Efficient grid-based Bayesian estimation of nonlinear low-dimensional systems with sparse non-Gaussian PDFs

DOI10.1016/J.AUTOMATICA.2012.02.039MaRDI QIDQ445948zbMATHOpenAlexWikidataFDO

Authors Thomas R. Bewley, Atul S. Sharma

Publication date 27 August 2012

Published in Automatica (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1301.4866

zbMATH Keywords

grid-based Bayesian estimationnonlinear/non-Gaussian observer design

Mathematics Subject Classification ID

Bayesian inference (62F15) Estimation and detection in stochastic control theory (93E10)

Abstract: Bayesian estimation strategies represent the most fundamental formulation of the state estimation problem available, and apply readily to nonlinear systems with non-Gaussian uncertainties. The present paper introduces a novel method for implementing grid-based Bayesian estimation which largely sidesteps the severe computational expense that has prevented the widespread use of such methods. The method represents the evolution of the probability density function (PDF) in phase space, $p_x(x^{\prime },t)$, discretized on a fixed Cartesian grid over {it all} of phase space, and consists of two main steps: (i) Between measurement times, $p_x(x^{\prime },t)$ is evolved via numerical discretization of the Kolmogorov forward equation, using a Godunov method with second-order corner transport upwind correction and a total variation diminishing flux limiter; (ii) at measurement times, $p_x(x^{\prime },t)$ is updated via Bayes' theorem. Computational economy is achieved by exploiting the localised nature of $p_x(x^{\prime },t)$. An ordered list of cells with non-negligible probability, as well as their immediate neighbours, is created and updated, and the PDF evolution is tracked {it only} on these active cells. %The grid-based discretization of $p_x(x^{\prime },t)$ in this approach avoids the requirement for resampling associated with particle-based representations of the PDF.

Recommendations

• Bayesian filtering estimation approach based on finite element method
• Estimation of dynamic systems using a method of characteristics filter
• Grid-based nonlinear estimation and its applications
• Point-mass filter: density specific grid design and implementation
• Advanced point-mass method for nonlinear state estimation

Cites work

• An essay towards solving a problem in the doctrine of chances. By the late Rev. Mr. Bayes, F. R. S. communicated by Mr. Price, in a letter to John Canton, A. M. F. R. S.
• Finite Volume Methods for Hyperbolic Problems
• On Information and Sufficiency
• Rate of Convegence of a Stochastic Particle Method for the Kolmogorov Equation with Variable Coefficients
• Recursive Bayesian estimation using piecewise constant approximations
• Stochastic processes and filtering theory

Cited in

(6)

• Point-mass filter: density specific grid design and implementation
• Sparse Bayesian Nonlinear System Identification Using Variational Inference
• Advanced point-mass method for nonlinear state estimation
• Optimal Transport for Parameter Identification of Chaotic Dynamics via Invariant Measures
• Bayesian filtering estimation approach based on finite element method
• Grid methods for Bayes-optimal continuous-discrete filtering and utilizing a functional tensor train representation

This page was built for publication: Efficient grid-based Bayesian estimation of nonlinear low-dimensional systems with sparse non-Gaussian PDFs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q445948)