Session | ||
YRM3: Dynamical Systems: Discretization, model reduction and learning
| ||
Session Abstract | ||
Dynamical systems are omnipresent in various physical processes. In view of analysis, simulation and (predictive) control, state-space models are often of central importance. To make these models accessible for computation, efficient discretization techniques and complexity reduction, e.g., by model order reduction, are often inevitable. If no state-space model is at hand, one can resort to data-based techniques to identify a surrogate model from measurements. The aim of this miniysymposium is to bring together young researchers working in the broad field of dynamical systems and their approximation via discretization, model reduction and data-based methods. In particular, we unite representatives and young experts from different communities such as operator and semigroup theory, model-order reduction, differential-algebraic equations, multi-body dynamics and optimization-based control to spark discussions and cooperations across these disciplines. | ||
Presentations | ||
4:30pm - 4:50pm
An operator-theoretic view on discretisation of random evolution equations Hamburg University of Technology 4:50pm - 5:10pm
Dynamic Control of a Soft Robot: Combining Data and Model Hamburg University of Technology 5:10pm - 5:30pm
Error bounds for Koopman-based control 1Max Planck Institute for Dynamics of Complex Technical Systems, Magdeburg; 2Paderborn University; 3TU Ilmenau 5:30pm - 5:50pm
Index aware learning of differential algebraic equations 1Eindhoven University of Technology; 2TU Darmstadt 5:50pm - 6:10pm
Model reduction on manifolds: a differential geometric framework 1University of Stuttgart; 2University of Twente, The Netherlands 6:10pm - 6:30pm
Stability and robustness in data-driven predictive control University of Stuttgart |