Von: Zentralbibliothek [bibliothek@gmd.de]
Gesendet: Montag, 2. Oktober 2000 10:39
An: deutscher@wotan.cs.uni-magdeburg.de
Betreff: Entropy-Based Markov Chains for Multisensor Fusion
Journals Table of Contents
Kluwer Homepage Customer Service Webmaster Journal Homepage Subscribe to this Journal Request Sample Copy Search Subject Related Journals Give us your comments Journal of Intelligent and Robotic Systems
29 (2):161-189, October 2000.
© Kluwer Academic Publishers

Entropy-Based Markov Chains for Multisensor Fusion

A. C. S. Chung
Department of Computer Science, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong; e-mail: helens@cs.ust.hk

H. C. Shen
Department of Computer Science, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong; e-mail: helens@cs.ust.hk

Abstract
This paper proposes an entropy based Markov chain (EMC) fusion technique and demonstrates its applications in multisensor fusion. Self-entropy and conditional entropy, which measure how uncertain a sensor is about its own observation and joint observations respectively, are adopted. We use Markov chain as an observation combination process because of two major reasons: (a) the consensus output is a linear combination of the weighted local observations; and (b) the weight is the transition probability assigned by one sensor to another sensor. Experimental results show that the proposed approach can reduce the measurement uncertainty by aggregating multiple observations. The major benefits of this approach are: (a) single observation distributions and joint observation distributions between any two sensors are represented in polynomial form; (b) the consensus output is the linear combination of the weighted observations; and (c) the approach suppresses noisy and unreliable observations in the combination process.

Keywords
decision making, entropy, Markov chains, multisensor fusion, uncertainty

Full Text PDF Article Download Free Adobe Acrobat 
Reader

ISSN 0921-0296