The National Science Foundation Division of Electrical, Communications and Cyber Systems awarded the University of South Alabama $304,351 on January 1, 2027, under the Engineering program (CFDA 47.041) to develop composite adaptive predictive safety via barrier-constrained optimization.
The project creates a mathematical framework that allows autonomous machines—including drones, mobile robots, and automated vehicles—to learn and adapt their internal models while remaining within certified safe operating envelopes, eliminating the current trade-off between cautious provably-safe operation and rapid learning. The technical approach couples online parameter adaptation, receding-horizon optimization, and statistical learning within a single safety certificate by merging a control barrier function (encoding safety constraints), a control Lyapunov function, and a parameter estimation error term into a composite energy function. This composite energy function serves as the terminal cost of a nonlinear model predictive controller.
Work is performed in Mobile, Alabama. The award runs through December 31, 2029. Deliverables include development of the composite adaptive predictive safety framework architecture, an open-source software toolbox for researchers and industry, graduate and undergraduate research training, and hands-on robotics workshops for high school students on the Gulf Coast.