Research On An Agent Model With Emotional Factor And Simulation Realization

An Agent is a software or hardware system with autonomy, reactivity, social abilityand proactivity. Studying the Agent provides an effective way for solving newdistributed application problems, provides reasonable concept models for studyingdistributed computing systems comprehensively and accurately. However, most of theexisting Agent structural models focus on logical reasoning abilities, whereasconsideration of the influence of emotions in the Agent models as well asdecision-making behaviors relatively less, thus needs for further study.In order to explore the emotions’ role in Agent architectural model as well asinfluences on the Agent’s reasoning and decision-making behavior, and address theissues of lacking of clear, reasonable semantic description of the emotional factor aswell as restraint and motivated relations between the emotional operator and otherinternal operators of the Agent model, an Agent Conceptual Model with Emotionalfactors (EACM) is proposed, which is on the basis of traditional Agent architecturemodels and the related studies of emotional Agents. This model differs from most ofthe existing emotional Agents just focusing on the role of emotions under thedeliberative mechanism or reactive mechanisms, it takes into consideration bothmechanisms under the condition of emotional effect and thus makes it more close tohuman intelligence including reasoning and decision-making behavioral reactions,which thus improves its adaptation and decision-making ability in complex anddynamic environment.The paper firstly analyzes and summarizes the related researches and existingproblems of Agent technology, affective computing and emotional Agents, thus pointsout the research background and significance of carrying out this article, andmeanwhile introduces the basic theories of Agent structural models, Agent formaldescription and emotional models.Secondly, designing and constructing an Agent Conceptual Model with Emotionalfactor (EACM). The formal description process and the emotional decision-makingprocess of EACM are designed and elaborated; an improved cognitive evaluationmethod based on OCC model is proposed, the emotional updating mechanism and thereasoning decision-making process of EACM from the perspective of the fuzzy reasoning are elaborated; the mapping implementation process between EACMconceptual model and the Agent modeling and simulation platform-Netlogo ispresented.Finally, from the perspective of simulation implementation and practicalapplication, a multi-goal simulation scenario of salesmen working at the marketing ofproducts based on the EACM model is designed and constructed. Utilizing theAgent-based modeling and simulation platform-Netlogo to model and simulate thesalesman Agent with the EACM conceptual model, and carrying out the comparativestudy with the ordinary salesman Agent. Simulation results show that the EACMconceptual model constructed in this paper is reasonable and effective, which canachieve the interaction between the emotional operator and other internal operators ofthe model, improve the reasoning and decision-making, and has certain advantagescompared with the ordinary Agent.