본문

서브메뉴

Towards a Comprehensive Benchmark for Embodied AI and Robotics
Towards a Comprehensive Benchmark for Embodied AI and Robotics
Towards a Comprehensive Benchmark for Embodied AI and Robotics

상세정보

자료유형  
 학위논문 서양
최종처리일시  
20260202104856
ISBN  
9798288814785
DDC  
530
저자명  
Li, Chengshu.
서명/저자  
Towards a Comprehensive Benchmark for Embodied AI and Robotics
발행사항  
[Sl] : Stanford University, 2025
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2025
형태사항  
314 p
주기사항  
Source: Dissertations Abstracts International, Volume: 87-02, Section: B.
주기사항  
Advisor: Li, Fei-Fei.
학위논문주기  
Thesis (Ph.D.)--Stanford University, 2025.
초록/해제  
요약This dissertation explores two key research directions toward enabling general-purpose embodied agents: the development of realistic, large-scale benchmarks and environments, and the design of learning frameworks-particularly action space representations-that support efficient policy learning for long-horizon mobile manipulation tasks. The first line of work establishes a closed-loop ecosystem for benchmarking and training embodied agents. Beginning with iGibson 1.0 and 2.0, we develop physically interactive 3D simulation platforms capable of supporting complex object interactions in realistic household environments. Building on these foundations, we introduce the BEHAVIOR and BEHAVIOR-1K benchmarks-comprising 100 and 1,000 everyday household activities, respectively-grounded in human time-use data, defined using a flexible logic-based language, and supported by human VR demonstrations. To enable scalable data-driven policy training, we propose MoMaGen, a demonstration generation method that synthesizes thousands of diverse trajectories from a single human demonstration. The second line of work investigates action space design as a source of inductive bias for solving long-horizon robotic tasks. We first present HRL4IN, a hierarchical reinforcement learning approach that decomposes interactive navigation via high-level end-effector goals. We then introduce ReLMoGen, a hybrid method that combines high-level exploration in spatial goal spaces with low-level motion generation for efficient execution. Finally, Chain of Code leverages large language models (LLMs) to generate executable code and pseudocode, enabling agents to blend algorithmic reasoning and commonsense inference for task completion. Together, these contributions advance the goal of building physically capable, semantically grounded, and human-aligned embodied agents.
일반주제명  
Physics
일반주제명  
Virtual reality
일반주제명  
Benchmarks
일반주제명  
Robotics
키워드  
Large language models
키워드  
Reinforcement learning approach
키워드  
Robotic tasks
기타저자  
Stanford University.
기본자료저록  
Dissertations Abstracts International. 87-02B.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008260126s2025        us                              c    eng  d
■001000017359257
■00520260202104856
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798288814785
■035    ▼a(MiAaPQ)AAI32201019
■035    ▼a(MiAaPQ)Stanfordwz861bd4280
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a530
■1001  ▼aLi,  Chengshu.
■24510▼aTowards  a  Comprehensive  Benchmark  for  Embodied  AI  and  Robotics
■260    ▼a[Sl]▼bStanford  University▼c2025
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2025
■300    ▼a314  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  87-02,  Section:  B.
■500    ▼aAdvisor:  Li,  Fei-Fei.
■5021  ▼aThesis  (Ph.D.)--Stanford  University,  2025.
■520    ▼aThis  dissertation  explores  two  key  research  directions  toward  enabling  general-purpose  embodied  agents:  the  development  of  realistic,  large-scale  benchmarks  and  environments,  and  the  design  of  learning  frameworks-particularly  action  space  representations-that  support  efficient  policy  learning  for  long-horizon  mobile  manipulation  tasks.  The  first  line  of  work  establishes  a  closed-loop  ecosystem  for  benchmarking  and  training  embodied  agents.  Beginning  with  iGibson  1.0  and  2.0,  we  develop  physically  interactive  3D  simulation  platforms  capable  of  supporting  complex  object  interactions  in  realistic  household  environments.  Building  on  these  foundations,  we  introduce  the  BEHAVIOR  and  BEHAVIOR-1K  benchmarks-comprising  100  and  1,000  everyday  household  activities,  respectively-grounded  in  human  time-use  data,  defined  using  a  flexible  logic-based  language,  and  supported  by  human  VR  demonstrations.  To  enable  scalable  data-driven  policy  training,  we  propose  MoMaGen,  a  demonstration  generation  method  that  synthesizes  thousands  of  diverse  trajectories  from  a  single  human  demonstration.  The  second  line  of  work  investigates  action  space  design  as  a  source  of  inductive  bias  for  solving  long-horizon  robotic  tasks.  We  first  present  HRL4IN,  a  hierarchical  reinforcement  learning  approach  that  decomposes  interactive  navigation  via  high-level  end-effector  goals.  We  then  introduce  ReLMoGen,  a  hybrid  method  that  combines  high-level  exploration  in  spatial  goal  spaces  with  low-level  motion  generation  for  efficient  execution.  Finally,  Chain  of  Code  leverages  large  language  models  (LLMs)  to  generate  executable  code  and  pseudocode,  enabling  agents  to  blend  algorithmic  reasoning  and  commonsense  inference  for  task  completion.  Together,  these  contributions  advance  the  goal  of  building  physically  capable,  semantically  grounded,  and  human-aligned  embodied  agents.
■590    ▼aSchool  code:  0212.
■650  4▼aPhysics
■650  4▼aVirtual  reality
■650  4▼aBenchmarks
■650  4▼aRobotics
■653    ▼aLarge  language  models
■653    ▼aReinforcement  learning  approach
■653    ▼aRobotic  tasks
■690    ▼a0605
■690    ▼a0800
■690    ▼a0771
■71020▼aStanford  University.
■7730  ▼tDissertations  Abstracts  International▼g87-02B.
■790    ▼a0212
■791    ▼aPh.D.
■792    ▼a2025
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17359257▼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
    TF15065 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.