The advancement of humanoid robots has been remarkable; however, they still encounter challenges in real-world settings, especially when faced with complex and rapidly changing situations. The collaborative research initiative known as the “RobOdin” project aims to address these limitations by merging artificial intelligence (AI) and robotics expertise from Germany and Denmark to develop a new breed of humanoid assistance systems. These systems are intended to provide support to individuals in their daily work routines in an intuitive, context-aware, and secure manner.
The German Research Center for Artificial Intelligence (DFKI) in Lübeck is a key contributor to the RobOdin project, offering an AI-based decision-making framework that facilitates intelligent and situation-specific actions by the humanoid robots. Together, universities, research institutions, and industry partners from northern Germany and southern Denmark are pooling their efforts to create high-performance humanoid robots tailored for industrial and social applications. The primary goal is to develop reliable systems capable of handling hazardous, repetitive, or physically strenuous tasks to enhance work safety and efficiency.
A significant emphasis is being placed on equipping these robots with interactive and social skills to enable natural and contextually appropriate interactions with humans. The robots are being designed to identify individuals, comprehend language, communicate effectively, and respond adeptly to dynamic situations. The University of Southern Denmark leads the project in collaboration with other partners such as DFKI Laboratory in Lübeck, Flensburg University of Applied Sciences, University of Lübeck, Christian-Albrechts-Universität zu Kiel, as well as industry and network partners like HARTING, Novo Nordisk, Diakonie Nord Nord Ost, the Schleswig-Flensburg Hazardous Materials Fire Brigade, navel robotics, and the University of Twente.
At the DFKI Laboratory in Lübeck, the AI for Assistive Health Technologies research department is spearheading the development of RobOdin’s customizable decision-making framework for humanoid robots. This system, relying on large language models, integrates outputs from various AI modules, including object and emotion recognition, to create a shared knowledge base. By leveraging this model to understand tasks, humanoid robot capabilities, and environmental context, the robots can devise context-specific action strategies to adapt flexibly to varying tasks and environments.
Other project partners are also making crucial technological contributions to the system. For instance, Christian-Albrechts-Universität zu Kiel is focusing on visual perception and social navigation, while the University of Lübeck is researching cognitive abilities to enhance situational understanding and decision-making for humanoid robots. The University of Southern Denmark is responsible for realizing the humanoid robot platform and integrating AI and robotics components developed by partners into a cohesive system.
Furthermore, the project involves testing the humanoid robot in practical scenarios across industries such as assembly, medicine, and emergency response operations. Collaborations with entities like the Schleswig-Flensburg hazardous materials fire brigade and Novo Nordisk ensure that the robots can assist in various tasks, including reconnaissance of hazardous situations and pharmaceutical procedures. The project also aims to establish a lasting network for robotics and AI research in the region, fostering collaboration between academia and industry to drive innovation and enhance regional capacities.
Funded by the European Union through the Germany–Denmark Interreg program, the RobOdin project is slated to run from October 1, 2025, to September 30, 2028, with a total budget of €2,369,227.52. For more information on the project, interested parties can visit the project website at https://robodin.eu/.
About the Author
