A reference architecture for situated multiagent systems

17Citations
Citations of this article
38Readers
Mendeley users who have this article in their library.
Get full text

Abstract

A reference architecture integrates a set of architectural patterns that have proven their value for a family of applications. Such family of applications is characterized by specific functionality and quality requirements. A reference architecture provides a blueprint for developing software architectures for applications that share that common base. As such, a reference architecture provides a means for large-scale reuse of architectural design. This paper gives an overview of a reference architecture for situated multiagent systems we have developed in our research. We discuss various architectural views of the reference architecture. Per view, we zoom in on the main view packets, each of them containing a bundle of information of a part of the reference architecture. For each view packet we explain the rationale for the design choices that were made and we give built-in mechanisms that describe how the view packet can be exercised to build a concrete software architecture. We illustrate the use of the reference architecture with an excerpt of the software architecture of an industrial AGV transportation system. © Springer-Verlag Berlin Heidelberg 2007.

Cite

CITATION STYLE

APA

Weyns, D., & Holvoet, T. (2007). A reference architecture for situated multiagent systems. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4389 LNAI, pp. 1–40). Springer Verlag. https://doi.org/10.1007/978-3-540-71103-2_1

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free