A Particle Filter Based Method for Evaluation of Information Gap between Dynamical Systems

Full Text PDF PDF
Author(s) Zhen Zhen | Jun Young Lee | Hyun Jun Kim | Abdus Saboor
Pages 588-592
Volume 3
Issue 5
Date May, 2013
Keywords Kullback-Leibler Divergence, Dynamical System, Particle Filter, Information Theory

Abstract

This paper presents a new result on evaluating the difference between two dynamical systems. Based on the idea of information theoretic gap developed in [12], a new numerical method is developed to compute the Kullback-Leibler (K-L) rate pseudo metric between dynamics systems. Our method is based on SIR particle filter and multimodal Gaussian representation of particles is used for the computation of K-L divergence. This proposed method is relatively easy to implement and is capable of handling nonlinear systems. Numerical experiments are provided to show the efficacy of this approach.

< Back to May Issue