Active State Estimation for Nonlinear Systems: A Neural Approximation Approach

Author

Scardovi, Luca ; Baglietto, Marco ; Parisini, Thomas

Author_Institution

Univ. of Liege, Liege

Volume

18

Issue

4

fYear

2007

fDate

7/1/2007 12:00:00 AM

Firstpage

1172

Lastpage

1184

Abstract

In this paper, we consider the problem of actively providing an estimate of the state of a stochastic dynamic system over a (possibly long) finite time horizon. The active estimation problem (AEP) is formulated as a stochastic optimal control one, in which the minimization of a suitable uncertainty measure is carried out. Toward this end, the use of the Renyi entropy as an information measure is proposed and motivated. A neural control scheme, based on the application of the extended Ritz method (ERIM) and on the use of a Gaussian sum filter (GSF), is then presented. Simulation results show the effectiveness of the proposed approach.

Keywords

Gaussian processes; entropy; infinite horizon; neurocontrollers; nonlinear programming; nonlinear systems; optimal control; probability; state estimation; stochastic systems; uncertain systems; Gaussian sum filter; Renyi entropy; active state estimation; conditional probability function; extended Ritz method; finite time horizon; information measure; neural approximation; neural control; nonlinear programming; nonlinear systems; stochastic dynamic system; stochastic optimal control; uncertainty measure minimization; Control systems; Covariance matrix; Entropy; Machine learning; Measurement uncertainty; Nonlinear systems; Optimal control; State estimation; Stochastic processes; Stochastic systems; Active estimation; entropy; neural networks (NNs); Algorithms; Computer Simulation; Decision Support Techniques; Feedback; Models, Theoretical; Neural Networks (Computer); Nonlinear Dynamics;

fLanguage

English

Journal_Title

Neural Networks, IEEE Transactions on

Publisher

ieee

ISSN

1045-9227

Type

jour

DOI

10.1109/TNN.2007.899251

Filename

4267718