Biosketch
Tzu-Yuan Huang received his M.Sc. degree in Electrical Engineering with a specialization in Motion planning & System Identification for Robot manipulators from National Cheng Kung University (NCKU), Taiwan in 2020. Following his graduation, he joined Syntec Technology as an R&D engineer from 2020 to 2023, where he developed trajectory planning algorithms of five-axis machine tool and robot using C++ and python.
Currently, Tzu-Yuan Huang is pursuing his Ph.D. degree at the Chair of Information-Oriented Control (ITR) at Technical University of Munich. He is also a member of the Konrad Zuse School of Excellence in Reliable Artificial Intelligence (relAI). His research interest is data-driven control in robotic system, and he contributes to the SeaClear project, which focuses on automating the collection of underwater waste.
relAI Research
Safe and Reliable Robot Foundation Models
Robot foundation models are emerging as a promising paradigm for building general-purpose robotic intelligence. By learning from large-scale, multimodal data, these models aim to generate flexible behaviors across diverse tasks, environments, and embodiments. However, their deployment in physical systems raises a fundamental challenge: generated actions must satisfy the safety, feasibility, and dynamic constraints of the real world. Therefore, a key open problem is to integrate formal constraint satisfaction into generative robot foundation models. Solving this problem is critical for bridging the gap between expressive behavior generation and trustworthy real-world robotic autonomy.
Publications
[1] Huang, T. Y., Lederer, A., Wu, D. J., Dai, X., Zhang, S., Sosnowski, S., & Hirche, S. (2025). SAD-Flower: Flow Matching for Safe, Admissible, and Dynamically Consistent Planning. arXiv preprint arXiv:2511.05355.