Agent-based architectural framework enhancing configurability, autonomy and scalability of context-aware pervasive services

  • Authors:

  • Nikolaos Dimakis;John Soldatos;Lazaros Polymenakos;Axel Bürkle;Uwe Pfirrmann;Gerhard Sutschet

  • Affiliations:

  • Intelligent Systems and Networks Group, Electrical and Electronic Engineering Department, Imperial College, London, UK SW7 2BT;Athens Information Technology, Peania, Greece 19002;Athens Information Technology, Peania, Greece 19002;Fraunhofer Institute for Information and Data Processing, Karlsruhe, Germany 76131;Fraunhofer Institute for Information and Data Processing, Karlsruhe, Germany 76131;Fraunhofer Institute for Information and Data Processing, Karlsruhe, Germany 76131

  • Venue:
  • Autonomous Agents and Multi-Agent Systems
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Multi-agent software architectures have gained in popularity due to their beneficial behavior in designing and implementing sophisticated applications. However, current approaches in implementing such architectures have led to application-specific, non-scalable implementations which limit the reusablity and improvement of the whole architecture. Moreover, these attempts lack features to enhance the user experience, thus slowing the adoption of the resulting services. In this paper we describe a fully-fledged multi-agent architecture covering a large variety of preferred features including capabilities of `plugging' ubiquitous services, servicing mobile users, interconnecting remote similar architectures and interfacing with advanced software components such as knowledge bases. This framework exploits a wide-range of context-aware components making it essentially context-aware, allowing for the development of ubiquitous context-aware human-centric services, which are the focus of our research. To illustrate the flexibility of this architectural framework, we present four services which were built using this architectural paradigm by different development teams and elaborate on their overall behavior.