Mechanical arm control method and device based on hybrid reinforcement learning and underwater robot

The invention relates to a mechanical arm control method and device based on hybrid reinforcement learning and an underwater robot, and belongs to the technical field of mechanical arm control. A positioning error obtained by making a difference between the expected operation pose and the current op...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: SUN ZHUO, ZHANG LIANYING, JING HUIXIANG, DU HONGWEI, JIAO RUOHONG, LI GUANGHUA, HOU DONGDONG, YAO BAOTAI, WANG KAI, LI ZHITAO, CAI JINSI
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator SUN ZHUO
ZHANG LIANYING
JING HUIXIANG
DU HONGWEI
JIAO RUOHONG
LI GUANGHUA
HOU DONGDONG
YAO BAOTAI
WANG KAI
LI ZHITAO
CAI JINSI
description The invention relates to a mechanical arm control method and device based on hybrid reinforcement learning and an underwater robot, and belongs to the technical field of mechanical arm control. A positioning error obtained by making a difference between the expected operation pose and the current operation pose of the mechanical arm serves as the input of a controller based on hybrid reinforcement learning, a control instruction serves as the output of the controller, and closed-loop control is conducted on the mechanical arm through the controller; the mixed reinforcement learning comprises construction of a reward function calculation model, a reward function comprises a position reward item and further comprises a posture reward item, and when the distance between the tail end of the current mechanical arm after action and the tail end of the expected mechanical arm is smaller than or equal to a position error set value, the posture reward item is rewarded; the closer the unit attitude quantity at the tail
format Patent
fullrecord <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_CN118181276A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>CN118181276A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_CN118181276A3</originalsourceid><addsrcrecordid>eNqNizEKwkAQRdNYiHqH8QAWUVBbCYqNVvZhsvM1C5uZMFkVb6-IB5BXvOa9cSEnhJY1Bk7E3lEwzW6JOuTWhFiFBI8YQA0PEDKl9tV4FHJEvZoHdNBMCewa9fY97irwJ2c4uTWWp8XoymnA7OdJMT_sL9Vxgd5qDD0HKHJdncty-2G5We9W_zRv3D0-9A</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Mechanical arm control method and device based on hybrid reinforcement learning and underwater robot</title><source>esp@cenet</source><creator>SUN ZHUO ; ZHANG LIANYING ; JING HUIXIANG ; DU HONGWEI ; JIAO RUOHONG ; LI GUANGHUA ; HOU DONGDONG ; YAO BAOTAI ; WANG KAI ; LI ZHITAO ; CAI JINSI</creator><creatorcontrib>SUN ZHUO ; ZHANG LIANYING ; JING HUIXIANG ; DU HONGWEI ; JIAO RUOHONG ; LI GUANGHUA ; HOU DONGDONG ; YAO BAOTAI ; WANG KAI ; LI ZHITAO ; CAI JINSI</creatorcontrib><description>The invention relates to a mechanical arm control method and device based on hybrid reinforcement learning and an underwater robot, and belongs to the technical field of mechanical arm control. A positioning error obtained by making a difference between the expected operation pose and the current operation pose of the mechanical arm serves as the input of a controller based on hybrid reinforcement learning, a control instruction serves as the output of the controller, and closed-loop control is conducted on the mechanical arm through the controller; the mixed reinforcement learning comprises construction of a reward function calculation model, a reward function comprises a position reward item and further comprises a posture reward item, and when the distance between the tail end of the current mechanical arm after action and the tail end of the expected mechanical arm is smaller than or equal to a position error set value, the posture reward item is rewarded; the closer the unit attitude quantity at the tail</description><language>chi ; eng</language><subject>CHAMBERS PROVIDED WITH MANIPULATION DEVICES ; EQUIPMENT FOR DWELLING OR WORKING UNDER WATER ; HAND TOOLS ; LAUNCHING, HAULING-OUT, OR DRY-DOCKING OF VESSELS ; LIFE-SAVING IN WATER ; MANIPULATORS ; MEANS FOR SALVAGING OR SEARCHING FOR UNDERWATER OBJECTS ; PERFORMING OPERATIONS ; PORTABLE POWER-DRIVEN TOOLS ; RELATED EQUIPMENT ; SHIPS OR OTHER WATERBORNE VESSELS ; TRANSPORTING</subject><creationdate>2024</creationdate><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&amp;date=20240614&amp;DB=EPODOC&amp;CC=CN&amp;NR=118181276A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,780,885,25564,76547</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&amp;date=20240614&amp;DB=EPODOC&amp;CC=CN&amp;NR=118181276A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>SUN ZHUO</creatorcontrib><creatorcontrib>ZHANG LIANYING</creatorcontrib><creatorcontrib>JING HUIXIANG</creatorcontrib><creatorcontrib>DU HONGWEI</creatorcontrib><creatorcontrib>JIAO RUOHONG</creatorcontrib><creatorcontrib>LI GUANGHUA</creatorcontrib><creatorcontrib>HOU DONGDONG</creatorcontrib><creatorcontrib>YAO BAOTAI</creatorcontrib><creatorcontrib>WANG KAI</creatorcontrib><creatorcontrib>LI ZHITAO</creatorcontrib><creatorcontrib>CAI JINSI</creatorcontrib><title>Mechanical arm control method and device based on hybrid reinforcement learning and underwater robot</title><description>The invention relates to a mechanical arm control method and device based on hybrid reinforcement learning and an underwater robot, and belongs to the technical field of mechanical arm control. A positioning error obtained by making a difference between the expected operation pose and the current operation pose of the mechanical arm serves as the input of a controller based on hybrid reinforcement learning, a control instruction serves as the output of the controller, and closed-loop control is conducted on the mechanical arm through the controller; the mixed reinforcement learning comprises construction of a reward function calculation model, a reward function comprises a position reward item and further comprises a posture reward item, and when the distance between the tail end of the current mechanical arm after action and the tail end of the expected mechanical arm is smaller than or equal to a position error set value, the posture reward item is rewarded; the closer the unit attitude quantity at the tail</description><subject>CHAMBERS PROVIDED WITH MANIPULATION DEVICES</subject><subject>EQUIPMENT FOR DWELLING OR WORKING UNDER WATER</subject><subject>HAND TOOLS</subject><subject>LAUNCHING, HAULING-OUT, OR DRY-DOCKING OF VESSELS</subject><subject>LIFE-SAVING IN WATER</subject><subject>MANIPULATORS</subject><subject>MEANS FOR SALVAGING OR SEARCHING FOR UNDERWATER OBJECTS</subject><subject>PERFORMING OPERATIONS</subject><subject>PORTABLE POWER-DRIVEN TOOLS</subject><subject>RELATED EQUIPMENT</subject><subject>SHIPS OR OTHER WATERBORNE VESSELS</subject><subject>TRANSPORTING</subject><fulltext>true</fulltext><rsrctype>patent</rsrctype><creationdate>2024</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNqNizEKwkAQRdNYiHqH8QAWUVBbCYqNVvZhsvM1C5uZMFkVb6-IB5BXvOa9cSEnhJY1Bk7E3lEwzW6JOuTWhFiFBI8YQA0PEDKl9tV4FHJEvZoHdNBMCewa9fY97irwJ2c4uTWWp8XoymnA7OdJMT_sL9Vxgd5qDD0HKHJdncty-2G5We9W_zRv3D0-9A</recordid><startdate>20240614</startdate><enddate>20240614</enddate><creator>SUN ZHUO</creator><creator>ZHANG LIANYING</creator><creator>JING HUIXIANG</creator><creator>DU HONGWEI</creator><creator>JIAO RUOHONG</creator><creator>LI GUANGHUA</creator><creator>HOU DONGDONG</creator><creator>YAO BAOTAI</creator><creator>WANG KAI</creator><creator>LI ZHITAO</creator><creator>CAI JINSI</creator><scope>EVB</scope></search><sort><creationdate>20240614</creationdate><title>Mechanical arm control method and device based on hybrid reinforcement learning and underwater robot</title><author>SUN ZHUO ; ZHANG LIANYING ; JING HUIXIANG ; DU HONGWEI ; JIAO RUOHONG ; LI GUANGHUA ; HOU DONGDONG ; YAO BAOTAI ; WANG KAI ; LI ZHITAO ; CAI JINSI</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_CN118181276A3</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>chi ; eng</language><creationdate>2024</creationdate><topic>CHAMBERS PROVIDED WITH MANIPULATION DEVICES</topic><topic>EQUIPMENT FOR DWELLING OR WORKING UNDER WATER</topic><topic>HAND TOOLS</topic><topic>LAUNCHING, HAULING-OUT, OR DRY-DOCKING OF VESSELS</topic><topic>LIFE-SAVING IN WATER</topic><topic>MANIPULATORS</topic><topic>MEANS FOR SALVAGING OR SEARCHING FOR UNDERWATER OBJECTS</topic><topic>PERFORMING OPERATIONS</topic><topic>PORTABLE POWER-DRIVEN TOOLS</topic><topic>RELATED EQUIPMENT</topic><topic>SHIPS OR OTHER WATERBORNE VESSELS</topic><topic>TRANSPORTING</topic><toplevel>online_resources</toplevel><creatorcontrib>SUN ZHUO</creatorcontrib><creatorcontrib>ZHANG LIANYING</creatorcontrib><creatorcontrib>JING HUIXIANG</creatorcontrib><creatorcontrib>DU HONGWEI</creatorcontrib><creatorcontrib>JIAO RUOHONG</creatorcontrib><creatorcontrib>LI GUANGHUA</creatorcontrib><creatorcontrib>HOU DONGDONG</creatorcontrib><creatorcontrib>YAO BAOTAI</creatorcontrib><creatorcontrib>WANG KAI</creatorcontrib><creatorcontrib>LI ZHITAO</creatorcontrib><creatorcontrib>CAI JINSI</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>SUN ZHUO</au><au>ZHANG LIANYING</au><au>JING HUIXIANG</au><au>DU HONGWEI</au><au>JIAO RUOHONG</au><au>LI GUANGHUA</au><au>HOU DONGDONG</au><au>YAO BAOTAI</au><au>WANG KAI</au><au>LI ZHITAO</au><au>CAI JINSI</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Mechanical arm control method and device based on hybrid reinforcement learning and underwater robot</title><date>2024-06-14</date><risdate>2024</risdate><abstract>The invention relates to a mechanical arm control method and device based on hybrid reinforcement learning and an underwater robot, and belongs to the technical field of mechanical arm control. A positioning error obtained by making a difference between the expected operation pose and the current operation pose of the mechanical arm serves as the input of a controller based on hybrid reinforcement learning, a control instruction serves as the output of the controller, and closed-loop control is conducted on the mechanical arm through the controller; the mixed reinforcement learning comprises construction of a reward function calculation model, a reward function comprises a position reward item and further comprises a posture reward item, and when the distance between the tail end of the current mechanical arm after action and the tail end of the expected mechanical arm is smaller than or equal to a position error set value, the posture reward item is rewarded; the closer the unit attitude quantity at the tail</abstract><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier
ispartof
issn
language chi ; eng
recordid cdi_epo_espacenet_CN118181276A
source esp@cenet
subjects CHAMBERS PROVIDED WITH MANIPULATION DEVICES
EQUIPMENT FOR DWELLING OR WORKING UNDER WATER
HAND TOOLS
LAUNCHING, HAULING-OUT, OR DRY-DOCKING OF VESSELS
LIFE-SAVING IN WATER
MANIPULATORS
MEANS FOR SALVAGING OR SEARCHING FOR UNDERWATER OBJECTS
PERFORMING OPERATIONS
PORTABLE POWER-DRIVEN TOOLS
RELATED EQUIPMENT
SHIPS OR OTHER WATERBORNE VESSELS
TRANSPORTING
title Mechanical arm control method and device based on hybrid reinforcement learning and underwater robot
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-23T16%3A37%3A06IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-epo_EVB&rft_val_fmt=info:ofi/fmt:kev:mtx:patent&rft.genre=patent&rft.au=SUN%20ZHUO&rft.date=2024-06-14&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3ECN118181276A%3C/epo_EVB%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true