01/2020 - present: PhD student at the Chair of Information-oriented Control, Faculty of Electrical Engineering and Information Technology, Technical University of Munich, Germany
10/2017 - 12/2019: M.Sc., Electrical Engineering and Information Technology, Technical University of Munich, Germany
09/2014 - 07/2017: B.Sc., Electrical Engineering and Information Technology, University of Zagreb, Croatia
Research Interests
Learning efficacious representations of dynamics
Koopman operator paradigm
Data-driven control in constrained and uncertain environments
Underwater robot Tortuga courtesy of Subsea Tech
https://subsea-tech.com
The oceans today contain tons of waste, majority of which is underwater. While effort have been put into collecting litter on the surface, so far there have been few attempts to collect litter on the ocean floor. In addition, such activities are mostly dependent on human divers.
As a part of the research team at ITR and in collaboration with partners across Europe, we are working on strategies for the automated collection of underwater litter as part of the SeaClear project. To this end, we develop concepts for autonomous robots that support us in this task.
Interesting challenges for our team include:
Hardware: Suitable mechanical grippers must be developed for collecting waste.
Robotics and Control: The modeling and control of the robots in an underwater environment as well as the cooperation with humans requires new approaches.
Artificial iIntelligence: Poor visibility in the underwater environment along with many unknown disturbances, lead to interesting possibilities for the use of artificial intelligence.
Working Field
Learning for control in H2020 project "SEarch, identificAtion and Collection of marine Litter with Autonomous Robots“ (SeaClear)
Student Theses
I am in constant search for motivated students interested in my research. Feel free to contact me if you are interested in a thesis, even if no topics are currently listed down below. Please let me know when you plan to start,attach your CV and Grade Transcript for suitably adjusted work content.
P. Bevanda, M. Beier, A. Lederer, S. Sosnowski, E. Hüllermeier, S. Hirche: Koopman Kernel Regression. Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS 2023) (NeurIPS Proceedings), 2023 mehr…BibTeX
Volltext (mediaTUM)
2022
P. Bevanda; J. Kirmayr; S. Sosnowski; S. Hirche: Learning the Koopman Eigendecomposition: A Diffeomorphic Approach. 2022 American Control Conference (ACC), IEEE, 2022 mehr…BibTeX
Volltext (
DOI
)
Volltext (mediaTUM)
P. Bevanda; M. Beier; S. Kerz; A. Lederer; S. Sosnowski; S. Hirche: Diffeomorphically Learning Stable Koopman Operators. IEEE Control Systems Letters (L-CSS) 6, 2022, 3427 - 3432 mehr…BibTeX
Volltext (
DOI
)
Volltext (mediaTUM)
