Abstract
Our analysis of the KQML (Knowledge Query and Manipulation Language) model yielded some conclusions on the knowledge level of communication in agent-oriented program. First, the agent state and transition model were given for analyzing the necessary conditions for interaction with the synchronal and asynchronous KQML model respectively. Second, we analyzed the deadlock and starvation problems in the KQML communication, and gave the solution. At last, the advantages and disadvantages of the synchronal and asynchronous KQML model were listed respectively, and the choosing principle was given.
Similar content being viewed by others
References
Angela Dalmonte, Mauro Gaspari, 1995. Modelling interaction in agent system, technical report UBLCS-95-7, laboratory for computer science, University of Bologna.
Mauro Gapari, Enrico Motta, 1994. Symbol-level Pequirment for agent-level rrogramming, technical report UBLCS-94-2, Laboratory for Computer Science, University of Bologna.
Michael, R. Genesereth, Steven, P. Ketchpel, 1994. Software Agents.Communication of ACM,37: 48–54.
Newell, A. 1982. The knowledge level.Artificial Intelligence,18: 87–127.
Shoham, Y., 1993. Agent-oriented programming.Artificial Intelligence,60: 51–92.
Stefan, K., 1994. STRICT: A blackboard based tool supporting the design of distributed PPC application.Expert Systems With Applications,7(1): 10–20.
Thomas, R. Gruber, 1992. A translation approach to portable ontology specifications. Computer Science Department, Stanford University Techn Report KSL 92-71.
Author information
Authors and Affiliations
Additional information
Project supported by the National High-Tech R&D Plan (9845-005)
Rights and permissions
About this article
Cite this article
Hai-long, L., Tie-jun, W. Analysis of the KQML model in multi-agent interaction. J. Zhejiang Univ. Sci. A 2, 132–136 (2001). https://doi.org/10.1631/jzus.2001.0132
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1631/jzus.2001.0132