서브메뉴
검색
Large-Scale Real-World Robotic Manipulation Using Diverse Data- [electronic resource]
![Large-Scale Real-World Robotic Manipulation Using Diverse Data - [electronic resource]](/Users/Baul/Images/book.png)
Large-Scale Real-World Robotic Manipulation Using Diverse Data- [electronic resource]
상세정보
- 자료유형
- 학위논문파일 국외
- 최종처리일시
- 20240214095926
- ISBN
- 9798380367677
- DDC
- 629.8
- 서명/저자
- Large-Scale Real-World Robotic Manipulation Using Diverse Data - [electronic resource]
- 발행사항
- [S.l.]: : University of California, Berkeley., 2022
- 발행사항
- Ann Arbor : : ProQuest Dissertations & Theses,, 2022
- 형태사항
- 1 online resource(127 p.)
- 주기사항
- Source: Dissertations Abstracts International, Volume: 85-03, Section: B.
- 주기사항
- Advisor: Levine, Sergey;Finn, Chelsea.
- 학위논문주기
- Thesis (Ph.D.)--University of California, Berkeley, 2022.
- 사용제한주기
- This item must not be sold to any third party vendors.
- 초록/해제
- 요약Recent breakthroughs in computer vision and natural language processing have been largely propelled by scaling up both dataset diversity as well as model capacity, leading to robust generalization. In this thesis I am addressing the question of 1) whether for learning-based robotic manipulation we can similarly scale up dataset diversity and model capacity in order to achieve generalization and adaptation to new scenes and environments, new objects, new tasks and even different types of robots, and 2) the question of how re-collecting data from scratch for every new task and environment can be avoided, since this often leads to poor generalization and performance. To answer these questions we propose two different methodologies, a model-based reinforcement learning approach based on video-prediction, and a model-free and imitation-learning-based approach. We collect several of the biggest robotic interaction datasets to date, and show that by leveraging and effectively reusing diverse prior datasets, we can allow an agent to generalize to never-before-seen objects, learn new tasks based on only a handful of demonstrations, and even adapt to new robot types.
- 일반주제명
- Robotics.
- 일반주제명
- Computer science.
- 키워드
- Diverse data
- 기타저자
- University of California, Berkeley Electrical Engineering & Computer Sciences
- 기본자료저록
- Dissertations Abstracts International. 85-03B.
- 기본자료저록
- Dissertation Abstract International
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008240612s2022 us |||||||||||||||c||eng d■001000016931148
■00520240214095926
■006m o d
■007cr#unu||||||||
■020 ▼a9798380367677
■035 ▼a(MiAaPQ)AAI29327855
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a629.8
■1001 ▼aEbert, Frederik D.
■24510▼aLarge-Scale Real-World Robotic Manipulation Using Diverse Data▼h[electronic resource]
■260 ▼a[S.l.]:▼bUniversity of California, Berkeley. ▼c2022
■260 1▼aAnn Arbor :▼bProQuest Dissertations & Theses, ▼c2022
■300 ▼a1 online resource(127 p.)
■500 ▼aSource: Dissertations Abstracts International, Volume: 85-03, Section: B.
■500 ▼aAdvisor: Levine, Sergey;Finn, Chelsea.
■5021 ▼aThesis (Ph.D.)--University of California, Berkeley, 2022.
■506 ▼aThis item must not be sold to any third party vendors.
■520 ▼aRecent breakthroughs in computer vision and natural language processing have been largely propelled by scaling up both dataset diversity as well as model capacity, leading to robust generalization. In this thesis I am addressing the question of 1) whether for learning-based robotic manipulation we can similarly scale up dataset diversity and model capacity in order to achieve generalization and adaptation to new scenes and environments, new objects, new tasks and even different types of robots, and 2) the question of how re-collecting data from scratch for every new task and environment can be avoided, since this often leads to poor generalization and performance. To answer these questions we propose two different methodologies, a model-based reinforcement learning approach based on video-prediction, and a model-free and imitation-learning-based approach. We collect several of the biggest robotic interaction datasets to date, and show that by leveraging and effectively reusing diverse prior datasets, we can allow an agent to generalize to never-before-seen objects, learn new tasks based on only a handful of demonstrations, and even adapt to new robot types.
■590 ▼aSchool code: 0028.
■650 4▼aRobotics.
■650 4▼aComputer science.
■653 ▼aRobotic manipulation
■653 ▼aDiverse data
■653 ▼aNatural language processing
■653 ▼aLearning-based approach
■690 ▼a0771
■690 ▼a0984
■690 ▼a0800
■71020▼aUniversity of California, Berkeley▼bElectrical Engineering & Computer Sciences.
■7730 ▼tDissertations Abstracts International▼g85-03B.
■773 ▼tDissertation Abstract International
■790 ▼a0028
■791 ▼aPh.D.
■792 ▼a2022
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T16931148▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
■980 ▼a202402▼f2024
