Seminar: "Model predictive control for uncertain systems - Robust & data-driven designs"

Abstract:

This talk addresses the design of model predictive controllers (MPC) that ensure the satisfaction of safety-critical constraints for uncertain dynamical systems. The approach leverages contraction metrics computed offline to enable efficient estimation of robust reachable sets under bounded disturbances and parametric uncertainty. A corresponding MPC formulation is provided that robustly ensures recursive feasibility, constraint satisfaction, and convergence. Generality of this framework will be demonstrated, by highlighting how the approach can also naturally address - unbounded stochastic noise - data-driven models and uncertainty quantification - online model adaptation (parametric and non-parametric)

Short Biography:

Johannes Köhler received the Ph.D. degree from the University of Stuttgart, Germany, in 2021. He is currently a postdoctoral researcher at ETH Zurich, Switzerland. His research interests are in the area of model predictive control and control and estimation for nonlinear uncertain systems. He is the recipient of the 2021 European Systems & Control PhD Thesis Award, the IEEE CSS George S. Axelby Outstanding Paper Award 2022, and the Journal of Process Control Paper Award 2023.