Loading…

Uncertainty-driven Exploration Strategies for Online Grasp Learning

Existing grasp prediction approaches are mostly based on offline learning, while, ignoring the exploratory grasp learning during online adaptation to new picking scenarios, i.e., objects that are unseen or out-of-domain (OOD), camera and bin settings, etc. In this paper, we present an uncertainty-ba...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2024-04
Main Authors: Shi, Yitian, Schillinger, Philipp, Gabriel, Miroslav, Qualmann, Alexander, Feldman, Zohar, Ziesche, Hanna, Ngo Anh Vien
Format: Article
Language:English
Subjects:
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Shi, Yitian
Schillinger, Philipp
Gabriel, Miroslav
Qualmann, Alexander
Feldman, Zohar
Ziesche, Hanna
Ngo Anh Vien
description Existing grasp prediction approaches are mostly based on offline learning, while, ignoring the exploratory grasp learning during online adaptation to new picking scenarios, i.e., objects that are unseen or out-of-domain (OOD), camera and bin settings, etc. In this paper, we present an uncertainty-based approach for online learning of grasp predictions for robotic bin picking. Specifically, the online learning algorithm with an effective exploration strategy can significantly improve its adaptation performance to unseen environment settings. To this end, we first propose to formulate online grasp learning as an RL problem that will allow us to adapt both grasp reward prediction and grasp poses. We propose various uncertainty estimation schemes based on Bayesian uncertainty quantification and distributional ensembles. We carry out evaluations on real-world bin picking scenes of varying difficulty. The objects in the bin have various challenging physical and perceptual characteristics that can be characterized by semi- or total transparency, and irregular or curved surfaces. The results of our experiments demonstrate a notable improvement of grasp performance in comparison to conventional online learning methods which incorporate only naive exploration strategies. Video: https://youtu.be/fPKOrjC2QrU
format article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2867416409</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2867416409</sourcerecordid><originalsourceid>FETCH-proquest_journals_28674164093</originalsourceid><addsrcrecordid>eNqNiksKwjAUAIMgWLR3CLgupEn6cV2qLgQX6roEfS0p5aW-pKK3twsP4GoGZhYskkqlSamlXLHY-14IIfNCZpmKWHXDO1AwFsMneZB9AfL6PQ6OTLAO-SXMAp0Fz1tH_IyDReAHMn7kJzCEFrsNW7Zm8BD_uGbbfX2tjslI7jmBD03vJsI5NbLMC53mWuzUf9cXTCU6ug</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2867416409</pqid></control><display><type>article</type><title>Uncertainty-driven Exploration Strategies for Online Grasp Learning</title><source>Publicly Available Content Database</source><creator>Shi, Yitian ; Schillinger, Philipp ; Gabriel, Miroslav ; Qualmann, Alexander ; Feldman, Zohar ; Ziesche, Hanna ; Ngo Anh Vien</creator><creatorcontrib>Shi, Yitian ; Schillinger, Philipp ; Gabriel, Miroslav ; Qualmann, Alexander ; Feldman, Zohar ; Ziesche, Hanna ; Ngo Anh Vien</creatorcontrib><description>Existing grasp prediction approaches are mostly based on offline learning, while, ignoring the exploratory grasp learning during online adaptation to new picking scenarios, i.e., objects that are unseen or out-of-domain (OOD), camera and bin settings, etc. In this paper, we present an uncertainty-based approach for online learning of grasp predictions for robotic bin picking. Specifically, the online learning algorithm with an effective exploration strategy can significantly improve its adaptation performance to unseen environment settings. To this end, we first propose to formulate online grasp learning as an RL problem that will allow us to adapt both grasp reward prediction and grasp poses. We propose various uncertainty estimation schemes based on Bayesian uncertainty quantification and distributional ensembles. We carry out evaluations on real-world bin picking scenes of varying difficulty. The objects in the bin have various challenging physical and perceptual characteristics that can be characterized by semi- or total transparency, and irregular or curved surfaces. The results of our experiments demonstrate a notable improvement of grasp performance in comparison to conventional online learning methods which incorporate only naive exploration strategies. Video: https://youtu.be/fPKOrjC2QrU</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Adaptation ; Algorithms ; Cameras ; Distance learning ; Grasping (robotics) ; Machine learning ; Picking ; Uncertainty</subject><ispartof>arXiv.org, 2024-04</ispartof><rights>2024. This work is published under http://creativecommons.org/licenses/by-nc-nd/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.proquest.com/docview/2867416409?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>780,784,25753,37012,44590</link.rule.ids></links><search><creatorcontrib>Shi, Yitian</creatorcontrib><creatorcontrib>Schillinger, Philipp</creatorcontrib><creatorcontrib>Gabriel, Miroslav</creatorcontrib><creatorcontrib>Qualmann, Alexander</creatorcontrib><creatorcontrib>Feldman, Zohar</creatorcontrib><creatorcontrib>Ziesche, Hanna</creatorcontrib><creatorcontrib>Ngo Anh Vien</creatorcontrib><title>Uncertainty-driven Exploration Strategies for Online Grasp Learning</title><title>arXiv.org</title><description>Existing grasp prediction approaches are mostly based on offline learning, while, ignoring the exploratory grasp learning during online adaptation to new picking scenarios, i.e., objects that are unseen or out-of-domain (OOD), camera and bin settings, etc. In this paper, we present an uncertainty-based approach for online learning of grasp predictions for robotic bin picking. Specifically, the online learning algorithm with an effective exploration strategy can significantly improve its adaptation performance to unseen environment settings. To this end, we first propose to formulate online grasp learning as an RL problem that will allow us to adapt both grasp reward prediction and grasp poses. We propose various uncertainty estimation schemes based on Bayesian uncertainty quantification and distributional ensembles. We carry out evaluations on real-world bin picking scenes of varying difficulty. The objects in the bin have various challenging physical and perceptual characteristics that can be characterized by semi- or total transparency, and irregular or curved surfaces. The results of our experiments demonstrate a notable improvement of grasp performance in comparison to conventional online learning methods which incorporate only naive exploration strategies. Video: https://youtu.be/fPKOrjC2QrU</description><subject>Adaptation</subject><subject>Algorithms</subject><subject>Cameras</subject><subject>Distance learning</subject><subject>Grasping (robotics)</subject><subject>Machine learning</subject><subject>Picking</subject><subject>Uncertainty</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>PIMPY</sourceid><recordid>eNqNiksKwjAUAIMgWLR3CLgupEn6cV2qLgQX6roEfS0p5aW-pKK3twsP4GoGZhYskkqlSamlXLHY-14IIfNCZpmKWHXDO1AwFsMneZB9AfL6PQ6OTLAO-SXMAp0Fz1tH_IyDReAHMn7kJzCEFrsNW7Zm8BD_uGbbfX2tjslI7jmBD03vJsI5NbLMC53mWuzUf9cXTCU6ug</recordid><startdate>20240424</startdate><enddate>20240424</enddate><creator>Shi, Yitian</creator><creator>Schillinger, Philipp</creator><creator>Gabriel, Miroslav</creator><creator>Qualmann, Alexander</creator><creator>Feldman, Zohar</creator><creator>Ziesche, Hanna</creator><creator>Ngo Anh Vien</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20240424</creationdate><title>Uncertainty-driven Exploration Strategies for Online Grasp Learning</title><author>Shi, Yitian ; Schillinger, Philipp ; Gabriel, Miroslav ; Qualmann, Alexander ; Feldman, Zohar ; Ziesche, Hanna ; Ngo Anh Vien</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_28674164093</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Adaptation</topic><topic>Algorithms</topic><topic>Cameras</topic><topic>Distance learning</topic><topic>Grasping (robotics)</topic><topic>Machine learning</topic><topic>Picking</topic><topic>Uncertainty</topic><toplevel>online_resources</toplevel><creatorcontrib>Shi, Yitian</creatorcontrib><creatorcontrib>Schillinger, Philipp</creatorcontrib><creatorcontrib>Gabriel, Miroslav</creatorcontrib><creatorcontrib>Qualmann, Alexander</creatorcontrib><creatorcontrib>Feldman, Zohar</creatorcontrib><creatorcontrib>Ziesche, Hanna</creatorcontrib><creatorcontrib>Ngo Anh Vien</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central</collection><collection>ProQuest Central</collection><collection>AUTh Library subscriptions: ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Shi, Yitian</au><au>Schillinger, Philipp</au><au>Gabriel, Miroslav</au><au>Qualmann, Alexander</au><au>Feldman, Zohar</au><au>Ziesche, Hanna</au><au>Ngo Anh Vien</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Uncertainty-driven Exploration Strategies for Online Grasp Learning</atitle><jtitle>arXiv.org</jtitle><date>2024-04-24</date><risdate>2024</risdate><eissn>2331-8422</eissn><abstract>Existing grasp prediction approaches are mostly based on offline learning, while, ignoring the exploratory grasp learning during online adaptation to new picking scenarios, i.e., objects that are unseen or out-of-domain (OOD), camera and bin settings, etc. In this paper, we present an uncertainty-based approach for online learning of grasp predictions for robotic bin picking. Specifically, the online learning algorithm with an effective exploration strategy can significantly improve its adaptation performance to unseen environment settings. To this end, we first propose to formulate online grasp learning as an RL problem that will allow us to adapt both grasp reward prediction and grasp poses. We propose various uncertainty estimation schemes based on Bayesian uncertainty quantification and distributional ensembles. We carry out evaluations on real-world bin picking scenes of varying difficulty. The objects in the bin have various challenging physical and perceptual characteristics that can be characterized by semi- or total transparency, and irregular or curved surfaces. The results of our experiments demonstrate a notable improvement of grasp performance in comparison to conventional online learning methods which incorporate only naive exploration strategies. Video: https://youtu.be/fPKOrjC2QrU</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2024-04
issn 2331-8422
language eng
recordid cdi_proquest_journals_2867416409
source Publicly Available Content Database
subjects Adaptation
Algorithms
Cameras
Distance learning
Grasping (robotics)
Machine learning
Picking
Uncertainty
title Uncertainty-driven Exploration Strategies for Online Grasp Learning
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T09%3A10%3A51IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Uncertainty-driven%20Exploration%20Strategies%20for%20Online%20Grasp%20Learning&rft.jtitle=arXiv.org&rft.au=Shi,%20Yitian&rft.date=2024-04-24&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2867416409%3C/proquest%3E%3Cgrp_id%3Ecdi_FETCH-proquest_journals_28674164093%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2867416409&rft_id=info:pmid/&rfr_iscdi=true