stochastic systems]. Informatika i ee Primeneniya - Inform. Appl. 10(2): 14-23.
- Sinitsyn, I. N., V. I. Sinitsyn, and E. R. Korepanov. 2017. Modifitsirovannye ellipsoidal'nye uslovno-optimal'nye fil'try dlya nelineynykh stokhasticheskikh sistem na mno-
goobraziyakh [Modificated ellipsoidal conditionally optimal filters for nonlinear stochastic systems on manifolds]. Informatika i ee Primeneniya - Inform. Appl. 11(2):101- 111.
- Borisov, A. V., A. V. Bosov, A. I. Kibzun, G. B. Miller, and K. V. Semenikhin. 2018. The conditionally minimax non-linear filtering method and modern approaches to state estimation in nonlinear stochastic systems. Automat. Rem. Contr. 79(1):1-11.
[+] About this article
Title
ON THE CONDITIONALLY MINIMAX NONLINEAR FILTERING CONCEPT DEVELOPMENT: FILTER MODIFICATION AND ANALYSIS
Journal
Informatics and Applications
2019, Volume 13, Issue 2, pp 7-15
Cover Date
2019-06-30
DOI
10.14357/19922264190202
Print ISSN
1992-2264
Publisher
Institute of Informatics Problems, Russian Academy of Sciences
Additional Links
Key words
nonlinear stochastic observation system in discrete time; conditionally minimax nonlinear filtering; Monte-Carlo simulation
Authors
A. V. Bosov  and G. B. Miller
Author Affiliations
Institute of Informatics Problems, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
|