A Real-Robot Dataset for Assessing Transferability of Learned Dynamics Models

Diego Agudelo-Espana, Andrii Zadaianchuk, Philippe Wenk, Aditya Garg, Joel Akpo, Felix Grimminger, Julian Viereck, Maximilien Naveau, Ludovic Righetti, Georg Martius, Andreas Krause, Bernhard Scholkopf, Stefan Bauer, Manuel Wuthrich

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In the context of model-based reinforcement learning and control, a large number of methods for learning system dynamics have been proposed in recent years. The purpose of these learned models is to synthesize new control policies. An important open question is how robust current dynamics-learning methods are to shifts in the data distribution due to changes in the control policy. We present a real-robot dataset which allows to systematically investigate this question. This dataset contains trajectories of a 3 degrees-of-freedom (DOF) robot being controlled by a diverse set of policies. For comparison, we also provide a simulated version of the dataset. Finally, we benchmark a few widely-used dynamics-learning methods using the proposed dataset. Our results show that the iid test error of a learned model is not necessarily a good indicator of its accuracy under control policies different from the one which generated the training data. This suggests that it may be important to evaluate dynamics-learning methods in terms of their transfer performance, rather than only their iid error.

Original languageEnglish (US)
Title of host publication2020 IEEE International Conference on Robotics and Automation, ICRA 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages8151-8157
Number of pages7
ISBN (Electronic)9781728173955
DOIs
StatePublished - May 2020
Event2020 IEEE International Conference on Robotics and Automation, ICRA 2020 - Paris, France
Duration: May 31 2020Aug 31 2020

Publication series

NameProceedings - IEEE International Conference on Robotics and Automation
ISSN (Print)1050-4729

Conference

Conference2020 IEEE International Conference on Robotics and Automation, ICRA 2020
CountryFrance
CityParis
Period5/31/208/31/20

ASJC Scopus subject areas

  • Software
  • Control and Systems Engineering
  • Artificial Intelligence
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'A Real-Robot Dataset for Assessing Transferability of Learned Dynamics Models'. Together they form a unique fingerprint.

Cite this