Federated Learning for Robot Picking (FLAIROP)

Funding Agency: BMWK (Germany) and NRC (Canada)
Partners: Festo SE & Co. KG (Germany), KIT (Germany), DarwinAI (Canada), University of Waterloo (Canada)
Duration: Februar 2021 - July 2023

Project Abstract

In the FLAIROP project, we are developing a federated learning framework that enables pick-and-place robots to collaboratively learn robust object recognition and grasping skills without sharing raw training data. The system leverages distributed experience to improve performance on known and unknown objects while respecting data protection requirements. This approach extends federated learning from its roots in medical imaging to Industry 4.0, advancing autonomous handling systems through scalable, privacy-preserving AI.

Links/ Media

- Press release KIT (August 2023)
- Press release Festo SE & Co. KG (August 2023)
- News BMWK (August 2022)