Home > Research > Publications & Outputs > Stellar

Electronic data

Links

Text available via DOI:

View graph of relations

Stellar: A Programming Model for Developing Protocol-Compliant Agents

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Published
Publication date14/07/2019
Host publicationEngineering Multi-Agent Systems - 6th International Workshop, EMAS 2018, Revised Selected Papers: 6th International Workshop, EMAS 2018, Stockholm, Sweden, July 14-15, 2018, Revised Selected Papers
EditorsViviana Mascardi, Danny Weyns, Alessandro Ricci
PublisherSpringer
Pages117-136
Number of pages20
ISBN (print)9783030256920
<mark>Original language</mark>English

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer
Volume11375

Abstract

An interaction protocol captures the rules of encounter in a multiagent system. Development of agents that comply with protocols is a central challenge of multiagent systems. Our contribution in this chapter is a programming model, Stellar, that simplifies development of agents compliant with information protocols specified in BSPL. A significant distinction of Stellar from similar approaches is that it does not rely upon extracting control flow structures from protocol specifications to ensure compliance. Instead, Stellar provides a set of fundamental operations to programmers for producing viable messages according to the correct flow of information between agents as specified by a protocol, enabling flexible design and implementation of protocol-compliant agents. Our main contributions are: (1) identification of a set of programming errors that commonly occur when developing agents for protocol-based multiagent system, (2) definition of Stellar’s operations and a simple yet effective pattern to develop protocol-compliant agents that avoid the identified errors, and (3) demonstration of Stellar’s effectiveness by presenting concrete agents in e-commerce and insurance policy domains.