Nonlinear state prediction by separation approach for continuous-discrete stochastic systems

Jaroslav Švácha; Miroslav Šimandl

Kybernetika (2008)

  • Volume: 44, Issue: 1, page 61-74
  • ISSN: 0023-5954

Abstract

top
The paper deals with a filter design for nonlinear continuous stochastic systems with discrete-time measurements. The general recursive solution is given by the Fokker–Planck equation (FPE) and by the Bayesian rule. The stress is laid on the computation of the predictive conditional probability density function from the FPE. The solution of the FPE and its integration into the estimation algorithm is the cornerstone for the whole recursive computation. A new usable numerical scheme for the FPE is designed. In the scheme, the separation technique based on the upwind volume method and the finite difference method for hyperbolic and parabolic part of the FPE is used. It is supposed that separation of the FPE and choice of a suitable numerical method for each part can achieve better estimation quality comparing to application of a single numerical method to the unseparated FPE. The approach is illustrated in some numerical examples.

How to cite

top

Švácha, Jaroslav, and Šimandl, Miroslav. "Nonlinear state prediction by separation approach for continuous-discrete stochastic systems." Kybernetika 44.1 (2008): 61-74. <http://eudml.org/doc/33913>.

@article{Švácha2008,
abstract = {The paper deals with a filter design for nonlinear continuous stochastic systems with discrete-time measurements. The general recursive solution is given by the Fokker–Planck equation (FPE) and by the Bayesian rule. The stress is laid on the computation of the predictive conditional probability density function from the FPE. The solution of the FPE and its integration into the estimation algorithm is the cornerstone for the whole recursive computation. A new usable numerical scheme for the FPE is designed. In the scheme, the separation technique based on the upwind volume method and the finite difference method for hyperbolic and parabolic part of the FPE is used. It is supposed that separation of the FPE and choice of a suitable numerical method for each part can achieve better estimation quality comparing to application of a single numerical method to the unseparated FPE. The approach is illustrated in some numerical examples.},
author = {Švácha, Jaroslav, Šimandl, Miroslav},
journal = {Kybernetika},
keywords = {stochastic systems; state estimation; nonlinear filters; Fokker –Planck equation; numerical solutions; finite volume method; finite difference method; stochastic systems; state estimation; nonlinear filters; Fokker-Planck equation; numerical solutions; finite volume method; finite difference method},
language = {eng},
number = {1},
pages = {61-74},
publisher = {Institute of Information Theory and Automation AS CR},
title = {Nonlinear state prediction by separation approach for continuous-discrete stochastic systems},
url = {http://eudml.org/doc/33913},
volume = {44},
year = {2008},
}

TY - JOUR
AU - Švácha, Jaroslav
AU - Šimandl, Miroslav
TI - Nonlinear state prediction by separation approach for continuous-discrete stochastic systems
JO - Kybernetika
PY - 2008
PB - Institute of Information Theory and Automation AS CR
VL - 44
IS - 1
SP - 61
EP - 74
AB - The paper deals with a filter design for nonlinear continuous stochastic systems with discrete-time measurements. The general recursive solution is given by the Fokker–Planck equation (FPE) and by the Bayesian rule. The stress is laid on the computation of the predictive conditional probability density function from the FPE. The solution of the FPE and its integration into the estimation algorithm is the cornerstone for the whole recursive computation. A new usable numerical scheme for the FPE is designed. In the scheme, the separation technique based on the upwind volume method and the finite difference method for hyperbolic and parabolic part of the FPE is used. It is supposed that separation of the FPE and choice of a suitable numerical method for each part can achieve better estimation quality comparing to application of a single numerical method to the unseparated FPE. The approach is illustrated in some numerical examples.
LA - eng
KW - stochastic systems; state estimation; nonlinear filters; Fokker –Planck equation; numerical solutions; finite volume method; finite difference method; stochastic systems; state estimation; nonlinear filters; Fokker-Planck equation; numerical solutions; finite volume method; finite difference method
UR - http://eudml.org/doc/33913
ER -

References

top
  1. Daum F. E., New exact nonlinear filters, In: Bayesian Analysis of Time Series and Dynamic Models (J. C. Spall, ed.), Marcel Dekker, New York 1988, pp. 199–226 (1988) 
  2. Higham D. J., Kloeden P. E., Maple and Matlab for Stochastic Differential Equations in Finance, Technical Report. University of Strathclyde 2001 
  3. Chang J. S., Cooper G., A practical difference scheme for Fokker–Planck equations, J. Comput. Phys. 6 (1970), 1–16 (1970) Zbl0221.65153
  4. Moral P. del, Jacod J., Interacting particle filtering with discrete observations, In: Sequential Monte Carlo Methods in Practice (A. Doucet, N. de Freitas, and N. Gordon, eds.), Springer-Verlag, New York 2001, pp. 43–75 MR1847786
  5. Jazwinski A. H., Stochastic Processes and Filtering Theory, Academic Press, New York 1970 Zbl0203.50101
  6. Kalman R. E., Bucy R. S., New results in linear filtering and prediction theory, J. Basic Engrg. 83 (1961), 95–108 (1961) MR0234760
  7. Kouritzin M. A., On exact filters for continuous signals with discrete observations, IEEE Trans. Automat. Control 43 (1998), 709–715 (1998) Zbl0908.93064MR1618075
  8. Kushner H. J., Budhijara A. S., A nonlinear filtering algorithm based on an approximation of the conditional distribution, IEEE Trans. Automat. Control 45 (2000), 580–585 MR1762882
  9. LeVeque R. J., Finite Volume Methods for Hyperbolic Problems, Cambridge University Press, New York 2002 Zbl1010.65040MR1925043
  10. Lototsky S. V., Rozovskii B. L., Recursive nonlinear filter for a continuous-discrete time model: Separation of parameters and observations, IEEE Trans. Automat. Control 43 (1998), 1154–1158 (1998) Zbl0957.93085MR1636479
  11. Mirkovic D., N -dimensional Finite Element Package, Technical Report. Department of Mathematics, Iowa State University 1996 
  12. Park B. T., Petrosian V., Fokker–Planck equations of stochastic acceleration: A study of numerical methods, Astrophys. J., Suppl. Ser. 103 (1996), 255–267 (1996) 
  13. Press W. H., Flannery B. P., Teukolsky S. A., Vetterling W. T., Numerical Recipes, Cambridge University Press, New York 1986 Zbl1132.65001MR0833288
  14. Risken H., The Fokker–Planck Equation, Springer–Verlag, Berlin 1984 Zbl0866.60071MR0749386
  15. Schmidt G. C., Designing nonlinear filters based on Daum’s theory, J. Guidance, Control and Dynamics 16 (1993), 371–376 (1993) Zbl0775.93283
  16. Sorenson H. W., Alspach D. L., Recursive Bayesian estimation using Gaussian sums, Automatica 7 (1971), 465–479 (1971) Zbl0219.93020MR0321581
  17. Spencer B. F., Bergman L. A., On the numerical solution of the Fokker–Planck equation for nonlinear stochastic systems, Nonlinear Dynamics 4 (1993), 357–372 (1993) 
  18. Spencer B. F., Wojtkiewicz S. F., Bergman L. A., Some experiments with massively parallel computation for Monte Carlo simulation of stochastic dynamical systems, In: Proc. Second Internat. Conference on Computational Stochastic Mechanics, Athens 1994 
  19. Šimandl M., Švácha J., Nonlinear filters for continuous-time processes, In: Proc. 5th Internat. Conference on Process Control, Kouty nad Desnou 2002 
  20. Šimandl M., Královec J., Filtering, prediction and smoothing with Gaussian sum Rrpresentation, In: Proc. Symposium on System Identification. Santa Barbara 2000 
  21. Šimandl M., Královec, J., Söderström T., Anticipative grid design in point-mass approach to nonlinear state estimation, IEEE Trans. Automat. Control 47 (2002), 699–702 MR1893533
  22. Šimandl M., Královec, J., Söderström T., Advanced point–mass method for nonlinear state estimation, Automatica 42 (2006), 1133–1145 Zbl1118.93052
  23. Šimandl M., Švácha J., Separation approach for numerical solution of the Fokker–Planck equation in estimation problem, In: Preprints of 16th IFAC World Congress. Prague 2005 
  24. Zhang D. S., Wei G. W., Kouri D. J., Hoffman D. K., Numerical method for the nonlinear Fokker–Planck equation, Amer. Physical Society 56 (1997), 1197–1206 (1997) 
  25. Zorzano M. P., Mais, H., Vazquez L., Numerical solution for Fokker–Planck equations in accelerators, Phys. D: Nonlinear Phenomena 113 (1998), 379–381 (1998) Zbl0962.82055

NotesEmbed ?

top

You must be logged in to post comments.