Research on Design and Modeling Technique for Complicated Systems Based BDI Agents
Li YAO,Yong GONG,Chang-ying WANG,Fang LIU
DOI: https://doi.org/10.3969/j.issn.1000-1220.2005.04.032
2005-01-01
Abstract:This paper presents an approach to model Multi-Agent Systems (MASs) for design by Belief-Desire-Intention (BDI) Structure. The objective of modeling for design is to model concretely the structure of every type of agent, which takes on the roles of a Multi-Agent Organization, so as to realize the MASs in existing programming environment easily. In this modeling method, each type agent needs three models: Agent Model, Interaction Model and Acquaintance Model. Agent Model consists of three sub-models: Belief Model, Goal Model and Planning Model, which is built according to the role models and cooperative working process models obtained from agent-oriented analysis phase. Each agent in MASs is an instance of some type of agent in Agent Model. Interaction Model specifies the detail of interaction between agents, such as interactive protocols, interactive language, interactive constrains, and so on. Acquaintance Model specifies all acquaintance and their properties for each type of agent. Now through this modeling approach, an effective application system, called Multiply Agent Equipment preparing Decision Support System (MAEDSS) , has been developed on a cooperative working platform MBOS.