New Horizons for Robot Learning: Closing the Innovation Loop

Workshop at the 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems

A workshop on “everything you’ve always wanted to ask about founding a start-up after your PhD, but were to afraid to ask”. These question will be asked at this workshop!

When & Where

Date: Friday 01.10.2021, 16:10-18:10, CET

Location: IROS, Hall 8, (If this link does not work, please find us in hall 8 on gcon:


Robots act in the real world. When deploying machine learning on robots, the continuous and dynamic nature of the physical world raises many challenges which are not encountered in purely digital domains such as Internet search and social networks. As an example, many learning algorithms rely on exploration, which, when deployed on real robots, may potentially result in physically dangerous behaviors and costly wear-and-tear.

The main objective of this workshop is to provide a forum for identifying such challenges, to present state-of-the-art research methods for addressing them, and to highlight best practices for transferring such methods from research to industry. An important question is to identify when learning algorithms could and should be used during a robot lifecycle, how they should be used, and what for.

This workshop thus brings together roboticists from academia and industry, to enable an interactive exchange on the future of robot learning, and best practices on transferring ideas to the real world of industrial competition. The speakers at this workshop all aim at real-world robotic applications of machine learning methods. Furthermore, many speakers are active both in academia and industry.

This workshop is the second edition of “New Horizons for Robot Learning”, the first which was held at IROS 2020 (



Dr. Freek Stulp German Aerospace Center (DLR), Institute for Robotics and Mechatronics, Oberpfaffenhofen-Weßling, Germany

Dr. Michael Spranger, Sony AI, Tokyo, Japan

Prof. Stephane Doncieux ISIR, Sorbonne Université/CNRS, Paris, France

Dr. Kim Listmann Bender GmbH

Format and schedule

The workshop is organized as a set of live interviews with global leaders in the field of machine learning and robotics. Questions are tailored to the main workshop topic, and will specifically focus on innovation, i.e. the transition from research ideas to their practical implementation as a product or service. The audience will be able to provide additional questions during the workshop. The insights from the interviews will form the basis for a panel discussion with all the invitees.

Time (CET)  
16:10 – 16:20 Introduction
16:20 – 16:40 Q&A Moritz Tenorth
16:40 – 17:00 Q&A David Ha
17:00 – 17:20 Q&A George Konidaris
17:20 – 17:40 Q&A Pieter Abbeel
17:40 – 18:05 Panel discussion. Moderator: Kim Listmann
18:05 – 18:10 Wrap-up and farewell


The workshop is endorsed by the IEEE RAS TC “Robot Learning”.

The workshop is supported by the EU H2020 project “VERtical Innovation in the Domain of Robotics Enabled by Artificial intelligence Methods” (