COLLOQUIUM
Department of Computer Science and Engineering
University of South Carolina

Towards The Matrix: Intelligent Agents and Augmented Reality to
Enhance Human Learning Performance

M. Brian Blake

Department of Computer Science
Georgetown University

Date: January 17, 2008
Time: 1500-1600
Place: Swearingen 1A03 (Faculty Lounge)

Abstract

In the blockbuster movie, "The Matrix," characters tether themselves
to a simulated world where they live their lives virtually.  A feature
of this sort of existence is that human beings can download training
directly into their conscious. In the real world, attaching a 4 inch
wire to a person's brain may be impractical.  However, using mobile
augmented reality headsets might be the next best thing to directly
deliver training routines.  Furthermore, intelligent agents with a
notion of process management can be used to provide specialized
training to individuals wearing the headsets. Our preliminary work
focuses on allowing agents to adaptively enhance these types of
training scenarios.  Training individuals from diverse backgrounds and
in changing environments requires customized training approaches that
align with the individual learning styles and ever-evolving
organizational needs.  Scaffolding is a well-established instructional
approach that facilitates learning by incrementally removing training
aids as the learner progresses.  By combining multiple training aids
(i.e. multimodal interfaces), a trainer, either human or virtual, must
make real-time decisions about which aids to remove throughout the
training scenario. A significant problem occurs in implementing
scaffolding techniques since the speed and choice of removing training
aids must be strongly correlated to the individual traits of a
specific trainee.  We detail an agent-based infrastructure that
supports the customization of scaffolding routines as triggered by the
performance of the trainee.  This talk will describe the integration
of this agent-based approach into a simulated augmented reality (AR)
environment.  Initial experiments using the simulated environment have
evaluated the proposed adaptive approach against static training
routines.  Results show that the proposed approach increases the
trainees' task familiarity and speed with negligible insertion of
errors.

M. Brian Blake is an Associate Professor and Chair in the Department
of Computer Science at Georgetown University.  Dr. Blake conducts
applied research in the development of automated approaches for the
sharing of information and capabilities across organizational
boundaries, sometimes referred to as enterprise integration. With
respect to this area of interest, his investigations cover the
spectrum of software engineering: design, specification, proof of
correctness, implementation/experimentation, performance evaluation,
and application. He has published over 80 journal articles and
refereed conference papers in the areas of service-oriented computing,
intelligent agents and workflow, enterprise systems integration,
component-based software engineering, distributed data management, and
software engineering education. His work has been funded by the
National Science Foundation, DARPA, Federal Aviation Administration,
the MITRE Corporation, Air Force Research Lab, SAIC, and the National
Institute of Health. More information about Dr. Blake can be found at
http://www.cs.georgetown.edu/~blakeb/