A knowledge-based approach for designing intelligent team training systems

To encode the team knowledge, a representation language, based on the BDI model, is developed, called MALLET, and the IARG (Inter-Agent Rule Generator) algorithm is introduced to detect information flow and generate team interactions.

Abstract

This paper presents a knowledge approach to designing team training systems using intelligent agents. We envision a computer-based training system in which teams are trained by putting them through scenarios, which allow them to practice their team skills. There are two important roles that intelligent agents can play; these are virtual team members, and tutors. To carry out these functions, these agents must be equipped with an understanding of the task domain, the team structure, the selected decision-making process and their beliefs about other team members' mental states. Even though existing agent teamwork models incorporate many of the elements listed above, they have not focused on analyzing information needs of team members to support proactive agent interactions. To encode the team knowledge, we have developed a representation language, based on the BDI model, called MALLET. A Petri Net model of an individual agent's plans and information needs can be derived from the role des...