ACM Home Page
Please provide us with feedback. Feedback
Further extensions of FIPA Contract Net Protocol: threshold plus DoA
Full text PdfPdf (348 KB)
Source Symposium on Applied Computing archive
Proceedings of the 2004 ACM symposium on Applied computing table of contents
Nicosia, Cyprus
SESSION: Agents, interactions, mobility, and systems (AIMS) table of contents
Pages: 45 - 51  
Year of Publication: 2004
ISBN:1-58113-812-1
Authors
Chen Xueguang  Huazhong Univ. of Sci. & Tech., Wuhan, HB, P. R. China
Song Haigang  Huazhong Univ. of Sci. & Tech., Wuhan, HB, P. R. China
Sponsor
SIGAPP: ACM Special Interest Group on Applied Computing
Publisher
ACM  New York, NY, USA
Bibliometrics
Downloads (6 Weeks): 13,   Downloads (12 Months): 38,   Citation Count: 0
Additional Information:

abstract   references   index terms   collaborative colleagues  

Tools and Actions: Request Permissions Request Permissions    Review this Article  
DOI Bookmark: Use this link to bookmark this Article: http://doi.acm.org/10.1145/967900.967914
What is a DOI?

ABSTRACT

Being easily understood and implemented for resource allocation or task assignment, FIPA Contract Net Protocol (CNP) has been widely applied to kinds of Multi-Agent System (MAS). Basing on a typical extension to the original CNP, which was named as Contract Net with Confirmation Protocol (CNCP), this paper made two further extensions to CNP by the ideas of Threshold and Degree of Availability (DoA). CNCP enabled the agents to make proposals or accept offers in a continuous time, but it also increased the risk of getting sub-optimal deal or nothing for individual agent. Our research is just aiming at relieving this risk. The idea of threshold comes from the fact that: no matter how much proposals one agent sent out, it will get only one task to do at last. By setting a proper threshold, we can save many unnecessary computational costs in making or evaluating proposals. The concept of DoA is derived from the principle of Maximum Expected-Utility (MEU). With this concept, one agent will evaluate proposals not only by the required costs, it also takes the DoA of a participant as an important factor in choosing its best deal, so initiators can act in a more rational way. We also presented an under-consideration idea "Deadline", which might become another feasible way to increase system effectiveness and efficiency.


REFERENCES

Note: OCR errors may be found in this Reference List extracted from the full text article. ACM has opted to expose the complete List rather than only correct and linked references.

 
1
 
2
FIPA Contract Net Interaction Protocol Specification. FIPA TC Communication. http://www.fipa.org/specs/pesspecs.tar.gz.
 
3
FIPA Iterated Contract Net Interaction Protocol Specification. FIPA TC Communication. http://www.fipa.org/specs/pesspecs.tar.gz
 
4
M. Fisher, M. Wooldridge. Specifying and Executing Protocols for Cooperative Action. In S. M. Deen, editor, CKBS-94-Proceedings of the Second International Working Conference on Cooperating Knowledge-Based Systems. Springer-Verlag, 1994.
 
5
 
6
T. Knabe, M. Schillo, K. Fisher. Improvements to the FIPA Contract Net Protocol for Performance Increase and Cascading Applications. http://citeseer.nj.nec.com/542858.html
 
7
 
8
 
9
T. Sandholm. An Implementation of the Contract Net Protocol Based on Marginal Calculations. In Eleventh National Conference on Artificial Intelligence, pages 256--262, 1993.
 
10
T. Sandholm, V. R. Lesser. Advantages of a Leveled Commitment Contracting Protocol. In AAAI, 126--133. 1996.
 
11
T. Sandholm, Y. Zhou. Surplus Equivalence of Leveled Commitment Contracts. Artificial Intelligence, 2002.
 
12
M. Schillo, K. Fischer, C. Kray. The Contract-Net with Confirmation Protocol: A solution to a fundamental problem of DAI. DFKI Technical Memo. 15, 2000.
 
13
M. Schillo, B. Fley, and M. Florian Self-Organization in Multi-Agent Systems: From Agent Interaction to Agent Organization. http://citeseer.nj.nec.com/schillo02selforganization.html,2002.
 
14
R. G. Smith. The Contract Net Protocol: High-Level Communication and Control in a Distributed Problem Solver. IEEE Transactions on Computers, Series C-29 (12): 1104--1113, 1980.
 
15

Collaborative Colleagues:
Chen Xueguang: colleagues
Song Haigang: colleagues