Abstract
This talk presents a unified framework for the data-driven design of fixed-structure controllers using only frequency response data of the plant. It begins with a general H-two/H-infinity synthesis method formulated as a convex optimization problem with Linear Matrix Inequalities (LMIs), showing that performance and stability objectives can be achieved without requiring a parametric model. The approach supported general controller structures in both continuous and discrete time.
Building on this foundation, the framework is extended to handle nonlinear and uncertain systems using the Integral Quadratic Constraint (IQC) formalism, enabling robust controller synthesis with both parametric and nonparametric multipliers. As a practical application, the method is used to reject sinusoidal disturbances with unknown and time-varying frequencies in a micro-vibration isolation platform for space applications. Finally, a data-driven technique was introduced for learning IQC multipliers directly from input/output trajectories, capturing nonlinear model mismatch and providing probabilistic robustness guarantees through scenario optimization. This perspective enables a scalable and model-free approach to robust control design for complex dynamical systems.
Biographical Information
Alireza Karimi received his Ph.D. degree from the National Polytechnic Institute of Grenoble, France, in 1997. From 1998 to 2000, he was an Assistant Professor in the Department of Electrical Engineering at Sharif University of Technology, Tehran, Iran. He then joined EPFL, Switzerland, where he is currently a Professor of Automatic Control and Head of the Data-Driven Modelling and Control Group within the Mechanical Engineering Institute. His research focuses on robust data-driven controller design for power grids and mechatronic systems. Prof. Karimi served as an Associate Editor of the European Journal of Control from 2004 to 2013 and is currently an Associate Editor of the IEEE Transactions on Control Systems Technology. Since 2018, he has been an active member of the Conference Editorial Board of the IEEE Control Systems Society.