ICRA2021 Virtual Workshop
Robot Learning in Real-world Applications: Beyond Proof of Concept

This workshop will be on-line at May 31st, starting from 13.00 JST (6.00 CEST, 21.00 PDT) to 19.00 JST (12.00 CEST, 3.00 PDT)


Robot learning is a regular focal point in research environments in academia and the industry, enabling technology to take robots beyond the laboratory into the real world. The progress is exceptional; however, parameters are still manually tweaked in the industry, service, and interaction robots have not taken over, and robots do not roam alone. Why is this the case when optimization/learning techniques have come so far?

This workshop aims to present robot learning applications in real-world scenarios that go beyond the proof of concept/laboratory, i. e., in the real-world or with explored potential. No specific learning methods or applications are targeted; they only need to change robot actions autonomously.

We are looking to answer if there are factors that limit the applicability of robot learning techniques in scenarios that go beyond proof of concept, illustrate differences between academic/real-world scenarios, and consider practical aspects of taking solutions out of the laboratory. The proposed full-day workshop will present invited presentations by distinguished speakers from real-world practitioners from both industry and academia.


This workshop is supported by the IEEE RAS Technical Committees on:

1) Robotics Research for Practicality

2) Robot Learning

3) Mobile Manipulation

4) Cognitive Robotics