Chapter

7th International Conference on Practical Applications of Agents and Multi-Agent Systems (PAAMS 2009)

Volume 55 of the series Advances in Intelligent and Soft Computing pp 430-439

Designing a Visual Sensor Network Using a Multi-agent Architecture

  • Federico CastanedoAffiliated withComputer Science Department Grupo de Inteligencia Artificial Aplicada (GIIA), University Carlos III of Madrid
  • , Jesús GarcíaAffiliated withComputer Science Department Grupo de Inteligencia Artificial Aplicada (GIIA), University Carlos III of Madrid
  • , Miguel A. PatricioAffiliated withComputer Science Department Grupo de Inteligencia Artificial Aplicada (GIIA), University Carlos III of Madrid
  • , José M. MolinaAffiliated withComputer Science Department Grupo de Inteligencia Artificial Aplicada (GIIA), University Carlos III of Madrid

* Final gross prices may vary according to local VAT.

Get Access

Abstract

An intelligent Visual Sensor Network (VSN) should consist of autonomous visual sensors, which exchange information with each other and have reasoning capabilities. The information exchanged must be fused and delivered to the end user as one unit. In this paper, we investigate the use of the Multi-Agent paradigm to enhance the fusion process in a VSN. A key issue in a VSN is to determine which information to exchange between nodes, what data to fuse and what information to present to the final user. These issues are investigated and reported in this paper and the benefits of an agent based VSN are also presented. The aim of the paper is to report how the multi-agent architecture contributes to solving VSNs problems. A real prototype of an intelligent VSN using the Multi-Agent paradigm has been implemented with the objective to enhance the data fusion process.