SKID RAW: Skill Discovery from Raw Trajectories

Daniel Tanneberg, Kai Ploeger, Elmar Rueckert, Jan Peters

Publikation: Beitrag in FachzeitschriftArtikelForschungBegutachtung

2 Zitate (Scopus)

Abstract

Integrating robots in complex everyday environments requires a multitude of problems to be solved. One crucial feature among those is to equip robots with a mechanism for teaching them a new task in an easy and natural way. When teaching tasks that involve sequences of different skills, with varying order and number of these skills, it is desirable to only demonstrate full task executions instead of all individual skills. For this purpose, we propose a novel approach that simultaneously learns to segment trajectories into reoccurring patterns and the skills to reconstruct these patterns from unlabelled demonstrations without further supervision. Moreover, the approach learns a skill conditioning that can be used to understand possible sequences of skills, a practical mechanism to be used in, for example, human-robot-interactions for a more intelligent and adaptive robot behaviour. The Bayesian and variational inference based approach is evaluated on synthetic and real human demonstrations with varying complexities and dimensionality, showing the successful learning of segmentations and skill libraries from unlabelled data.
OriginalspracheEnglisch
Aufsatznummer9387162
Seiten (von - bis)4696-4703
Seitenumfang8
Fachzeitschrift IEEE robotics and automation letters
Jahrgang6
Ausgabenummer3
Frühes Online-Datum25 März 2021
DOIs
PublikationsstatusVeröffentlicht - Juli 2021

Bibliographische Notiz

Funding Information:
Manuscript received December 2, 2020; accepted March 8, 2021. Date of publication March 25, 2021; date of current version April 13, 2021. This letter was recommended for publication by Associate Editor M. Burke and Editor D. Kulic upon evaluation of the reviewers’ comments. This work was supported in part by the European Union’s Horizon 2020 Research and Innovation Programme under Grant 713010 (GOAL-Robots) and 640554 (SKILLS4ROBOTS), in part by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) -No. 430054590 (TRAIN), and in part by NVIDIA. (Corresponding author: Daniel Tanneberg.) Daniel Tanneberg and Kai Ploeger are with the Intelligent Autonomous Systems, Technische Universität Darmstadt, 64289 Darmstadt, Germany (e-mail: daniel@robot-learning.de; kai.ploeger@tu-darmstadt.de).

Publisher Copyright:
© 2016 IEEE.

Schlagwörter

  • robot motion control
  • learning
  • movement primitives

Dieses zitieren