“This year's Alfried Krupp Award honours Professor Georgia Chalvatzaki and her pioneering research into AI, machine learning, and robotics,” says Professor Ursula Gather, Chair of the Krupp Foundation's Board of Trustees. “With her innovative approach, she views robots, humans and their environment as an integrated system. Her aim is to develop robots that learn from experience and continuously adapt their behaviour. Her scientific findings have the potential to make an effective impact in areas such as healthcare, logistics, and sustainable agriculture. The Krupp Foundation is delighted to support Professor Chalvatzaki in her future scientific endeavours."

'It makes me proud that a scientist from TU Darmstadt is receiving the Alfried Krupp Prize, one of the most prestigious awards for young scientists and engineers holding a first professorship.' This award once again underlines Ms Chalvatzaki's scientific excellence in the field of robotics, artificial intelligence, and human–robot collaboration. She is thus making a decisive contribution to the success of hessian.AI, our Information and Intelligence (I+I) research field, and to the future viability of our university,' says Professor Tanja Brühl, President of TU Darmstadt.

Professor Georgia Chalvatzaki is researching robots that can learn from their interactions with their environment and with humans in real time. This groundbreaking approach puts humans at the centre of robotics and paves the way for a new generation of robots. Her research in interactive robot perception and learning aims to create robotic systems capable of performing complex tasks in dynamic, unstructured environments.

A central focus of her work is mobile humanoid manipulators — robots with mobile bases and anthropomorphic arms that can move in human environments, perceive their surroundings, and perform skilful grasping and manipulation tasks. These systems are based on the interplay between perception, planning, and action. To achieve this, Chalvatzaki combines classical model-based robotics with modern learning methods, such as reinforcement learning, imitation learning, and graph-based neural networks. Through structured representations of the environment, the robots gain an in-depth understanding of their surroundings, enabling them to adapt their behaviour in a context-sensitive manner. A significant achievement in this area of research is the introduction of SE(3) diffusion models — geometry-based learning methods that incorporate the structure of three-dimensional space (position and orientation) into generative model training. These models allow robots to perform precise and fluid gripping and movement sequences, even in cluttered or crowded environments.

Human-centred robotics

A central goal of Georgia Chalvatzaki's research is to develop human-centred robotics, where robots are seen as trustworthy partners for humans, particularly in sensitive areas such as care, mobility and education. Her learning systems should deliver technical precision and enable safe, intuitive interaction through transparency and adaptability. The social added value of her research is evident in its specific applications. In healthcare, for instance, their mobile assistance systems facilitate safe human-robot co-navigation, with robots able to dynamically adapt to a person's walking speed and movement patterns — an important development for use in caring for the elderly. In industrial logistics, Chalvatzaki's adaptive systems contribute to the automation of supermarket, supply chain and airport processes. In sustainable agriculture, they are developing solutions for robot-assisted harvesting, transplanting, and maintaining hydro and solar plants.