This $500,000 project grant, awarded on January 1, 2024 by the National Science Foundation (NSF) under the Computer and Information Science and Engineering (CISE) program (CFDA 47.070), aims to address the urgent need for end-to-end safety in learning-enabled autonomous systems across various application scenarios, such as self-driving cars and urban air mobility.
The project, titled "COLLABORATIVE RESEARCH: SLES: GUARANTEED TUBES FOR SAFE LEARNING ACROSS AUTONOMY ARCHITECTURES," proposes a novel solution called "Data-Enabled Simplex" or "DESIMPLEX" that is built on solid mathematical principles and systematic methods for data collection and utilization to improve system performance and ensure safety, even in the face of extreme events or environmental hazards. The research will focus on (i) improving high-performance autonomy with reliable uncertainty quantification methods, and (ii) developing high-assurance autonomy architectures and switching rules for verifiable observability and controllability. The proposed framework will be validated through modular simulation testing and integration and deployment on real aerial and ground vehicles.