Soft robot perception using embedded soft sensors and recurrent neural networks

Recent work has begun to explore the design of biologically inspired soft robots composed of soft, stretchable materials for applications including the handling of delicate materials and safe interaction with humans. However, the solid-state sensors traditionally used in robotics are unable to captu...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Science robotics 2019-01, Vol.4 (26)
Hauptverfasser: Thuruthel, Thomas George, Shih, Benjamin, Laschi, Cecilia, Tolley, Michael Thomas
Format: Artikel
Sprache:eng
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue 26
container_start_page
container_title Science robotics
container_volume 4
creator Thuruthel, Thomas George
Shih, Benjamin
Laschi, Cecilia
Tolley, Michael Thomas
description Recent work has begun to explore the design of biologically inspired soft robots composed of soft, stretchable materials for applications including the handling of delicate materials and safe interaction with humans. However, the solid-state sensors traditionally used in robotics are unable to capture the high-dimensional deformations of soft systems. Embedded soft resistive sensors have the potential to address this challenge. However, both the soft sensors-and the encasing dynamical system-often exhibit nonlinear time-variant behavior, which makes them difficult to model. In addition, the problems of sensor design, placement, and fabrication require a great deal of human input and previous knowledge. Drawing inspiration from the human perceptive system, we created a synthetic analog. Our synthetic system builds models using a redundant and unstructured sensor topology embedded in a soft actuator, a vision-based motion capture system for ground truth, and a general machine learning approach. This allows us to model an unknown soft actuated system. We demonstrate that the proposed approach is able to model the kinematics of a soft continuum actuator in real time while being robust to sensor nonlinearities and drift. In addition, we show how the same system can estimate the applied forces while interacting with external objects. The role of action in perception is also presented. This approach enables the development of force and deformation models for soft robotic systems, which can be useful for a variety of applications, including human-robot interaction, soft orthotics, and wearable robotics.
doi_str_mv 10.1126/scirobotics.aav1488
format Article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_2457281585</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2457281585</sourcerecordid><originalsourceid>FETCH-LOGICAL-c416t-44c57554a15387dd00ddf887c1cd757200263b640a90d7d3722269aec57d07183</originalsourceid><addsrcrecordid>eNpNkMtOwzAQRS0EolXpFyAhL9kE_IrtLlHFS6rUBbCOHHuCAklcPAmIvyelBXV1Z3HPHekQcs7ZFedCX6OvUyxjX3u8cu6TK2uPyFQow7KFMvr44J6QOeIbY4wbLbUSp2QiJZfGaDEl66dY9fR3im4gedj0dezogHX3SqEtIQQIFLclhA5jQuq6QBP4ISXoetrBkFwzRv8V0zuekZPKNQjzfc7Iy93t8_IhW63vH5c3q8wrrvtMKZ-bPFeO59KaEBgLobLWeO6DyY1gTGhZasXcggUTpBFC6IWDkQrMcCtn5HK3u0nxYwDsi7ZGD03jOogDFkKNK5bnNh-rclf1KSImqIpNqluXvgvOiq3M4kBmsZc5Uhf7B0PZQvhn_tTJH_ZFdFs</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2457281585</pqid></control><display><type>article</type><title>Soft robot perception using embedded soft sensors and recurrent neural networks</title><source>American Association for the Advancement of Science</source><creator>Thuruthel, Thomas George ; Shih, Benjamin ; Laschi, Cecilia ; Tolley, Michael Thomas</creator><creatorcontrib>Thuruthel, Thomas George ; Shih, Benjamin ; Laschi, Cecilia ; Tolley, Michael Thomas</creatorcontrib><description>Recent work has begun to explore the design of biologically inspired soft robots composed of soft, stretchable materials for applications including the handling of delicate materials and safe interaction with humans. However, the solid-state sensors traditionally used in robotics are unable to capture the high-dimensional deformations of soft systems. Embedded soft resistive sensors have the potential to address this challenge. However, both the soft sensors-and the encasing dynamical system-often exhibit nonlinear time-variant behavior, which makes them difficult to model. In addition, the problems of sensor design, placement, and fabrication require a great deal of human input and previous knowledge. Drawing inspiration from the human perceptive system, we created a synthetic analog. Our synthetic system builds models using a redundant and unstructured sensor topology embedded in a soft actuator, a vision-based motion capture system for ground truth, and a general machine learning approach. This allows us to model an unknown soft actuated system. We demonstrate that the proposed approach is able to model the kinematics of a soft continuum actuator in real time while being robust to sensor nonlinearities and drift. In addition, we show how the same system can estimate the applied forces while interacting with external objects. The role of action in perception is also presented. This approach enables the development of force and deformation models for soft robotic systems, which can be useful for a variety of applications, including human-robot interaction, soft orthotics, and wearable robotics.</description><identifier>ISSN: 2470-9476</identifier><identifier>EISSN: 2470-9476</identifier><identifier>DOI: 10.1126/scirobotics.aav1488</identifier><identifier>PMID: 33137762</identifier><language>eng</language><publisher>United States</publisher><ispartof>Science robotics, 2019-01, Vol.4 (26)</ispartof><rights>Copyright © 2019 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim to original U.S. Government Works.</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c416t-44c57554a15387dd00ddf887c1cd757200263b640a90d7d3722269aec57d07183</citedby><cites>FETCH-LOGICAL-c416t-44c57554a15387dd00ddf887c1cd757200263b640a90d7d3722269aec57d07183</cites><orcidid>0000-0001-5248-1043 ; 0000-0001-7821-7777 ; 0000-0003-0571-1672 ; 0000-0001-8463-2694</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,2884,2885,27924,27925</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/33137762$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Thuruthel, Thomas George</creatorcontrib><creatorcontrib>Shih, Benjamin</creatorcontrib><creatorcontrib>Laschi, Cecilia</creatorcontrib><creatorcontrib>Tolley, Michael Thomas</creatorcontrib><title>Soft robot perception using embedded soft sensors and recurrent neural networks</title><title>Science robotics</title><addtitle>Sci Robot</addtitle><description>Recent work has begun to explore the design of biologically inspired soft robots composed of soft, stretchable materials for applications including the handling of delicate materials and safe interaction with humans. However, the solid-state sensors traditionally used in robotics are unable to capture the high-dimensional deformations of soft systems. Embedded soft resistive sensors have the potential to address this challenge. However, both the soft sensors-and the encasing dynamical system-often exhibit nonlinear time-variant behavior, which makes them difficult to model. In addition, the problems of sensor design, placement, and fabrication require a great deal of human input and previous knowledge. Drawing inspiration from the human perceptive system, we created a synthetic analog. Our synthetic system builds models using a redundant and unstructured sensor topology embedded in a soft actuator, a vision-based motion capture system for ground truth, and a general machine learning approach. This allows us to model an unknown soft actuated system. We demonstrate that the proposed approach is able to model the kinematics of a soft continuum actuator in real time while being robust to sensor nonlinearities and drift. In addition, we show how the same system can estimate the applied forces while interacting with external objects. The role of action in perception is also presented. This approach enables the development of force and deformation models for soft robotic systems, which can be useful for a variety of applications, including human-robot interaction, soft orthotics, and wearable robotics.</description><issn>2470-9476</issn><issn>2470-9476</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2019</creationdate><recordtype>article</recordtype><recordid>eNpNkMtOwzAQRS0EolXpFyAhL9kE_IrtLlHFS6rUBbCOHHuCAklcPAmIvyelBXV1Z3HPHekQcs7ZFedCX6OvUyxjX3u8cu6TK2uPyFQow7KFMvr44J6QOeIbY4wbLbUSp2QiJZfGaDEl66dY9fR3im4gedj0dezogHX3SqEtIQQIFLclhA5jQuq6QBP4ISXoetrBkFwzRv8V0zuekZPKNQjzfc7Iy93t8_IhW63vH5c3q8wrrvtMKZ-bPFeO59KaEBgLobLWeO6DyY1gTGhZasXcggUTpBFC6IWDkQrMcCtn5HK3u0nxYwDsi7ZGD03jOogDFkKNK5bnNh-rclf1KSImqIpNqluXvgvOiq3M4kBmsZc5Uhf7B0PZQvhn_tTJH_ZFdFs</recordid><startdate>20190130</startdate><enddate>20190130</enddate><creator>Thuruthel, Thomas George</creator><creator>Shih, Benjamin</creator><creator>Laschi, Cecilia</creator><creator>Tolley, Michael Thomas</creator><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><orcidid>https://orcid.org/0000-0001-5248-1043</orcidid><orcidid>https://orcid.org/0000-0001-7821-7777</orcidid><orcidid>https://orcid.org/0000-0003-0571-1672</orcidid><orcidid>https://orcid.org/0000-0001-8463-2694</orcidid></search><sort><creationdate>20190130</creationdate><title>Soft robot perception using embedded soft sensors and recurrent neural networks</title><author>Thuruthel, Thomas George ; Shih, Benjamin ; Laschi, Cecilia ; Tolley, Michael Thomas</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c416t-44c57554a15387dd00ddf887c1cd757200263b640a90d7d3722269aec57d07183</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Thuruthel, Thomas George</creatorcontrib><creatorcontrib>Shih, Benjamin</creatorcontrib><creatorcontrib>Laschi, Cecilia</creatorcontrib><creatorcontrib>Tolley, Michael Thomas</creatorcontrib><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><jtitle>Science robotics</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Thuruthel, Thomas George</au><au>Shih, Benjamin</au><au>Laschi, Cecilia</au><au>Tolley, Michael Thomas</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Soft robot perception using embedded soft sensors and recurrent neural networks</atitle><jtitle>Science robotics</jtitle><addtitle>Sci Robot</addtitle><date>2019-01-30</date><risdate>2019</risdate><volume>4</volume><issue>26</issue><issn>2470-9476</issn><eissn>2470-9476</eissn><abstract>Recent work has begun to explore the design of biologically inspired soft robots composed of soft, stretchable materials for applications including the handling of delicate materials and safe interaction with humans. However, the solid-state sensors traditionally used in robotics are unable to capture the high-dimensional deformations of soft systems. Embedded soft resistive sensors have the potential to address this challenge. However, both the soft sensors-and the encasing dynamical system-often exhibit nonlinear time-variant behavior, which makes them difficult to model. In addition, the problems of sensor design, placement, and fabrication require a great deal of human input and previous knowledge. Drawing inspiration from the human perceptive system, we created a synthetic analog. Our synthetic system builds models using a redundant and unstructured sensor topology embedded in a soft actuator, a vision-based motion capture system for ground truth, and a general machine learning approach. This allows us to model an unknown soft actuated system. We demonstrate that the proposed approach is able to model the kinematics of a soft continuum actuator in real time while being robust to sensor nonlinearities and drift. In addition, we show how the same system can estimate the applied forces while interacting with external objects. The role of action in perception is also presented. This approach enables the development of force and deformation models for soft robotic systems, which can be useful for a variety of applications, including human-robot interaction, soft orthotics, and wearable robotics.</abstract><cop>United States</cop><pmid>33137762</pmid><doi>10.1126/scirobotics.aav1488</doi><orcidid>https://orcid.org/0000-0001-5248-1043</orcidid><orcidid>https://orcid.org/0000-0001-7821-7777</orcidid><orcidid>https://orcid.org/0000-0003-0571-1672</orcidid><orcidid>https://orcid.org/0000-0001-8463-2694</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 2470-9476
ispartof Science robotics, 2019-01, Vol.4 (26)
issn 2470-9476
2470-9476
language eng
recordid cdi_proquest_miscellaneous_2457281585
source American Association for the Advancement of Science
title Soft robot perception using embedded soft sensors and recurrent neural networks
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T04%3A26%3A53IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Soft%20robot%20perception%20using%20embedded%20soft%20sensors%20and%20recurrent%20neural%20networks&rft.jtitle=Science%20robotics&rft.au=Thuruthel,%20Thomas%20George&rft.date=2019-01-30&rft.volume=4&rft.issue=26&rft.issn=2470-9476&rft.eissn=2470-9476&rft_id=info:doi/10.1126/scirobotics.aav1488&rft_dat=%3Cproquest_cross%3E2457281585%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2457281585&rft_id=info:pmid/33137762&rfr_iscdi=true