Endre søk
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Global Search with Bernoulli Alternation Kernel for Task-oriented Grasping Informed by Simulation
Robotics, Perception and Learning, CSC, Royal Institute of Technology, Stockholm, Sweden.
Robotics, Perception and Learning, CSC, Royal Institute of Technology, Stockholm, Sweden.
Robotics, Perception and Learning, CSC, Royal Institute of Technology, Stockholm, Sweden. (AASS)ORCID-id: 0000-0003-3958-6179
Robotics, Perception and Learning, CSC, Royal Institute of Technology, Stockholm, Sweden.
2018 (engelsk)Inngår i: Proceedings of Machine Learning Research: Conference on Robot Learning 2018, PMLR , 2018, Vol. 87, s. 641-650Konferansepaper, Publicerat paper (Fagfellevurdert)
Abstract [en]

We develop an approach that benefits from large simulated datasets and takes full advantage of the limited online data that is most relevant. We propose a variant of Bayesian optimization that alternates between using informed and uninformed kernels. With this Bernoulli Alternation Kernel we ensure that discrepancies between simulation and reality do not hinder adapting robot control policies online. The proposed approach is applied to a challenging real-world problem of task-oriented grasping with novel objects. Our further contribution is a neural network architecture and training pipeline that use experience from grasping objects in simulation to learn grasp stability scores. We learn task scores from a labeled dataset with a convolutional network, which is used to construct an informed kernel for our variant of Bayesian optimization. Experiments on an ABB Yumi robot with real sensor data demonstrate success of our approach, despite the challenge of fulfilling task requirements and high uncertainty over physical properties of objects.

sted, utgiver, år, opplag, sider
PMLR , 2018. Vol. 87, s. 641-650
Serie
Proceedings of Machine Learning Research, ISSN 2640-3498
Emneord [en]
Bayesian optimization, Deep learning, Task-oriented grasping
HSV kategori
Identifikatorer
URN: urn:nbn:se:oru:diva-71550OAI: oai:DiVA.org:oru-71550DiVA, id: diva2:1280223
Konferanse
The 2nd Conference on Robot Learning (CoRL), Zürich, Switzerland, October 29-31, 2018
Forskningsfinansiär
Knut and Alice Wallenberg FoundationTilgjengelig fra: 2019-01-18 Laget: 2019-01-18 Sist oppdatert: 2019-01-24bibliografisk kontrollert

Open Access i DiVA

Fulltekst mangler i DiVA

Andre lenker

Electronic full texthttp://proceedings.mlr.press/v87/antonova18a.html

Søk i DiVA

Av forfatter/redaktør
Stork, Johannes Andreas

Søk utenfor DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric

urn-nbn
Totalt: 101 treff
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf