Loading…
Learning to acquire whole-body humanoid CoM movements to achieve dynamic tasks
This paper presents a novel approach to acquire dynamic whole-body movements on humanoid robots focused on learning a control policy for the center of mass. A policy-gradient method is used to acquire a CoM movement as a control policy for achieving a desired dynamic task. A CoM-Jacobian-based redun...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Conference Proceeding |
| Language: | English |
| Subjects: | |
| Online Access: | Request full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| cited_by | |
|---|---|
| cites | |
| container_end_page | 2693 |
| container_issue | |
| container_start_page | 2688 |
| container_title | |
| container_volume | |
| creator | Matsubara, T. Morimoto, J. Nakanishi, J. Sang-Ho Hyon Hale, J.G. Cheng, G. |
| description | This paper presents a novel approach to acquire dynamic whole-body movements on humanoid robots focused on learning a control policy for the center of mass. A policy-gradient method is used to acquire a CoM movement as a control policy for achieving a desired dynamic task. A CoM-Jacobian-based redundancy resolution is then used to compute angular velocities for all joints in order to achieve a whole-body movement consistent with the CoM movement acquired through learning. To demonstrate the effectiveness of our method, we apply it in simulation to the learning of a strong punching movement on the Fujitsu humanoid robot, Hoap-2. |
| doi_str_mv | 10.1109/ROBOT.2007.363871 |
| format | conference_proceeding |
| fullrecord | <record><control><sourceid>ieee_6IE</sourceid><recordid>TN_cdi_ieee_primary_4209489</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>4209489</ieee_id><sourcerecordid>4209489</sourcerecordid><originalsourceid>FETCH-LOGICAL-i175t-87f20ec89d7282e0f3ee68843996a3e6b9af383ab14a9d709490b6a242facd9c3</originalsourceid><addsrcrecordid>eNpFzEtLw0AUBeDxBdbaHyBu5g-k3nlkHksNvqBakAruyk1yY0abRJO0kn9voIJncxbn4zB2IWAuBPirl-XNcjWXAHaujHJWHLAzoaXWYECKQzaRsbUROPt29D8IdcwmAmKItJX-lM267gPGxGDB-Al7XhC2dajfed9wzL63oSX-UzYbitImH3i5rbBuQs6T5olXzY4qqvtuj8tAO-L5UGMVMt5j99mds5MCNx3N_nrKXu9uV8lDtFjePybXiygIG_eRs4UEypzPrXSSoFBExjmtvDeoyKQeC-UUpkLjaMBrD6lBqWWBWe4zNWWX-99AROuvNlTYDmstR-m8-gWLjVLh</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Learning to acquire whole-body humanoid CoM movements to achieve dynamic tasks</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Matsubara, T. ; Morimoto, J. ; Nakanishi, J. ; Sang-Ho Hyon ; Hale, J.G. ; Cheng, G.</creator><creatorcontrib>Matsubara, T. ; Morimoto, J. ; Nakanishi, J. ; Sang-Ho Hyon ; Hale, J.G. ; Cheng, G.</creatorcontrib><description>This paper presents a novel approach to acquire dynamic whole-body movements on humanoid robots focused on learning a control policy for the center of mass. A policy-gradient method is used to acquire a CoM movement as a control policy for achieving a desired dynamic task. A CoM-Jacobian-based redundancy resolution is then used to compute angular velocities for all joints in order to achieve a whole-body movement consistent with the CoM movement acquired through learning. To demonstrate the effectiveness of our method, we apply it in simulation to the learning of a strong punching movement on the Fujitsu humanoid robot, Hoap-2.</description><identifier>ISSN: 1050-4729</identifier><identifier>ISBN: 1424406013</identifier><identifier>ISBN: 9781424406012</identifier><identifier>EISSN: 2577-087X</identifier><identifier>EISBN: 1424406021</identifier><identifier>EISBN: 9781424406029</identifier><identifier>DOI: 10.1109/ROBOT.2007.363871</identifier><language>eng</language><publisher>IEEE</publisher><subject>Angular velocity ; Computational modeling ; Equations ; Humanoid robot ; Humanoid robots ; Humans ; Learning ; Legged locomotion ; Orbital robotics ; Policy-gradient method ; Reinforcement learning ; Robotics and automation ; Weight control ; Whole-body movement</subject><ispartof>Proceedings 2007 IEEE International Conference on Robotics and Automation, 2007, p.2688-2693</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/4209489$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,2058,27925,54555,54920,54932</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/4209489$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Matsubara, T.</creatorcontrib><creatorcontrib>Morimoto, J.</creatorcontrib><creatorcontrib>Nakanishi, J.</creatorcontrib><creatorcontrib>Sang-Ho Hyon</creatorcontrib><creatorcontrib>Hale, J.G.</creatorcontrib><creatorcontrib>Cheng, G.</creatorcontrib><title>Learning to acquire whole-body humanoid CoM movements to achieve dynamic tasks</title><title>Proceedings 2007 IEEE International Conference on Robotics and Automation</title><addtitle>ROBOT</addtitle><description>This paper presents a novel approach to acquire dynamic whole-body movements on humanoid robots focused on learning a control policy for the center of mass. A policy-gradient method is used to acquire a CoM movement as a control policy for achieving a desired dynamic task. A CoM-Jacobian-based redundancy resolution is then used to compute angular velocities for all joints in order to achieve a whole-body movement consistent with the CoM movement acquired through learning. To demonstrate the effectiveness of our method, we apply it in simulation to the learning of a strong punching movement on the Fujitsu humanoid robot, Hoap-2.</description><subject>Angular velocity</subject><subject>Computational modeling</subject><subject>Equations</subject><subject>Humanoid robot</subject><subject>Humanoid robots</subject><subject>Humans</subject><subject>Learning</subject><subject>Legged locomotion</subject><subject>Orbital robotics</subject><subject>Policy-gradient method</subject><subject>Reinforcement learning</subject><subject>Robotics and automation</subject><subject>Weight control</subject><subject>Whole-body movement</subject><issn>1050-4729</issn><issn>2577-087X</issn><isbn>1424406013</isbn><isbn>9781424406012</isbn><isbn>1424406021</isbn><isbn>9781424406029</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2007</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNpFzEtLw0AUBeDxBdbaHyBu5g-k3nlkHksNvqBakAruyk1yY0abRJO0kn9voIJncxbn4zB2IWAuBPirl-XNcjWXAHaujHJWHLAzoaXWYECKQzaRsbUROPt29D8IdcwmAmKItJX-lM267gPGxGDB-Al7XhC2dajfed9wzL63oSX-UzYbitImH3i5rbBuQs6T5olXzY4qqvtuj8tAO-L5UGMVMt5j99mds5MCNx3N_nrKXu9uV8lDtFjePybXiygIG_eRs4UEypzPrXSSoFBExjmtvDeoyKQeC-UUpkLjaMBrD6lBqWWBWe4zNWWX-99AROuvNlTYDmstR-m8-gWLjVLh</recordid><startdate>200704</startdate><enddate>200704</enddate><creator>Matsubara, T.</creator><creator>Morimoto, J.</creator><creator>Nakanishi, J.</creator><creator>Sang-Ho Hyon</creator><creator>Hale, J.G.</creator><creator>Cheng, G.</creator><general>IEEE</general><scope>6IE</scope><scope>6IH</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIO</scope></search><sort><creationdate>200704</creationdate><title>Learning to acquire whole-body humanoid CoM movements to achieve dynamic tasks</title><author>Matsubara, T. ; Morimoto, J. ; Nakanishi, J. ; Sang-Ho Hyon ; Hale, J.G. ; Cheng, G.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i175t-87f20ec89d7282e0f3ee68843996a3e6b9af383ab14a9d709490b6a242facd9c3</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2007</creationdate><topic>Angular velocity</topic><topic>Computational modeling</topic><topic>Equations</topic><topic>Humanoid robot</topic><topic>Humanoid robots</topic><topic>Humans</topic><topic>Learning</topic><topic>Legged locomotion</topic><topic>Orbital robotics</topic><topic>Policy-gradient method</topic><topic>Reinforcement learning</topic><topic>Robotics and automation</topic><topic>Weight control</topic><topic>Whole-body movement</topic><toplevel>online_resources</toplevel><creatorcontrib>Matsubara, T.</creatorcontrib><creatorcontrib>Morimoto, J.</creatorcontrib><creatorcontrib>Nakanishi, J.</creatorcontrib><creatorcontrib>Sang-Ho Hyon</creatorcontrib><creatorcontrib>Hale, J.G.</creatorcontrib><creatorcontrib>Cheng, G.</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan (POP) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Electronic Library Online</collection><collection>IEEE Proceedings Order Plans (POP) 1998-present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Matsubara, T.</au><au>Morimoto, J.</au><au>Nakanishi, J.</au><au>Sang-Ho Hyon</au><au>Hale, J.G.</au><au>Cheng, G.</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Learning to acquire whole-body humanoid CoM movements to achieve dynamic tasks</atitle><btitle>Proceedings 2007 IEEE International Conference on Robotics and Automation</btitle><stitle>ROBOT</stitle><date>2007-04</date><risdate>2007</risdate><spage>2688</spage><epage>2693</epage><pages>2688-2693</pages><issn>1050-4729</issn><eissn>2577-087X</eissn><isbn>1424406013</isbn><isbn>9781424406012</isbn><eisbn>1424406021</eisbn><eisbn>9781424406029</eisbn><abstract>This paper presents a novel approach to acquire dynamic whole-body movements on humanoid robots focused on learning a control policy for the center of mass. A policy-gradient method is used to acquire a CoM movement as a control policy for achieving a desired dynamic task. A CoM-Jacobian-based redundancy resolution is then used to compute angular velocities for all joints in order to achieve a whole-body movement consistent with the CoM movement acquired through learning. To demonstrate the effectiveness of our method, we apply it in simulation to the learning of a strong punching movement on the Fujitsu humanoid robot, Hoap-2.</abstract><pub>IEEE</pub><doi>10.1109/ROBOT.2007.363871</doi><tpages>6</tpages></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 1050-4729 |
| ispartof | Proceedings 2007 IEEE International Conference on Robotics and Automation, 2007, p.2688-2693 |
| issn | 1050-4729 2577-087X |
| language | eng |
| recordid | cdi_ieee_primary_4209489 |
| source | IEEE Electronic Library (IEL) Conference Proceedings |
| subjects | Angular velocity Computational modeling Equations Humanoid robot Humanoid robots Humans Learning Legged locomotion Orbital robotics Policy-gradient method Reinforcement learning Robotics and automation Weight control Whole-body movement |
| title | Learning to acquire whole-body humanoid CoM movements to achieve dynamic tasks |
| url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-04T17%3A09%3A57IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_6IE&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Learning%20to%20acquire%20whole-body%20humanoid%20CoM%20movements%20to%20achieve%20dynamic%20tasks&rft.btitle=Proceedings%202007%20IEEE%20International%20Conference%20on%20Robotics%20and%20Automation&rft.au=Matsubara,%20T.&rft.date=2007-04&rft.spage=2688&rft.epage=2693&rft.pages=2688-2693&rft.issn=1050-4729&rft.eissn=2577-087X&rft.isbn=1424406013&rft.isbn_list=9781424406012&rft_id=info:doi/10.1109/ROBOT.2007.363871&rft.eisbn=1424406021&rft.eisbn_list=9781424406029&rft_dat=%3Cieee_6IE%3E4209489%3C/ieee_6IE%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i175t-87f20ec89d7282e0f3ee68843996a3e6b9af383ab14a9d709490b6a242facd9c3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=4209489&rfr_iscdi=true |