Interview: Cornelia Freund
ED: Congratulations on your professorship. What academic path did you take?
Prof. Paul Kotyczka: Thank you very much! After studying electrical engineering and information technology at TUM, I was looking for a doctoral position. After a brief stopover at the Institute for Machine Tools and Industrial Management, I moved to the then newly established Chair of Automatic Control under Prof. Boris Lohmann on January 1, 2006. There I was able to develop many new ideas as a researcher.
I established my “Energy-Based Modeling and Control” working group with my doctorate in 2010. From 2015 to 2017, I was a postdoc at the University of Lyon thanks to a Marie Skłodowska-Curie fellowship - a valuable time that had a strong influence on my professional career. In 2019, I habilitated and on April 22, 2025, TUM appointed me as an adjunct professor.
ED: What sparked your interest in Port-Hamiltonian systems and what fascinates you about them?
Prof. Paul Kotyczka: As a doctoral candidate, I was able to choose from many current topics in control theory. I was particularly fascinated by the idea of using energy to design complex control systems. Among other things, the publication “Putting Energy Back in Control” from 2001 inspired me greatly. Over time, new areas of focus were added alongside control design, such as structure-preserving numerical methods or issues from process engineering or robotics.
It is interesting to describe different physical systems in a standardized, mathematical formulation and to use these structures for control tasks. And what excites me personally is that this field has taught me a lot about modeling different classes of systems. After my studies, I would never have thought that I would be dealing with heat transfer in metallic foams or non-linear continuum mechanics. The fact that I can constantly learn new things, together with students, doctoral candidates and colleagues, is a privilege and gives me great joy.
ED: For those who are not familiar with the subject: Could you explain what Port-Hamiltonian systems are?
Prof. Paul Kotyczka: Port-Hamiltonian systems are mathematical models that combine two important and elegant concepts from classical mechanics and electrical engineering: The structure of the differential equations maps exchanges and transformations of the total system energy - the Hamiltonian. And just as in network theory, models of any complexity can be modularly assembled from subsystems of different physical nature via the power interfaces - the ports.
ED: Why are Port-Hamiltonian systems important in various scientific fields such as mathematics and engineering?
Prof. Paul Kotyczka: Due to their modularity, they are initially interesting for modeling and simulating complex, coupled systems. At the same time, numerical methods are required to keep the models consistent with regard to the energy flows between the subsystems. This consistency is achieved by choosing the right variables and discretizing them correctly. This avoids the occurrence of unphysical numerical effects. As mentioned at the beginning, the Port-Hamiltonian system description also inspires energy-based control designs.
ED: Can you give examples of practical applications of Port-Hamiltonian systems?
Prof. Paul Kotyczka: The approach is extremely versatile, especially for modeling coupled systems such as energy grids. For the adaptive lightweight high-rise building at the University of Stuttgart with a hydraulically actuated support structure, for example, a Port-Hamiltonian model was used for energy-optimized control. In soft robotics and medical technology, the couplings in electro-active polymers, for example for endoscopes, can be modeled. Control models for the plasma in Tokamak reactors such as the ITER in southern France are impressive. The Institut de Recherche et de Coordination Acoustique/Musique (IRCAM) in Paris uses Port-Hamiltonian models to digitally recreate the sound of old electronic instruments. One amusing (commercial!) application is software that - based on physical models - plays music as if it came from a cassette tape from the 1980s. Our cable-actuated continuum manipulator can also be modeled in a Port-Hamiltonian way, dual to the Lagrangian approach.
ED: What are the current challenges and how do you see developments in the field of Port-Hamiltonian systems in the coming years?
Prof. Paul Kotyczka: Our community brings together researchers from the fields of systems theory, automatic control and applied mathematics. There are great initiatives, such as our doctoral program, which is now entering its second funding phase [see info below] and a fresh collaborative research center at the University of Wuppertal. There is a lot of expertise in energy-based methods in the engineering sciences, without the “Port-Hamiltonian” label. There is also powerful software, both commercial and open source. We should make better use of these resources in future to tackle increasingly complex issues. As always, it is worth thinking outside the box.
ED: Are there any future projects that you are particularly looking forward to?
Prof. Paul Kotyczka: Yes, I am very much looking forward to a project with Prof. Peter Betsch from the Karlsruhe Institute of Technology, which deals with modeling and control for systems with irreversible processes.
A new area of research for me revolves around renewable energy grids and their control. Last year, I started a cooperation with Prof. Anca Hansen and Prof. Kaushik Das from the Department of Wind and Energy Systems at DTU. I am also in contact with my TUM colleague Dr. Anurag Mohapatra from the Chair of Renewable and Sustainable Energy Systems - the CoSES laboratory offers many exciting questions and application possibilities for control problems in the energy transition. On the one hand, this is thematically a “return to the roots”, on the other hand I'm learning new things again - a nice perspective.
Further Information:
Franco-German Doctoral College Port-Hamiltonian Systems
The doctoral college funded by the Franco-German University is aimed at doctoral candidates from Germany, France and the partner countries of the Netherlands and Austria who are working on structured, energy-based methods for modeling, numerical approximation and control of multiphysical systems. It promotes cooperation between research groups in the participating countries, provides an institutional framework for structured training and supports the mobility of young scientists. The biannual workshops are open to all interested individuals.
Spring School on Theory and Applications of Port-Hamiltonian Systems
At the international spring school, which is organized in cooperation with the Bavarian-French University Center, doctoral candidates, postdocs and advanced master students from the engineering sciences and applied mathematics have been given an introduction to Port-Hamiltonian systems as well as current research topics and applications since 2019. The next spring school will take place in 2028.