Autonomous Intelligent Agents for Team Training

Christopher Myers, Jerry Ball, Nancy Cooke, Mary Freiman, Michelle Caisse, Stuart Rodgers, Mustafa Demir, Nathan McNeese

Research output: Contribution to journalArticle

Abstract

The rise in autonomous system research and development combined with the maturation of computational cognitive architectures holds the promise of high-cognitive-fidelity agents capable of operating as team members for training. We report an ACT-R model capable of operating as a team member within a remotely piloted aerial system, and provide results from a first-of-its-kind controlled, randomized empirical evaluation in which teams that worked with an AST were compared against all-human teams. Our results demonstrate that ASTs can be incorporated into human teams, providing training opportunities when teammates are unavailable. We conclude with issues faced in developing ASTs and lessons learned for future and current developers.

Original languageEnglish (US)
Article number8574973
Pages (from-to)3-14
Number of pages12
JournalIEEE Intelligent Systems
Volume34
Issue number2
DOIs
StatePublished - Mar 1 2019

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Artificial Intelligence

Fingerprint Dive into the research topics of 'Autonomous Intelligent Agents for Team Training'. Together they form a unique fingerprint.

  • Cite this

    Myers, C., Ball, J., Cooke, N., Freiman, M., Caisse, M., Rodgers, S., Demir, M., & McNeese, N. (2019). Autonomous Intelligent Agents for Team Training. IEEE Intelligent Systems, 34(2), 3-14. [8574973]. https://doi.org/10.1109/MIS.2018.2886670