Soar Technology's key research into large-scale agents with human-level intelligence began a decade ago in conjunction with the University of Michigan's Artificial Intelligence Laboratory with support from the Defense Advanced Research Projects Agency (DARPA). This work developing training solutions for the Air Force resulted in TacAir-Soar, the most highly developed model of tactical fighter pilots in action ever created. Our work has since branched beyond pilot training applications to include several new and distinct lines of research:
Extending the philosophy of Tac-Air Soar, these highly capable, autonomous
agents enable human-level reasoning, knowledge, and interaction in both
simulated and operational environments.
[ Read More ... ]
Groups of knowledge-rich intelligent agents, each with unique specialties
and capabilities are combined in well-defined and dynamic problem-solving
teams for adaptive user interfaces and automated workflow applications.
[ Read More ... ]
Collections of intelligent agents are used to model cultural,
social, and other aspects of human or other organizations to
reveal and predict emergent properties of inter-organization
and cross-organization relationships.
[ Read More ... ]
We pursue informed research into human behavior, performance,
and error, and cognitive architectures. Of particular interest
are two areas: the interaction between reasoning, knowledge and
learning; and the tools and technologies that will enable better,
faster, as well as cheaper, engineering knowledge-rich intelligent
agent systems.
[ Read More ... ]
©2008 Soar Technology, Inc. All rights reserved
Legal Notice | Site Map
- Research Scientists