Engineering Self-Organising Systems
The Fourth International Workshop on Engineering Self-Organizing Appli- tions (ESOA) was held on May 9, 2006 in conjunction with the 2006 Conference on Autonomous Agents and Multi-Agent Systems (AAMAS 2006),in Hakodate, Japan. The present post-proceedings volume contains revised versions of the seven papers presented at the workshop, and six additional invited papers. C- tinuing the tradition of previous editions, this book discusses a broad variety of topics in an e?ort to allow room for new ideas and discussion, and eventually a better understanding of the important directions and techniques of our ?eld. In “Hybrid Multi-Agent Systems: Integrating Swarming and BDI Agents”— an article based on an invited talk at the workshop by Van Parunak—Parunak et al. address an important question facing the ESOA community: how should self-organizingswarm-likeagentapproachesrelatetothetechniquesofthemul- agent community at large? ESOA techniques primarily rely on simple reactive agents, whose intelligence emerges at the group level via carefully designed - teraction rules. These simple agents might have some internal state that allows them to remember the history of their interactions at some (low) level of detail, but generally the complexity in such systems arises from the dynamics. In c- trast, the mainstream multi-agent systems community uses intelligent agents, which apply sophisticated algorithms to build up internal models of their - vironments and complex protocols to communicate about their models. This general approach,of which the BDI frameworksare an example, warranta more cognitive analogythan the typical ESOAideas. Parunaket al. ’s workshowshow the two approaches could pro?tably interact.