Torque tracking position control of DLR-HIT II robotic hand using a real-time physics-informed neural network

Al-Shahrabi, Ali; JAHANBAKHSHJAVID, MASOUD; Abdo, Ashraf Fahmy; Griffiths, Christian; Li, Chunxu

doi:10.1016/j.apm.2025.116110

Journal article 551 views 161 downloads

Torque tracking position control of DLR-HIT II robotic hand using a real-time physics-informed neural network

Ali Al-Shahrabi, MASOUD JAHANBAKHSHJAVID, Ashraf Fahmy Abdo

, Christian Griffiths, Chunxu Li

Applied Mathematical Modelling, Volume: 145, Start page: 116110

Swansea University Authors: Ali Al-Shahrabi, MASOUD JAHANBAKHSHJAVID, Ashraf Fahmy Abdo , Christian Griffiths, Chunxu Li

PDF | Version of Record

© 2025 The Author(s). This is an open access article under the CC BY license.
Download (3.34MB)

Check full text

DOI (Published version): 10.1016/j.apm.2025.116110

Abstract

This paper presents a novel approach for controlling the DLR-HIT II robotic hand by leveraging physics-informed neural networks (PINNs) for torque and position control. This method eliminates the need for additional control inputs or external controllers, achieving high precision and simplified dyna...

Full description

Published in:	Applied Mathematical Modelling
ISSN:	0307-904X 1872-8480
Published:	Elsevier BV 2025
Online Access:	Check full text
URI:	https://cronfa.swan.ac.uk/Record/cronfa69140

first_indexed	2025-03-24T09:33:23Z
last_indexed	2025-04-10T06:17:20Z
id	cronfa69140
recordtype	SURis
fullrecord	<?xml version="1.0"?><rfc1807><datestamp>2025-04-09T15:35:29.6002190</datestamp><bib-version>v2</bib-version><id>69140</id><entry>2025-03-24</entry><title>Torque tracking position control of DLR-HIT II robotic hand using a real-time physics-informed neural network</title><swanseaauthors><author><sid>d01e641793738bce8cdd4778eea4f80c</sid><firstname>Ali</firstname><surname>Al-Shahrabi</surname><name>Ali Al-Shahrabi</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>232109ad0f98f5f0b730dd41ece5032b</sid><firstname>MASOUD</firstname><surname>JAHANBAKHSHJAVID</surname><name>MASOUD JAHANBAKHSHJAVID</name><active>true</active><ethesisStudent>true</ethesisStudent></author><author><sid>b952b837f8a8447055210d209892b427</sid><ORCID>0000-0003-1624-1725</ORCID><firstname>Ashraf</firstname><surname>Fahmy Abdo</surname><name>Ashraf Fahmy Abdo</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>84c202c256a2950fbc52314df6ec4914</sid><ORCID/><firstname>Christian</firstname><surname>Griffiths</surname><name>Christian Griffiths</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>e6ed70d02c25b05ab52340312559d684</sid><ORCID>0000-0001-7851-0260</ORCID><firstname>Chunxu</firstname><surname>Li</surname><name>Chunxu Li</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2025-03-24</date><abstract>This paper presents a novel approach for controlling the DLR-HIT II robotic hand by leveraging physics-informed neural networks (PINNs) for torque and position control. This method eliminates the need for additional control inputs or external controllers, achieving high precision and simplified dynamics, which is validated through extensive simulations that closely replicate experimental conditions, demonstrating the system's ability to handle external disturbances and maintain accurate trajectory tracking. The strategy only requires time and joint position data as inputs, allowing the network to compute velocity and acceleration internally. Time normalization enhances the model's ability to generalize across different time scales and ensures stable training. The method demonstrates strong generalization from a limited training set and successfully performs across diverse trajectory types. This simplification significantly reduces computational complexity and facilitates real-time control in advanced robotic applications.</abstract><type>Journal Article</type><journal>Applied Mathematical Modelling</journal><volume>145</volume><journalNumber/><paginationStart>116110</paginationStart><paginationEnd/><publisher>Elsevier BV</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>0307-904X</issnPrint><issnElectronic>1872-8480</issnElectronic><keywords>Machine Learning in Robotics; Computed torque control; Real-time Applications of Dexterous Manipulation; Neural Network-based Control; DLR-HIT II hand; Normalization method</keywords><publishedDay>1</publishedDay><publishedMonth>9</publishedMonth><publishedYear>2025</publishedYear><publishedDate>2025-09-01</publishedDate><doi>10.1016/j.apm.2025.116110</doi><url/><notes/><college>COLLEGE NANME</college><CollegeCode>COLLEGE CODE</CollegeCode><institution>Swansea University</institution><apcterm>SU Library paid the OA fee (TA Institutional Deal)</apcterm><funders>The authors express their sincere gratitude to the Iraqi Ministry of Higher Education for their support. Special thanks to Al-Nahrain University for the financial assistance that enabled this study. We also extend our appreciation to Swansea University, particularly to Professor Johann Sienz, Deputy Executive Dean of Science and Engineering, for providing essential facilities, workspace, and financial support. Additionally, we would like to thank Professor Dr. Dunhui Xiao and Associate Professor Dr. Rowan Brown for reviewing the mathematical aspects of our work and providing valuable feedback as experts in mathematics.</funders><projectreference/><lastEdited>2025-04-09T15:35:29.6002190</lastEdited><Created>2025-03-24T09:28:15.8394361</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Mechanical Engineering</level></path><authors><author><firstname>Ali</firstname><surname>Al-Shahrabi</surname><order>1</order></author><author><firstname>MASOUD</firstname><surname>JAHANBAKHSHJAVID</surname><order>2</order></author><author><firstname>Ashraf</firstname><surname>Fahmy Abdo</surname><orcid>0000-0003-1624-1725</orcid><order>3</order></author><author><firstname>Christian</firstname><surname>Griffiths</surname><orcid/><order>4</order></author><author><firstname>Chunxu</firstname><surname>Li</surname><orcid>0000-0001-7851-0260</orcid><order>5</order></author></authors><documents><document><filename>69140__33976__4899862b9fd94d53833f71c1769b638c.pdf</filename><originalFilename>69140.VoR.pdf</originalFilename><uploaded>2025-04-09T15:32:46.9054510</uploaded><type>Output</type><contentLength>3506273</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>© 2025 The Author(s). This is an open access article under the CC BY license.</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>http://creativecommons.org/licenses/by/4.0/).</licence></document></documents><OutputDurs/></rfc1807>
spelling	2025-04-09T15:35:29.6002190 v2 69140 2025-03-24 Torque tracking position control of DLR-HIT II robotic hand using a real-time physics-informed neural network d01e641793738bce8cdd4778eea4f80c Ali Al-Shahrabi Ali Al-Shahrabi true false 232109ad0f98f5f0b730dd41ece5032b MASOUD JAHANBAKHSHJAVID MASOUD JAHANBAKHSHJAVID true true b952b837f8a8447055210d209892b427 0000-0003-1624-1725 Ashraf Fahmy Abdo Ashraf Fahmy Abdo true false 84c202c256a2950fbc52314df6ec4914 Christian Griffiths Christian Griffiths true false e6ed70d02c25b05ab52340312559d684 0000-0001-7851-0260 Chunxu Li Chunxu Li true false 2025-03-24 This paper presents a novel approach for controlling the DLR-HIT II robotic hand by leveraging physics-informed neural networks (PINNs) for torque and position control. This method eliminates the need for additional control inputs or external controllers, achieving high precision and simplified dynamics, which is validated through extensive simulations that closely replicate experimental conditions, demonstrating the system's ability to handle external disturbances and maintain accurate trajectory tracking. The strategy only requires time and joint position data as inputs, allowing the network to compute velocity and acceleration internally. Time normalization enhances the model's ability to generalize across different time scales and ensures stable training. The method demonstrates strong generalization from a limited training set and successfully performs across diverse trajectory types. This simplification significantly reduces computational complexity and facilitates real-time control in advanced robotic applications. Journal Article Applied Mathematical Modelling 145 116110 Elsevier BV 0307-904X 1872-8480 Machine Learning in Robotics; Computed torque control; Real-time Applications of Dexterous Manipulation; Neural Network-based Control; DLR-HIT II hand; Normalization method 1 9 2025 2025-09-01 10.1016/j.apm.2025.116110 COLLEGE NANME COLLEGE CODE Swansea University SU Library paid the OA fee (TA Institutional Deal) The authors express their sincere gratitude to the Iraqi Ministry of Higher Education for their support. Special thanks to Al-Nahrain University for the financial assistance that enabled this study. We also extend our appreciation to Swansea University, particularly to Professor Johann Sienz, Deputy Executive Dean of Science and Engineering, for providing essential facilities, workspace, and financial support. Additionally, we would like to thank Professor Dr. Dunhui Xiao and Associate Professor Dr. Rowan Brown for reviewing the mathematical aspects of our work and providing valuable feedback as experts in mathematics. 2025-04-09T15:35:29.6002190 2025-03-24T09:28:15.8394361 Faculty of Science and Engineering School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Mechanical Engineering Ali Al-Shahrabi 1 MASOUD JAHANBAKHSHJAVID 2 Ashraf Fahmy Abdo 0000-0003-1624-1725 3 Christian Griffiths 4 Chunxu Li 0000-0001-7851-0260 5 69140__33976__4899862b9fd94d53833f71c1769b638c.pdf 69140.VoR.pdf 2025-04-09T15:32:46.9054510 Output 3506273 application/pdf Version of Record true © 2025 The Author(s). This is an open access article under the CC BY license. true eng http://creativecommons.org/licenses/by/4.0/).
title	Torque tracking position control of DLR-HIT II robotic hand using a real-time physics-informed neural network
spellingShingle	Torque tracking position control of DLR-HIT II robotic hand using a real-time physics-informed neural network Ali Al-Shahrabi MASOUD JAHANBAKHSHJAVID Ashraf Fahmy Abdo Christian Griffiths Chunxu Li
title_short	Torque tracking position control of DLR-HIT II robotic hand using a real-time physics-informed neural network
title_full	Torque tracking position control of DLR-HIT II robotic hand using a real-time physics-informed neural network
title_fullStr	Torque tracking position control of DLR-HIT II robotic hand using a real-time physics-informed neural network
title_full_unstemmed	Torque tracking position control of DLR-HIT II robotic hand using a real-time physics-informed neural network
title_sort	Torque tracking position control of DLR-HIT II robotic hand using a real-time physics-informed neural network
author_id_str_mv	d01e641793738bce8cdd4778eea4f80c 232109ad0f98f5f0b730dd41ece5032b b952b837f8a8447055210d209892b427 84c202c256a2950fbc52314df6ec4914 e6ed70d02c25b05ab52340312559d684
author_id_fullname_str_mv	d01e641793738bce8cdd4778eea4f80c_*_Ali Al-Shahrabi 232109ad0f98f5f0b730dd41ece5032b__MASOUD JAHANBAKHSHJAVID b952b837f8a8447055210d209892b427__Ashraf Fahmy Abdo 84c202c256a2950fbc52314df6ec4914__Christian Griffiths e6ed70d02c25b05ab52340312559d684_**_Chunxu Li
author	Ali Al-Shahrabi MASOUD JAHANBAKHSHJAVID Ashraf Fahmy Abdo Christian Griffiths Chunxu Li
author2	Ali Al-Shahrabi MASOUD JAHANBAKHSHJAVID Ashraf Fahmy Abdo Christian Griffiths Chunxu Li
format	Journal article
container_title	Applied Mathematical Modelling
container_volume	145
container_start_page	116110
publishDate	2025
institution	Swansea University
issn	0307-904X 1872-8480
doi_str_mv	10.1016/j.apm.2025.116110
publisher	Elsevier BV
college_str	Faculty of Science and Engineering
hierarchytype
hierarchy_top_id	facultyofscienceandengineering
hierarchy_top_title	Faculty of Science and Engineering
hierarchy_parent_id	facultyofscienceandengineering
hierarchy_parent_title	Faculty of Science and Engineering
department_str	School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Mechanical Engineering{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Mechanical Engineering
document_store_str	1
active_str	0
description	This paper presents a novel approach for controlling the DLR-HIT II robotic hand by leveraging physics-informed neural networks (PINNs) for torque and position control. This method eliminates the need for additional control inputs or external controllers, achieving high precision and simplified dynamics, which is validated through extensive simulations that closely replicate experimental conditions, demonstrating the system's ability to handle external disturbances and maintain accurate trajectory tracking. The strategy only requires time and joint position data as inputs, allowing the network to compute velocity and acceleration internally. Time normalization enhances the model's ability to generalize across different time scales and ensures stable training. The method demonstrates strong generalization from a limited training set and successfully performs across diverse trajectory types. This simplification significantly reduces computational complexity and facilitates real-time control in advanced robotic applications.
published_date	2025-09-01T05:23:09Z
_version_	1851641130770235392
score	11.089905

Torque tracking position control of DLR-HIT II robotic hand using a real-time physics-informed neural network

Similar Items