This research theme tackles the fundamental theoretical and practical challenges in developing data-driven modeling tools for dynamical systems starting from measured, noisy, input-output data. Deep learning tools and theory are combined with systems and control approaches to reach this goal.

Funding for this research theme is acquired through university research institutes (e.g. Eindhoven AI Systems Institute) the NWO (Dutch national research organization) and the European Union (e.g. Marie Sklodowska-Curie and ERC).

High-tech systems continuously push the boundary of what is possible in terms of accuracy, energy requirements, and system complexity and design. State-of-the-art data-driven modeling and control are of key importance in developing the required models and controllers to reach this performance increase.

Together with PhD and MSc students, we develop and implement novel system identification and feedforward/feedback control strategies over various projects in collaboration with the Eindhoven Brainport industry and beyond.

This project aims to develop a platform for the characterization of cellular system dynamics for tissue disease staging and programming for repair. System identification tools and techniques will be developed to extract the relevant cellular dynamics to obtain quantitative insight into the complex cross-relationships between biomechanical, biochemical, and mechanobiological mechanisms.

This project is coordinated by Koen Reesing from Maastricht University, The Netherlands. We are one of the consortium partners. More information can be found here, or in this interview with Koen.