Intelligent Agents for Education and Training Applications

Period of Performance: 04/02/1998 - 04/02/2000

$744K

Phase 2 SBIR

Recipient Firm

Intelligent Systems Technology, Inc.
1212 Victoria Ave
Los Angeles, CA 90066
Principal Investigator

Abstract

Training agents (i.e., computer-animated characters that assist in the learning process) are becoming central to creating affordable and effective Intelligent Tutoring Systems (ITSs). While several other required components for intelligent tutoring already exist in the research community, what is conspicuously absent in ITSs today, and ultimately has the highest payoff in terms of system affordability and instructional effectiveness, is an intelligent pedagogical agent that capitalizes on advances in autonomous agent technologies and representation of pedagogical knowledge to provide an effective substitute for a human instructor. The focus of this SBIR is the intelligent pedagogical agent for use in third party intelligent tutoring systems. The ISTI/USC team proposes to develop an architecture and authoring toolkit, called IMPART, for creating complete pedagogical agents. Using this framework, courseware developers will be able to incorporate agents into a wide range of training applications. Both high-end systems, such as distributed virtual environments, and low-end PC-based systems will be able to benefit from this technology, both for military and civilian training applications. This particular capability is central to distrubed training system affordability and effectiveness. It is also an innovation that is made possible for the first time given advances in autonomous agent technology, low-cost computer animation, and distribued object technology. The combination makes "author once, tutor anyone, anywhere, anytime" possible.