Enabling Robust Persistent Autonomy in Robots
Abstract
This project investigates the science of persistent autonomy for robotic systems, towards the goal of enhanced machine intelligence. In order to achieve persistent autonomy, a robot must be able to operate for long periods without the aid of human intervention. Persistent autonomy is characterized by the abilities to: (a) continually progress with mission objectives (sense or effect the world state); (b) apply generally-stated human preferences appropriately in specific situations; (c) adapt to novel circumstances; (d) avoid hazards; and (e) resolve failures. The challenges of persistent autonomy include limited or denied communication with humans, unknown and changing environments, hazardous conditions, and software failures. This project focuses specifically on the problems of structural adaptation to the unknown and resolution of unanticipated failures.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Feb 06, 2017
- Source ID
- FA95501710109
Entities
People
- Ross Knepper
Organizations
- Air Force Office of Scientific Research
- Cornell University
- United States Air Force