Project Grant 2534263

Washington State University received a $225,000 Project Grant award from the National Science Foundation's Directorate for Engineering (CFDA 47.041) effective July 1, 2026, with completion targeted for June 30, 2029. The award supports collaborative research on reinforcement learning (RL) with high-probability safety constraints, addressing the critical gap between current learning-enabled systems that optimize average performance and the stringent safety requirements of real-world cyber-physical systems such as unmanned aerial vehicles, mobile robots, and networked infrastructures. The project will develop new theoretical foundations and algorithms for high-confidence autonomy that explicitly enforce safety as probabilistic guarantees rather than average-case constraints, enabling safe decision-making under model uncertainty, limited data, and changing operating conditions.

The deliverables include peer-reviewed research advancing RL theory and methodology for safety-critical applications, open-source software tools and benchmark scenarios to support broader adoption of safe autonomy research, and educational curriculum modules and student research experiences that enhance participation in science, technology, engineering, and mathematics (STEM) disciplines. The work addresses national priorities in reliable autonomous systems relevant to defense, infrastructure security, and public welfare, with research performance conducted at Washington State University's Pullman, Washington campus.