RoboSAPIENS: Freethinking Industrial Robots

With funding from the European Commission, researchers are using artificial intelligence to develop freethinking industrial robots.

The RoboSAPIENS EU project has secured a EUR 7.5 million grant from Horizon Europe, the EU’s research and innovation funding program. 

The European RoboSAPIENS are not related to the popular toy robot from the 2000s created by WowWee. The 2024 RoboSAPIENS stands for Robotic Safe Adaptation In Unprecedented Situations. 

Project overview from RoboSapiens EU YouTube.

The project kicked off in January and is funded through December 2026. Aarhus University in Denmark is leading and coordinating the project. 

Professor Peter Gorm Larsen, the project coordinator, said in an announcement, “We see great benefits if robots can adapt to conditions, learn from their surroundings and then change their behavior to better suit the task.”

He said the project aims to unlock the ability for robots to gradually improve on their own. Researchers want to demonstrate that safety can be achieved without recertifications, and that industrial robots can adapt appropriately with intelligent safety systems and control software. 

Larsen is a vice head at Aarhus University’s Department of Electrical and Computer Engineering and an expert in complex cyber-physical systems. 

Mikkel Labori Olsen, a consultant who’s participating on the behalf of the Danish Technological Institute, said the project aligns with Europe’s circular economy initiatives. Manufacturers have used robots to boost productivity for a long time, he explained, and now they’re evolving to help recycle, upgrade, and extend the life of products rather than just making new ones.

They’ll specifically explore automating laptop refurbishment. He said researchers are considering screen repairs now but plan to expand scope. 

The project’s stated goals are to:

  • Help robots learn to adjust by themselves when confronted with new and unexpected situations
  • Ensure the robotic adjustments are safe and reliable
  • Leverage deep machine learning to assist robots making smarter decisions
  • Develop ‘first-time-right’ tools to help robots complete tasks right and safely at first attempt 

They will demonstrate four industry-scale use cases centered around industrial disassembly, a warehouse robotic swarm, an autonomous vessel’s prolonged hull, and a human-robot interaction. 

Other organizations involved with the project include the University of Antwerp, Aristotle University of Thessaloniki (AUTH), the Norwegian University of Science and Technology (NTNU), the University of York, PAL Robotics, Fraunhofer IFF, ISDI Accelerator, and Simula Research Laboratory.

Featured Videos

Digital Twins: The Next 75 Years [Future Applications from 2025 - 2100]

Kalil 4.0 May 13, 2024 3:22 am

Navy's Slow Adoption of Digital Tools Hurting Shipbuilding Programs: GAO Report

Kalil 4.0 May 8, 2024 3:21 am