Biosketch
Julian Balletshofer received his bachelor’s degree in mechanical engineering from the Baden-Wuerttemberg Cooperative State University (DHBW) in cooperation with BSH Home Appliances Corporation in 2020. Afterwards, he enrolled at the Technical University of Munich to pursue his master’s degree in Robotics, Cognition, and Intelligence. During his master’s program, he completed a research internship at Disney Zurich, where he focused on computer vision. In 2024, he joined the Cyber-Physical Systems Group as a PhD candidate under the supervision of Prof. Dr.-Ing. Matthias Althoff.
His research focuses on safe human-robot interaction using machine learning and formal methods techniques.
relAI Research
Safe human-robot interaction using machine learning and formal methods techniques
My research focuses on enabling safe and efficient human–robot interaction for autonomous robotic systems operating in close proximity to humans. To achieve this, I investigate three key aspects of robot safety. First, I develop collision models that accurately characterize energy transfer during impacts and improve the prediction of potential injury risk in human–robot collisions. Second, I integrate these models into formal safety filters that provide provable guarantees on safe robot behavior during physical interaction. Third, I study how such safety filters can be adapted and optimized for modern machine learning-based control approaches, with the goal of safeguarding autonomous decision-making while maintaining high task performance and efficiency. By combining physically grounded collision modeling, formal safety guarantees, and learning-based robotics, this research aims to establish practical frameworks for safe, reliable, and high-performing collaborative robotic systems.
Publications
https://arxiv.org/abs/2412.10180
https://arxiv.org/pdf/2511.06385
https://ieeexplore.ieee.org/document/11247577