서브메뉴
검색
Robotics As Sensorimotor Sequence Modeling
Robotics As Sensorimotor Sequence Modeling
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202105103
- ISBN
- 9798297601536
- DDC
- 629.8
- 서명/저자
- Robotics As Sensorimotor Sequence Modeling
- 발행사항
- [Sl] : University of California, Berkeley, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 141 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-04, Section: B.
- 주기사항
- Advisor: Malik, Jitendra.
- 학위논문주기
- Thesis (Ph.D.)--University of California, Berkeley, 2025.
- 초록/해제
- 요약Today's robotics is marked by a diversity of approaches---we have different algorithms for different problems. This dissertation presents a single approach that can address multiple robotics problems. We view robotics through the lens of sequence modeling. We consider sequences of sensory observations and motor commands, interleaved over time. Our approach models this data using deep neural networks trained via sensorimotor sequence prediction and reinforcement learning. We address multiple robotic problems using this sensorimotor sequence modeling approach. In locomotion, we show humanoid robots walking over challenging terrain, including hiking in the Berkeley Hills and climbing the steepest streets in San Francisco. In manipulation, we show dexterous tasks, such as folding laundry with hands. Beyond these capabilities, we show that behaviors and representations emerge as a byproduct of learning.
- 일반주제명
- Robotics
- 일반주제명
- Computer science
- 일반주제명
- Electrical engineering
- 키워드
- Motor commands
- 키워드
- Robotic problems
- 기타저자
- University of California, Berkeley Electrical Engineering & Computer Sciences
- 기본자료저록
- Dissertations Abstracts International. 87-04B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017359333
■00520260202105103
■006m o d
■007cr#unu||||||||
■020 ▼a9798297601536
■035 ▼a(MiAaPQ)AAI32236374
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a629.8
■1001 ▼aRadosavovic, Ilija.
■24510▼aRobotics As Sensorimotor Sequence Modeling
■260 ▼a[Sl]▼bUniversity of California, Berkeley▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a141 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-04, Section: B.
■500 ▼aAdvisor: Malik, Jitendra.
■5021 ▼aThesis (Ph.D.)--University of California, Berkeley, 2025.
■520 ▼aToday's robotics is marked by a diversity of approaches---we have different algorithms for different problems. This dissertation presents a single approach that can address multiple robotics problems. We view robotics through the lens of sequence modeling. We consider sequences of sensory observations and motor commands, interleaved over time. Our approach models this data using deep neural networks trained via sensorimotor sequence prediction and reinforcement learning. We address multiple robotic problems using this sensorimotor sequence modeling approach. In locomotion, we show humanoid robots walking over challenging terrain, including hiking in the Berkeley Hills and climbing the steepest streets in San Francisco. In manipulation, we show dexterous tasks, such as folding laundry with hands. Beyond these capabilities, we show that behaviors and representations emerge as a byproduct of learning.
■590 ▼aSchool code: 0028.
■650 4▼aRobotics
■650 4▼aComputer science
■650 4▼aElectrical engineering
■653 ▼aSensorimotor sequence
■653 ▼aSensory observations
■653 ▼aMotor commands
■653 ▼aRobotic problems
■690 ▼a0800
■690 ▼a0771
■690 ▼a0984
■690 ▼a0544
■71020▼aUniversity of California, Berkeley▼bElectrical Engineering & Computer Sciences.
■7730 ▼tDissertations Abstracts International▼g87-04B.
■790 ▼a0028
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17359333▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
