Toward Undifferentiated Cognitive Agents: Translating Instructions to Knowledge
Abstract
Autonomous systems are a new frontier for pushing sociotechnical advancement. Such systems will eventually become pervasive, involved in everything from manufacturing, healthcare, defense, and even research itself. However, proliferation is stifled by the high development costs and the resulting inflexibility of the produced systems. The current time needed to create and integrate state of the art autonomous systems that operate as team members in complex situations is a 3-15 year development period, often requiring humans to adapt to limitations in the resulting systems. A new research thrust in interactive task learning (ITL; Laird, Gluck, et al., 2017) has begun, calling for natural human-autonomy interaction to facilitate system flexibility and minimize users’ complexity in providing autonomous systems with new tasks.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jul 11, 2018
- Source ID
- FA95501810371
Entities
People
- Dario D Salvucci
Organizations
- Air Force Office of Scientific Research
- Drexel University
- United States Air Force