No Cover Image

Journal article 747 views 123 downloads

Deep Recurrent Neural Networks Based Obstacle Avoidance Control for Redundant Manipulators

Zhihao Xu, Xuefeng Zhou, Shuai Li Orcid Logo

Frontiers in Neurorobotics, Volume: 13

Swansea University Author: Shuai Li Orcid Logo

  • xu2019(2).pdf

    PDF | Version of Record

    Released under the terms of a Creative Commons Attribution License (CC-BY).

    Download (1.17MB)

Check full text

DOI (Published version): 10.3389/fnbot.2019.00047

Abstract

Obstacle avoidance is an important subject in the control of robot manipulators, but is remains challenging for robots with redundant degrees of freedom, especially when there exist complex physical constraints. In this paper, we propose a novel controller based on deep recurrent neural networks. By...

Full description

Published in: Frontiers in Neurorobotics
ISSN: 1662-5218 1662-5218
Published: 2019
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa52000
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2019-09-23T14:18:30Z
last_indexed 2023-02-22T04:00:09Z
id cronfa52000
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2023-02-21T16:48:58.7299556</datestamp><bib-version>v2</bib-version><id>52000</id><entry>2019-09-23</entry><title>Deep Recurrent Neural Networks Based Obstacle Avoidance Control for Redundant Manipulators</title><swanseaauthors><author><sid>42ff9eed09bcd109fbbe484a0f99a8a8</sid><ORCID>0000-0001-8316-5289</ORCID><firstname>Shuai</firstname><surname>Li</surname><name>Shuai Li</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2019-09-23</date><deptcode>MECH</deptcode><abstract>Obstacle avoidance is an important subject in the control of robot manipulators, but is remains challenging for robots with redundant degrees of freedom, especially when there exist complex physical constraints. In this paper, we propose a novel controller based on deep recurrent neural networks. By abstracting robots and obstacles into critical point sets respectively, the distance between the robot and obstacles can be described in a simpler way, then the obstacle avoidance strategy is established in form of inequality constraints by general class-K functions. Using minimal-velocity-norm (MVN) scheme, the control problem is formulated as a quadratic-programming case under multiple constraints. Then a deep recurrent neural network considering system models is established to solve the QP problem online. Theoretical conduction and numerical simulations show that the controller is capable of avoiding static or dynamic obstacles, while tracking the predefined trajectories under physical constraints.</abstract><type>Journal Article</type><journal>Frontiers in Neurorobotics</journal><volume>13</volume><journalNumber/><paginationStart/><paginationEnd/><publisher/><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>1662-5218</issnPrint><issnElectronic>1662-5218</issnElectronic><keywords>recurrent neural network, redundant manipulator, obstacle avoidance, zeroing neural network, motion plan</keywords><publishedDay>4</publishedDay><publishedMonth>7</publishedMonth><publishedYear>2019</publishedYear><publishedDate>2019-07-04</publishedDate><doi>10.3389/fnbot.2019.00047</doi><url/><notes/><college>COLLEGE NANME</college><department>Mechanical Engineering</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>MECH</DepartmentCode><institution>Swansea University</institution><apcterm/><funders/><projectreference/><lastEdited>2023-02-21T16:48:58.7299556</lastEdited><Created>2019-09-23T11:43:30.6060571</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>Zhihao</firstname><surname>Xu</surname><order>1</order></author><author><firstname>Xuefeng</firstname><surname>Zhou</surname><order>2</order></author><author><firstname>Shuai</firstname><surname>Li</surname><orcid>0000-0001-8316-5289</orcid><order>3</order></author></authors><documents><document><filename>0052000-10102019114100.pdf</filename><originalFilename>xu2019(2).pdf</originalFilename><uploaded>2019-10-10T11:41:00.3870000</uploaded><type>Output</type><contentLength>1426793</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>Released under the terms of a Creative Commons Attribution License (CC-BY).</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>http://creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling 2023-02-21T16:48:58.7299556 v2 52000 2019-09-23 Deep Recurrent Neural Networks Based Obstacle Avoidance Control for Redundant Manipulators 42ff9eed09bcd109fbbe484a0f99a8a8 0000-0001-8316-5289 Shuai Li Shuai Li true false 2019-09-23 MECH Obstacle avoidance is an important subject in the control of robot manipulators, but is remains challenging for robots with redundant degrees of freedom, especially when there exist complex physical constraints. In this paper, we propose a novel controller based on deep recurrent neural networks. By abstracting robots and obstacles into critical point sets respectively, the distance between the robot and obstacles can be described in a simpler way, then the obstacle avoidance strategy is established in form of inequality constraints by general class-K functions. Using minimal-velocity-norm (MVN) scheme, the control problem is formulated as a quadratic-programming case under multiple constraints. Then a deep recurrent neural network considering system models is established to solve the QP problem online. Theoretical conduction and numerical simulations show that the controller is capable of avoiding static or dynamic obstacles, while tracking the predefined trajectories under physical constraints. Journal Article Frontiers in Neurorobotics 13 1662-5218 1662-5218 recurrent neural network, redundant manipulator, obstacle avoidance, zeroing neural network, motion plan 4 7 2019 2019-07-04 10.3389/fnbot.2019.00047 COLLEGE NANME Mechanical Engineering COLLEGE CODE MECH Swansea University 2023-02-21T16:48:58.7299556 2019-09-23T11:43:30.6060571 Faculty of Science and Engineering School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Mechanical Engineering Zhihao Xu 1 Xuefeng Zhou 2 Shuai Li 0000-0001-8316-5289 3 0052000-10102019114100.pdf xu2019(2).pdf 2019-10-10T11:41:00.3870000 Output 1426793 application/pdf Version of Record true Released under the terms of a Creative Commons Attribution License (CC-BY). true eng http://creativecommons.org/licenses/by/4.0/
title Deep Recurrent Neural Networks Based Obstacle Avoidance Control for Redundant Manipulators
spellingShingle Deep Recurrent Neural Networks Based Obstacle Avoidance Control for Redundant Manipulators
Shuai Li
title_short Deep Recurrent Neural Networks Based Obstacle Avoidance Control for Redundant Manipulators
title_full Deep Recurrent Neural Networks Based Obstacle Avoidance Control for Redundant Manipulators
title_fullStr Deep Recurrent Neural Networks Based Obstacle Avoidance Control for Redundant Manipulators
title_full_unstemmed Deep Recurrent Neural Networks Based Obstacle Avoidance Control for Redundant Manipulators
title_sort Deep Recurrent Neural Networks Based Obstacle Avoidance Control for Redundant Manipulators
author_id_str_mv 42ff9eed09bcd109fbbe484a0f99a8a8
author_id_fullname_str_mv 42ff9eed09bcd109fbbe484a0f99a8a8_***_Shuai Li
author Shuai Li
author2 Zhihao Xu
Xuefeng Zhou
Shuai Li
format Journal article
container_title Frontiers in Neurorobotics
container_volume 13
publishDate 2019
institution Swansea University
issn 1662-5218
1662-5218
doi_str_mv 10.3389/fnbot.2019.00047
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 Obstacle avoidance is an important subject in the control of robot manipulators, but is remains challenging for robots with redundant degrees of freedom, especially when there exist complex physical constraints. In this paper, we propose a novel controller based on deep recurrent neural networks. By abstracting robots and obstacles into critical point sets respectively, the distance between the robot and obstacles can be described in a simpler way, then the obstacle avoidance strategy is established in form of inequality constraints by general class-K functions. Using minimal-velocity-norm (MVN) scheme, the control problem is formulated as a quadratic-programming case under multiple constraints. Then a deep recurrent neural network considering system models is established to solve the QP problem online. Theoretical conduction and numerical simulations show that the controller is capable of avoiding static or dynamic obstacles, while tracking the predefined trajectories under physical constraints.
published_date 2019-07-04T04:04:07Z
_version_ 1763753325849739264
score 11.012678

Similar Items