Inversion of nonlinear stochastic models for parameter estimation
2000 (English)In: Proceedings of the IEEE Conference on Decision and Control, 2000, Vol. 2, 1591-1596 p.Conference paper (Refereed)
Prediction error and maximum likelihood estimation of nonlinear stochastic models requires inversion of the model, a step which may require substantial efforts, either in terms of manual calculations or through the use of software capable of symbolic computations. In this paper we show that model inversion can be easily implemented in numerical software such as, e.g., Simulink and Matrixx, by means of a feedback connection based on the model. We derive sufficient conditions for the existence of a stable causal inverse as well as sufficient conditions for the initial transient to decay. These conditions are given in terms of properties for a linear time-varying system associated with the nonlinear model. The method is illustrated on a numerical example.
Place, publisher, year, edition, pages
2000. Vol. 2, 1591-1596 p.
Discrete time control systems, Mathematical models, Maximum likelihood estimation, Nonlinear control systems, Probability density function, Stochastic control systems, Model inversion, Nonlinear stochastic models, Parameter estimation
Research subject SRA - ICT
IdentifiersURN: urn:nbn:se:kth:diva-28539DOI: 10.1109/CDC.2000.912087OAI: oai:DiVA.org:kth-28539DiVA: diva2:387558
39th IEEE Confernce on Decision and Control; Sydney, NSW; 12 December 2000 through 15 December 2000
QC 20110114 NR 201408042011-01-142011-01-142012-01-13Bibliographically approved