서브메뉴
검색
Multimodal Spatio-Semantic Perception and Team Performance Evaluation for Hierarchical Human-Robot Interaction in Dynamic Environments
Multimodal Spatio-Semantic Perception and Team Performance Evaluation for Hierarchical Human-Robot Interaction in Dynamic Environments
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260311091534.5
- ISBN
- 9798270231798
- DDC
- 302
- 서명/저자
- Multimodal Spatio-Semantic Perception and Team Performance Evaluation for Hierarchical Human-Robot Interaction in Dynamic Environments / John Alexander Duncan
- 발행사항
- [Sl] : The University of Texas at Austin, 2025
- 형태사항
- 1 electronic resource (143 pages)
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-06, Section: B.
- 주기사항
- Advisors: Pryor, Mitchell; Alambeigi, Farshid Committee members: Majewicz-Fey, Ann; Warnell, Garrett.
- 학위논문주기
- - Ph.D. : The University of Texas at Austin, 2025.
- 초록/해제
- 요약Robots are increasingly deployed alongside humans in dynamic environments. However, effectiveness in hierarchical teams-such as those in defense or disaster response-is hindered by a lack of specialized perception systems and "in-the-wild" teaming studies. This dissertation addresses these limitations through three primary research thrusts: 1) the design and evaluation of a multi-object tracking framework to provide robust spatial perception of nearby persons in real-time; 2) the development of a novel spatio-semantic multimodal fusion method to enable hierarchical interaction by simultaneously estimating human roles, commands, and positions; and 3) an empirical user study to evaluate the effects of human-robot team structure on task performance. The tracking evaluation identified key performance tradeoffs for deployment on robot hardware and demonstrated that augmenting onboard robot sensors with data from human-worn sensors unilaterally improves tracking accuracy. The spatio-semantic perception system proved viable for fusing hierarchical information, but requires more advanced modeling to accurately estimate temporal communications from speech and gesture. Finally, the user study confirmed t hat team structure is a significant factor on human-robot team performance, and that human-robot team performance is strongly mediated by team communication and strategy selection. Taken together, this dissertation demonstrates that effective hierarchical human-robot teaming should combine robust, context-aware spatio-semantic perception with an empirical understanding of team and environment dynamics. The findings provide a foundation and actionable insights for the design and deployment of human-robot teams in complex, realistic operating environments.
- 언어주기
- English
- 일반주제명
- Computer engineering
- 일반주제명
- Robotics
- 키워드
- Robots
- 기타저자
- The University of Texas at Austin Mechanical Engineering
- 기본자료저록
- Dissertations Abstracts International. 87-06B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260311s2025 us eng d■001000017361203
■00520260311091534.5
■006m o d
■007cr|nu||||||||
■020 ▼a9798270231798
■040 ▼aMiAaPQD▼beng▼cMiAaPQD▼erda
■082 ▼a302
■1001 ▼aDuncan, John Alexander▼eauthor.
■24510▼aMultimodal Spatio-Semantic Perception and Team Performance Evaluation for Hierarchical Human-Robot Interaction in Dynamic Environments ▼cJohn Alexander Duncan
■260 ▼a[Sl]▼bThe University of Texas at Austin▼c2025
■264 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a1 electronic resource (143 pages)
■336 ▼atext▼btxt▼2rdacontent
■337 ▼acomputer▼bc▼2rdamedia
■338 ▼aonline resource▼bcr▼2rdacarrier
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-06, Section: B.
■500 ▼aAdvisors: Pryor, Mitchell; Alambeigi, Farshid Committee members: Majewicz-Fey, Ann; Warnell, Garrett.
■5021 ▼bPh.D.▼cThe University of Texas at Austin▼d2025.
■520 ▼aRobots are increasingly deployed alongside humans in dynamic environments. However, effectiveness in hierarchical teams-such as those in defense or disaster response-is hindered by a lack of specialized perception systems and "in-the-wild" teaming studies. This dissertation addresses these limitations through three primary research thrusts: 1) the design and evaluation of a multi-object tracking framework to provide robust spatial perception of nearby persons in real-time; 2) the development of a novel spatio-semantic multimodal fusion method to enable hierarchical interaction by simultaneously estimating human roles, commands, and positions; and 3) an empirical user study to evaluate the effects of human-robot team structure on task performance. The tracking evaluation identified key performance tradeoffs for deployment on robot hardware and demonstrated that augmenting onboard robot sensors with data from human-worn sensors unilaterally improves tracking accuracy. The spatio-semantic perception system proved viable for fusing hierarchical information, but requires more advanced modeling to accurately estimate temporal communications from speech and gesture. Finally, the user study confirmed t hat team structure is a significant factor on human-robot team performance, and that human-robot team performance is strongly mediated by team communication and strategy selection. Taken together, this dissertation demonstrates that effective hierarchical human-robot teaming should combine robust, context-aware spatio-semantic perception with an empirical understanding of team and environment dynamics. The findings provide a foundation and actionable insights for the design and deployment of human-robot teams in complex, realistic operating environments.
■546 ▼aEnglish
■590 ▼aSchool code: 0227
■650 4▼aComputer engineering
■650 4▼aEnvironmental engineering
■650 4▼aRobotics
■653 ▼aRobots
■653 ▼aDisaster response
■653 ▼aHuman-robot teams
■653 ▼aTemporal communications
■653 ▼aHierarchical interaction
■7102 ▼aThe University of Texas at Austin▼bMechanical Engineering.▼edegree granting institution.
■7201 ▼aPryor, Mitchell▼edegree supervisor.
■7201 ▼aAlambeigi, Farshid ▼edegree supervisor.
■7730 ▼tDissertations Abstracts International▼g87-06B.
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17361203▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
