Academic Journal
View in EDS

Variable impedance skill learning for contact-rich manipulation

Bibliographic Details
Title: Variable impedance skill learning for contact-rich manipulation
Authors: Yang, Quantao, Dürr, Alexander, Topp, Elin Anna, Stork, Johannes, Stoyanov, Todor
Contributors: Lund University, Faculty of Engineering, LTH, Departments at LTH, Department of Computer Science, Robotics and Semantic Systems, Lunds universitet, Lunds Tekniska Högskola, Institutioner vid LTH, Institutionen för datavetenskap, Robotik och Semantiska System, Originator, Lund University, Faculty of Engineering, LTH, LTH Profile areas, LTH Profile Area: AI and Digitalization, Lunds universitet, Lunds Tekniska Högskola, LTH profilområden, LTH profilområde: AI och digitalisering, Originator, Lund University, Profile areas and other strong research environments, Strategic research areas (SRA), ELLIIT: the Linköping-Lund initiative on IT and mobile communication, Lunds universitet, Profilområden och andra starka forskningsmiljöer, Strategiska forskningsområden (SFO), ELLIIT: the Linköping-Lund initiative on IT and mobile communication, Originator, Lund University, Faculty of Engineering, LTH, Departments at LTH, Department of Computer Science, Lunds universitet, Lunds Tekniska Högskola, Institutioner vid LTH, Institutionen för datavetenskap, Originator
Source: IEEE Robotics and Automation Letters. 7(3):8391-8398
Subject Terms: Engineering and Technology, Electrical Engineering, Electronic Engineering, Information Engineering, Control Engineering, Teknik, Elektroteknik och elektronik, Reglerteknik, Robotics and automation, Robotik och automation, Natural Sciences, Computer and Information Sciences, Computer graphics and computer vision, Naturvetenskap, Data- och informationsvetenskap (Datateknik), Datorgrafik och datorseende
Description: Contact-rich manipulation tasks remain a hard problem in robotics that requires interaction with unstructured environments. Reinforcement Learning (RL) is one potential solution to such problems, as it has been successfully demonstrated on complex continuous control tasks. Nevertheless, current state-of-the-art methods require policy training in simulation to prevent undesired behavior and later domain transfer even for simple skills involving contact. In this paper, we address the problem of learning contact-rich manipulation policies by extending an existing skill-based RL framework with a variable impedance action space. Our method leverages a small set of suboptimal demonstration trajectories and learns from both position, but also crucially impedance-space information. We evaluate our method on a number of peg-in-hole task variants with a Franka Panda arm and demonstrate that learning variable impedance actions for RL in Cartesian space can be deployed directly on the real robot, without resorting to learning in simulation.
Access URL: https://doi.org/10.1109/LRA.2022.3187276
Database: SwePub
FullText Links:
  – Type: other
Text:
  Availability: 0
CustomLinks:
  – Url: https://doi.org/10.1109/LRA.2022.3187276#
    Name: EDS - SwePub (ns324271)
    Category: fullText
    Text: View record in SwePub
Header DbId: edsswe
DbLabel: SwePub
An: edsswe.oai.portal.research.lu.se.publications.0d10c436.e705.4ed3.8748.87f92ff3588e
RelevancyScore: 1015
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 1015.01336669922
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Variable impedance skill learning for contact-rich manipulation
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Yang%2C+Quantao%22">Yang, Quantao</searchLink><br /><searchLink fieldCode="AR" term="%22Dürr%2C+Alexander%22">Dürr, Alexander</searchLink><br /><searchLink fieldCode="AR" term="%22Topp%2C+Elin+Anna%22">Topp, Elin Anna</searchLink><br /><searchLink fieldCode="AR" term="%22Stork%2C+Johannes%22">Stork, Johannes</searchLink><br /><searchLink fieldCode="AR" term="%22Stoyanov%2C+Todor%22">Stoyanov, Todor</searchLink>
– Name: Author
  Label: Contributors
  Group: Au
  Data: Lund University, Faculty of Engineering, LTH, Departments at LTH, Department of Computer Science, Robotics and Semantic Systems, Lunds universitet, Lunds Tekniska Högskola, Institutioner vid LTH, Institutionen för datavetenskap, Robotik och Semantiska System, Originator<br />Lund University, Faculty of Engineering, LTH, LTH Profile areas, LTH Profile Area: AI and Digitalization, Lunds universitet, Lunds Tekniska Högskola, LTH profilområden, LTH profilområde: AI och digitalisering, Originator<br />Lund University, Profile areas and other strong research environments, Strategic research areas (SRA), ELLIIT: the Linköping-Lund initiative on IT and mobile communication, Lunds universitet, Profilområden och andra starka forskningsmiljöer, Strategiska forskningsområden (SFO), ELLIIT: the Linköping-Lund initiative on IT and mobile communication, Originator<br />Lund University, Faculty of Engineering, LTH, Departments at LTH, Department of Computer Science, Lunds universitet, Lunds Tekniska Högskola, Institutioner vid LTH, Institutionen för datavetenskap, Originator
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <i>IEEE Robotics and Automation Letters</i>. 7(3):8391-8398
– Name: Subject
  Label: Subject Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Engineering+and+Technology%22">Engineering and Technology</searchLink><br /><searchLink fieldCode="DE" term="%22Electrical+Engineering%22">Electrical Engineering</searchLink><br /><searchLink fieldCode="DE" term="%22Electronic+Engineering%22">Electronic Engineering</searchLink><br /><searchLink fieldCode="DE" term="%22Information+Engineering%22">Information Engineering</searchLink><br /><searchLink fieldCode="DE" term="%22Control+Engineering%22">Control Engineering</searchLink><br /><searchLink fieldCode="DE" term="%22Teknik%22">Teknik</searchLink><br /><searchLink fieldCode="DE" term="%22Elektroteknik+och+elektronik%22">Elektroteknik och elektronik</searchLink><br /><searchLink fieldCode="DE" term="%22Reglerteknik%22">Reglerteknik</searchLink><br /><searchLink fieldCode="DE" term="%22Robotics+and+automation%22">Robotics and automation</searchLink><br /><searchLink fieldCode="DE" term="%22Robotik+och+automation%22">Robotik och automation</searchLink><br /><searchLink fieldCode="DE" term="%22Natural+Sciences%22">Natural Sciences</searchLink><br /><searchLink fieldCode="DE" term="%22Computer+and+Information+Sciences%22">Computer and Information Sciences</searchLink><br /><searchLink fieldCode="DE" term="%22Computer+graphics+and+computer+vision%22">Computer graphics and computer vision</searchLink><br /><searchLink fieldCode="DE" term="%22Naturvetenskap%22">Naturvetenskap</searchLink><br /><searchLink fieldCode="DE" term="%22Data-+och+informationsvetenskap+%28Datateknik%29%22">Data- och informationsvetenskap (Datateknik)</searchLink><br /><searchLink fieldCode="DE" term="%22Datorgrafik+och+datorseende%22">Datorgrafik och datorseende</searchLink>
– Name: Abstract
  Label: Description
  Group: Ab
  Data: Contact-rich manipulation tasks remain a hard problem in robotics that requires interaction with unstructured environments. Reinforcement Learning (RL) is one potential solution to such problems, as it has been successfully demonstrated on complex continuous control tasks. Nevertheless, current state-of-the-art methods require policy training in simulation to prevent undesired behavior and later domain transfer even for simple skills involving contact. In this paper, we address the problem of learning contact-rich manipulation policies by extending an existing skill-based RL framework with a variable impedance action space. Our method leverages a small set of suboptimal demonstration trajectories and learns from both position, but also crucially impedance-space information. We evaluate our method on a number of peg-in-hole task variants with a Franka Panda arm and demonstrate that learning variable impedance actions for RL in Cartesian space can be deployed directly on the real robot, without resorting to learning in simulation.
– Name: URL
  Label: Access URL
  Group: URL
  Data: <link linkTarget="URL" linkTerm="https://doi.org/10.1109/LRA.2022.3187276" linkWindow="_blank">https://doi.org/10.1109/LRA.2022.3187276</link>
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsswe&AN=edsswe.oai.portal.research.lu.se.publications.0d10c436.e705.4ed3.8748.87f92ff3588e
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1109/LRA.2022.3187276
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 8
        StartPage: 8391
    Subjects:
      – SubjectFull: Engineering and Technology
        Type: general
      – SubjectFull: Electrical Engineering
        Type: general
      – SubjectFull: Electronic Engineering
        Type: general
      – SubjectFull: Information Engineering
        Type: general
      – SubjectFull: Control Engineering
        Type: general
      – SubjectFull: Teknik
        Type: general
      – SubjectFull: Elektroteknik och elektronik
        Type: general
      – SubjectFull: Reglerteknik
        Type: general
      – SubjectFull: Robotics and automation
        Type: general
      – SubjectFull: Robotik och automation
        Type: general
      – SubjectFull: Natural Sciences
        Type: general
      – SubjectFull: Computer and Information Sciences
        Type: general
      – SubjectFull: Computer graphics and computer vision
        Type: general
      – SubjectFull: Naturvetenskap
        Type: general
      – SubjectFull: Data- och informationsvetenskap (Datateknik)
        Type: general
      – SubjectFull: Datorgrafik och datorseende
        Type: general
    Titles:
      – TitleFull: Variable impedance skill learning for contact-rich manipulation
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Yang, Quantao
      – PersonEntity:
          Name:
            NameFull: Dürr, Alexander
      – PersonEntity:
          Name:
            NameFull: Topp, Elin Anna
      – PersonEntity:
          Name:
            NameFull: Stork, Johannes
      – PersonEntity:
          Name:
            NameFull: Stoyanov, Todor
      – PersonEntity:
          Name:
            NameFull: Lund University, Faculty of Engineering, LTH, Departments at LTH, Department of Computer Science, Robotics and Semantic Systems, Lunds universitet, Lunds Tekniska Högskola, Institutioner vid LTH, Institutionen för datavetenskap, Robotik och Semantiska System, Originator
      – PersonEntity:
          Name:
            NameFull: Lund University, Faculty of Engineering, LTH, LTH Profile areas, LTH Profile Area: AI and Digitalization, Lunds universitet, Lunds Tekniska Högskola, LTH profilområden, LTH profilområde: AI och digitalisering, Originator
      – PersonEntity:
          Name:
            NameFull: Lund University, Profile areas and other strong research environments, Strategic research areas (SRA), ELLIIT: the Linköping-Lund initiative on IT and mobile communication, Lunds universitet, Profilområden och andra starka forskningsmiljöer, Strategiska forskningsområden (SFO), ELLIIT: the Linköping-Lund initiative on IT and mobile communication, Originator
      – PersonEntity:
          Name:
            NameFull: Lund University, Faculty of Engineering, LTH, Departments at LTH, Department of Computer Science, Lunds universitet, Lunds Tekniska Högskola, Institutioner vid LTH, Institutionen för datavetenskap, Originator
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 30
              M: 06
              Type: published
              Y: 2022
          Identifiers:
            – Type: issn-print
              Value: 23773766
            – Type: issn-locals
              Value: SWEPUB_FREE
            – Type: issn-locals
              Value: LU_SWEPUB
          Numbering:
            – Type: volume
              Value: 7
            – Type: issue
              Value: 3
          Titles:
            – TitleFull: IEEE Robotics and Automation Letters
              Type: main
ResultId 1