Title
Beyond Models: Data-Driven Prediction and Control for the Robots of Tomorrow
Abstract
Over the past decade, Model Predictive Control (MPC) has become a powerful tool in robotics, enabling applications such as walking robots, autonomous driving, and agile drone flight. Its success comes from three key strengths: it can systematically handle physical and safety constraints, it works well for complex systems, and it allows reliable and predictable behavior.
This keynote offers an introduction to modern predictive control methods, covering both traditional model-based approaches and newer data-driven techniques. We begin by explaining the basic ideas behind MPC in an intuitive way, highlighting both its advantages and the challenges that arise when applying it to real robotic systems.
We then explore recent developments in data-driven MPC, where control decisions are derived directly from measured data with little or no reliance on explicit mathematical models. These approaches open up new possibilities for adaptive and learning-based control.
Using illustrative examples — including an autonomous e-scooter platform — the talk demonstrates how these methods perform in practice, as well as their current limitations.
The talk aims to equip the audience with both conceptual understanding and practical insights into the rapidly evolving landscape of predictive control, offering guidance for future research at the intersection of models, data, and robotics.
Short CV
Frank Allgöwer is Director of the Institute for Systems Theory and Automatic Control at the University of Stuttgart, where he also holds a professorship in the Department of Mechanical Engineering. His research focuses on data-driven control, optimization-based control, and networked control systems.
He has authored more than 500 scientific publications and has received numerous honors, including the IFAC Outstanding Service Award, the IEEE Control Systems Society Distinguished Member Award, the State Teaching Award of Baden-Württemberg, and the Leibniz Prize of the Deutsche Forschungsgemeinschaft, as well as multiple Fellow distinctions from leading professional societies.
Professor Allgöwer served as President of the International Federation of Automatic Control from 2017 to 2020. He was Editor of the journal Automatica from 2001 to 2015 and is currently an editor of the Springer Lecture Notes in Control and Information Sciences series. He also served as Vice President for Technical Activities of the IEEE Control Systems Society (2013–2014) and as a member of the European Control Association Council (2001–2004). From 2012 to 2020, he was Vice President of the German Research Foundation.