GLACIER Lab

The Geometric Learning for Autonomy, Control, Inference, and Estimation under Risk (GLACIER) Lab develops mathematical and computational foundations for intelligent autonomous systems that learn, reason, and act reliably under uncertainty. Our research integrates geometry, control and estimation, machine learning, and optimization to uncover structure in complex dynamical systems and exploit it for scalable inference and decision-making. We develop geometry-informed learning methods, reinforcement learning and policy optimization algorithms, risk-aware and distributed control strategies, and principled approaches to estimation and inference. Our goal is to build autonomous systems that are adaptive, resilient, and trustworthy, with rigorous guarantees on their behavior in uncertain and evolving environments.

This page is under construction. Please check back soon.