본문

서브메뉴

Towards Human-Level Dexterity via Robot Learning
Towards Human-Level Dexterity via Robot Learning
Towards Human-Level Dexterity via Robot Learning

상세정보

자료유형  
 학위논문 서양
최종처리일시  
20250211153025
ISBN  
9798342747479
DDC  
629.8
저자명  
Khandate, Gagan Muralidhar.
서명/저자  
Towards Human-Level Dexterity via Robot Learning
발행사항  
[Sl] : Columbia University, 2024
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2024
형태사항  
132 p
주기사항  
Source: Dissertations Abstracts International, Volume: 86-05, Section: B.
주기사항  
Advisor: Ciocarlie, Matei.
학위논문주기  
Thesis (Ph.D.)--Columbia University, 2024.
초록/해제  
요약Dexterous intelligence-the ability to perform complex interactions with multi-fingered hands-is a pinnacle of human physical intelligence and emergent higher-order cognitive skills. However, contrary to Moravec's paradox, dexterous intelligence in humans appears simple only superficially. Many million years were spent co-evolving the human brain and hands including rich tactile sensing. Achieving human-level dexterity with robotic hands has long been a fundamental goal in robotics and represents a critical milestone toward general embodied intelligence. In this pursuit, computational sensorimotor learning has made significant progress, enabling feats such as arbitrary in-hand object reorientation. However, we observe that achieving higher levels of dexterity requires overcoming very fundamental limitations of computational sensorimotor learning.I develop robot learning methods for highly dexterous multi-fingered manipulation by directly addressing these limitations at their root cause. Chiefly, through key studies, this dissertation progressively builds an effective framework for reinforcement learning of dexterous multi-fingered manipulation skills. These methods adopt structured exploration, effectively overcoming the limitations of random exploration in reinforcement learning. The insights gained culminate in a highly effective reinforcement learning that incorporates sampling-based planning for direct exploration. Additionally, this thesis explores a new paradigm of using visuo-tactile human demonstrations for dexterity, introducing corresponding imitation learning techniques.
일반주제명  
Robotics
일반주제명  
Computer engineering
일반주제명  
Computer science
키워드  
Dexterous manipulation
키워드  
In-hand reorientation
키워드  
Reinforcement learning
키워드  
Robot learning
키워드  
Sampling-based planning
키워드  
Tactile sensing
기타저자  
Columbia University Computer Science
기본자료저록  
Dissertations Abstracts International. 86-05B.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008250123s2024        us                              c    eng  d
■001000017164633
■00520250211153025
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798342747479
■035    ▼a(MiAaPQ)AAI31633673
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a629.8
■1001  ▼aKhandate,  Gagan  Muralidhar.
■24510▼aTowards  Human-Level  Dexterity  via  Robot  Learning
■260    ▼a[Sl]▼bColumbia  University▼c2024
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2024
■300    ▼a132  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  86-05,  Section:  B.
■500    ▼aAdvisor:  Ciocarlie,  Matei.
■5021  ▼aThesis  (Ph.D.)--Columbia  University,  2024.
■520    ▼aDexterous  intelligence-the  ability  to  perform  complex  interactions  with  multi-fingered  hands-is  a  pinnacle  of  human  physical  intelligence  and  emergent  higher-order  cognitive  skills.  However,  contrary  to  Moravec's  paradox,  dexterous  intelligence  in  humans  appears  simple  only  superficially.  Many  million  years  were  spent  co-evolving  the  human  brain  and  hands  including  rich  tactile  sensing.  Achieving  human-level  dexterity  with  robotic  hands  has  long  been  a  fundamental  goal  in  robotics  and  represents  a  critical  milestone  toward  general  embodied  intelligence.  In  this  pursuit,  computational  sensorimotor  learning  has  made  significant  progress,  enabling  feats  such  as  arbitrary  in-hand  object  reorientation.  However,  we  observe  that  achieving  higher  levels  of  dexterity  requires  overcoming  very  fundamental  limitations  of  computational  sensorimotor  learning.I  develop  robot  learning  methods  for  highly  dexterous  multi-fingered  manipulation  by  directly  addressing  these  limitations  at  their  root  cause.  Chiefly,  through  key  studies,  this  dissertation  progressively  builds  an  effective  framework  for  reinforcement  learning  of  dexterous  multi-fingered  manipulation  skills.  These  methods  adopt  structured  exploration,  effectively  overcoming  the  limitations  of  random  exploration  in  reinforcement  learning.  The  insights  gained  culminate  in  a  highly  effective  reinforcement  learning  that  incorporates  sampling-based  planning  for  direct  exploration.  Additionally,  this  thesis  explores  a  new  paradigm  of  using  visuo-tactile  human  demonstrations  for  dexterity,  introducing  corresponding  imitation  learning  techniques.
■590    ▼aSchool  code:  0054.
■650  4▼aRobotics
■650  4▼aComputer  engineering
■650  4▼aComputer  science
■653    ▼aDexterous  manipulation
■653    ▼aIn-hand  reorientation
■653    ▼aReinforcement  learning
■653    ▼aRobot  learning
■653    ▼aSampling-based  planning
■653    ▼aTactile  sensing
■690    ▼a0771
■690    ▼a0800
■690    ▼a0984
■690    ▼a0464
■71020▼aColumbia  University▼bComputer  Science.
■7730  ▼tDissertations  Abstracts  International▼g86-05B.
■790    ▼a0054
■791    ▼aPh.D.
■792    ▼a2024
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17164633▼nKERIS▼z이  자료의  원문은  한국교육학술정보원에서  제공합니다.

미리보기

내보내기

chatGPT토론

Ai 추천 관련 도서


    신착도서 더보기
    최근 3년간 통계입니다.

    소장정보

    • 예약
    • 소재불명신고
    • 나의폴더
    • 우선정리요청
    • 비도서대출신청
    • 야간 도서대출신청
    소장자료
    등록번호 청구기호 소장처 대출가능여부 대출정보
    TF13772 전자도서 대출가능 마이폴더 부재도서신고 비도서대출신청 야간 도서대출신청

    * 대출중인 자료에 한하여 예약이 가능합니다. 예약을 원하시면 예약버튼을 클릭하십시오.

    해당 도서를 다른 이용자가 함께 대출한 도서

    관련 인기도서

    로그인 후 이용 가능합니다.