We introduce a simple language for multi-agent systems that lends itself to intuitive design of local specifications. Agents operate on (parts of) a decentralized data structure, the stigmergy, that contains their (partial) knowledge. Such knowledge is asynchronously propagated across local stigmergies. In this way, local changes may influence global behavior. The main novelty is that our interaction mechanism combines stigmergic interaction with attribute-based communication. Specific conditions for interaction can be expressed in the form of predicates over exposed features of the agents. Additionally, agents may access a global environment. After presenting the language, we show its expressiveness by considering some illustrative case studies. We also include preliminary results towards automated verification via a mechanizable symbolic encoding that enables us to exploit verification tools developed for mainstream languages

Multi-agent systems with virtual stigmergy

Inverso O
2020

Abstract

We introduce a simple language for multi-agent systems that lends itself to intuitive design of local specifications. Agents operate on (parts of) a decentralized data structure, the stigmergy, that contains their (partial) knowledge. Such knowledge is asynchronously propagated across local stigmergies. In this way, local changes may influence global behavior. The main novelty is that our interaction mechanism combines stigmergic interaction with attribute-based communication. Specific conditions for interaction can be expressed in the form of predicates over exposed features of the agents. Additionally, agents may access a global environment. After presenting the language, we show its expressiveness by considering some illustrative case studies. We also include preliminary results towards automated verification via a mechanizable symbolic encoding that enables us to exploit verification tools developed for mainstream languages
Multi-agent systems, Stigmergic interaction, Emergent behavior, Attribute-based communication, Agent-based modeling
File in questo prodotto:
File Dimensione Formato  
2020_SciCoProg_187_denicola.pdf

non disponibili

Tipologia: Versione Editoriale (PDF)
Licenza: Non pubblico
Dimensione 666.76 kB
Formato Adobe PDF
666.76 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12571/6951
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 13
  • ???jsp.display-item.citation.isi??? ND
social impact