Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.652672
Title: Metaphor-based negotiation and its application in AGV movement planning
Author: Huang, X.
Awarding Body: University of Edinburgh
Current Institution: University of Edinburgh
Date of Award: 1996
Availability of Full Text:
Full text unavailable from EThOS. Please contact the current institution’s library for further details.
Abstract:
The theme of this thesis is "metaphor-based negotiation". By metaphor-based negotiation I mean a category of approaches for problem-solving in Distributed Artificial Intelligence (DAI) that mimic some aspects of human negotiation behaviour. The research in this dissertation is divided into two closely related parts. Cooperative interaction among agents in a multiagent system (MAS) is discussed in general, and the discussion leads to a formal definition of metaphor-based negotiation. Then, as a specific application, a "spring-based" computational model for metaphor-based negotiation is developed as an approach to solving movement planning, specifically the AGV scheduling problem (AGVSP) - determining the timings of AGV's activities of automated guided vehicles (AGVs) in a factory. By formally addressing the multi-agent cooperative interaction problem and assuming that agents in a MAS are rational, benevolent and fully informed, an initial strategy set of cooperative interaction can be primed to strategy set by eliminating strategies that are irrational in a group sense. However, it is proved in this dissertation that, in the remaining strategy set, no unique strategy can be found that is acceptable to all agents according their individual preferences. More specifically, in this smaller strategy set, if one agent moves from one strategy to another in an attempt to better its individual goal achievement, then there is at least one agent whose goal achievement will be negatively affected by such a move. So, the cooperative interaction problem can only be partially solved if no further knowledge is given to those agents. The idea of a common sense principle is introduced in this dissertation to overcome the deficiencies of the assumptions of rationality, benevolence and full-informedness. In reality, the assumption of full-informedness of agents may not be practical. Communication is needed for agents to (1) exchange their local problem solving information and (2) exchange proposals for global problem solving, when their views are in conflict. Based on the discussion of cooperative interaction, a formal definition of metaphor-based negotiation is proposed to formally indicate what is a proposal and what is the condition for accepting a proposal from another agent.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.652672  DOI: Not available
Share: