Using the Koopman operator, nonlinear systems can be expressed as infinite-dimensional linear systems. Data-driven methods can then be used to approximate a finite-dimensional Koopman operator, which is particularly useful for system identification, control, and state estimation tasks. However, approximating large Koopman operators is numerically challenging, leading to unstable Koopman operators being identified for otherwise stable systems. Presented are a selection of techniques to regularize the Koopman regression problem, including a novel H-infinity norm regularizer. The authors' open-source Koopman operator identification library, pykoop, is also presented (https://github.com/decarsg/pykoop).

The presentation will be given in English and the slides will be in English.

Bio

Steven Dahdah is a Ph.D. student in the department of Mechanical Engineering at McGill University. He is a member of the DECAR systems group, which, under the guidance of Prof. James Richard Forbes, conducts research in the dynamics, estimation, and control of aerospace and robotic systems. He received a B.Eng. in Electrical Engineering from McGill University in 2019. His research explores data-driven modelling and control techniques for industrial robots.

Zoom Meeting
To obtain the Zoom meeting web link, please contact:
Annette.Schwerdtfeger@gel.ulaval.ca