This five-year National Science Foundation Project Grant of $1,238,246 will fund research at Carnegie Mellon University toward developing safe autonomous systems. The goal is to enable robots and other autonomous systems to operate safely while interacting closely with humans, as required for applications like next-generation manufacturing infrastructure. The university will develop a new algorithmic framework for assuring safety in autonomous robotic systems. The framework aims to optimize performance when safety can be managed, anticipate and compensate for inevitable failures when it cannot, and learn from past mistakes. It is intended to increase trust in autonomous systems while minimizing human effort for deployment and maintenance. The research also aims to contribute to cross-task safeguard approaches that augment hardware without manual tuning, monitor and optimize nominal task-oriented controls to satisfy safety requirements, and accomplish these objectives under time-varying uncertainty. Evaluation will utilize simulation and experiments with autonomous vehicles interacting with human-driven vehicles and robotic arm manipulators sharing space with humans and other robots. Funded under the NSF's Engineering program (CFDA 47.041), this award supports the program's goals of fostering innovation and excellence in engineering research.
Mod # | Description | Reason For Modification | Federal Obligation (Click to sort descending) | Date (Click to sort ascending) |
|---|---|---|---|---|
| Not listed | $494.1k | 6/3/22 | ||
| Not listed | $250.0k | 1/25/22 |