We are proud to announce the publication of the article “A roadmap for AI in robotics” in the journal Nature Machine Intelligence — the result of collaborative work between the euROBIN Strategic Research Agenda (SRA) and leading academic voices in the robotics and AI communities.
This Perspective piece offers a comprehensive and thoughtful roadmap for the future of artificial intelligence in robotics. As excitement continues to grow around the potential of AI to address critical challenges in robotics, this publication takes a step back to ask: Which AI approaches are truly suited to the physical, interactive, and dynamic nature of robotics?
Unlike AI systems that analyze static data, robots must act and sense in complex real-world environments. The paper highlights how this adds layers of complexity — from adapting models to individual robot designs and physical tasks, to deploying AI across varied platforms and scenarios.
The authors examine:
-
The progress made in integrating AI into robotics since the 1990s
-
The need for updated and representative datasets for diverse robot tasks
-
How AI can be designed to generalize across platforms while being tailored to robotics-specific problems
-
The importance of explainability, transparency, and trust in human–robot interaction
-
The long-term goals of lifelong learning, safe deployment, and sustainable development
This article is not only a reflection of how far the field has come — it also serves as a strategic reference for the research and development roadmap being shaped within the euROBIN network and beyond.
You can access the full article here: https://www.nature.com/articles/s42256-025-01050-6
A huge thank you to the authors: Aude Billard, Alin Albu-Schaeffer, Michael Beetz, Wolfram Burgard, Peter Corke, Matei Ciocarlie, Ravinder Dahiya, Danica Kragic, Ken Goldberg, Yukie Nagai & Davide Scaramuzza — and to all contributors involved in this milestone work.