Niklas Hanselmann
machine learning scientist | computer vision and robotics
I am a machine learning scientist with the Scene Understanding Group at Mercedes-Benz R&D, where I work as a tech lead on both pre-development of next-gen ADAS functions as well as AI research for automated driving.
Previously, I was a PhD student with the same group, as well as the University of Tübingen and the Max Planck Institute for Intelligent Systems, where I was advised by Prof. Andreas Geiger.
Before my graduate studies, I was a master’s student at the Karlsruhe Institute of Technology, where I completed my master’s thesis at the Department of Measurement and Control, supervised by Prof. Christoph Stiller.
My research interests lie at the intersection of machine learning, robotics and computer vision. In particular, I am interested in learning robust representations for perception and planning from simulation.
news
| Apr 02, 2025 | EMPERROR, our paper on probing the robustness of learned planning using a generative model of perception errors has been accepted to IEEE Robotics and Automation Letters (RA-L). It will be presented at the International Conference on Robotics and Automation (ICRA) 2026 in Vienna. |
|---|---|
| Jun 30, 2024 | STAR-Track, our RA-L paper on latent motion models has been selected for oral presentation at IROS this year! |
| Jul 15, 2022 | I will be presenting KING at the ICML 2022 Workshop on Safe Learning for Autonomous Driving (SL4AD) next week. If you’re around, come say hi! |
| Jul 07, 2022 | KING, our paper on gradient-based generation of safety-critical driving scenarios, has been accepted to this year’s European Conference on Computer Vision (ECCV) as an oral. |
