서브메뉴
검색
Graph Optimization and Dual Quaternions for Spacecraft Autonomy During Close Proximity Operations
Graph Optimization and Dual Quaternions for Spacecraft Autonomy During Close Proximity Operations
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260209102904
- ISBN
- 9798265400222
- DDC
- 629.13
- 서명/저자
- Graph Optimization and Dual Quaternions for Spacecraft Autonomy During Close Proximity Operations
- 발행사항
- [Sl] : Georgia Institute of Technology, 2023
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2023
- 형태사항
- 168 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-05, Section: B.
- 주기사항
- Advisor: Tsiotras, Panagiotis;Dellaert, Frank.
- 학위논문주기
- Thesis (Ph.D.)--Georgia Institute of Technology, 2023.
- 초록/해제
- 요약Over the last decade, the space servicing industry has become an increasingly profitable and tantalizing domain for both the private and public market sectors. For instance, the on-orbit satellite servicing market size is projected to grow from $2.4 billion to $5.1 billion from 2023 to 2030, pushing the demand for dedicated satellite servicing Mission Extension Vehicles (MEVs) and projects. As the scope and ambitions of spacecraft servicing missions grow beyond basic module-exchanging servicers for both lower Earth and geostationary orbits, the need to enable the next generation of robotic servicing technology has become increasingly relevant.Although the mechanical, electrical, dexterous, and sensing capabilities of MEVs have reached a technology readiness level for appropriately servicing both new and heritage flightproven architectures during proximity operations, the outstanding needed core infrastructure for bridging the interactions between all of these different components is that of machine intelligence or autonomy.In this dissertation, graph optimization and dual quaternion modeling and control techniques for enabling the next generation of spacecraft autonomy for future space missions during close proximity operations are investigated. Specifically, this work contributes to enabling next-generation spacecraft autonomy by advancing two key areas of intelligent machine decision-making for space systems: a) task performance, or the capacity to model and control one's self to execute predefined tasks or goals, and b) navigation, or the capacity to model, localize, and navigate environments.The first main contribution of this dissertation is to show that dual quaternions enable task performance for multibody systems via a compact 6-Degree-Of-Freedom (DOF) formulation of position and attitude or pose for modeling and control of both ground-base and SpacecraftMounted Robotic Systems (SMRSs). In particular, this thesis presents results for both a multibody robust hybrid global dual quaternion controller for simultaneous pose-tracking of a spacecraft base and end-effector of SMRSs, along with a novel allocation technique to mitigate problems such as system singularities in a simulation environment, and hardware verification of dual quaternion kinematics on a ground-based manipulator system in the Dynamics and Control Systems Laboratory (DCSL).The other significant contribution of this thesis consists of work about graph-based optimization, notably, a deterministic approach, i.e., A* search, and a probabilistic factor graph optimization technique, for online trajectory generation, optimal control, state estimation, collision avoidance, and object detection for spacecraft navigation. More specifically, leveraging the dual quaternion algebra with the A* search algorithm, an attitude-constrained, collision-avoiding, path-planning approach for 6-DOF spacecraft navigation in an environment with moving objects is developed and verified in simulation. Additionally, we present a new algorithm for spacecraft navigation, Simultaneous Control And Trajectory Estimation (SCATE), which concurrently solves probabilistic optimal control, and sensor fusion for state estimation and object localization problems using a factor graph-based optimization approach. The SCATE algorithm's flexibility for constrained motion planning is demonstrated online to solve an optimal control problem constrained by two-norm minimized control input threshold-limiting, attitude-pointing, collision avoidance, and waypoint navigation, as well as the algorithm's capacity for object localization and state estimation of spacecraft via air-bearing robotic platform, with their respective onboard sensors, in the DCSL.
- 일반주제명
- Aeronautics
- 일반주제명
- Kinematics
- 일반주제명
- Optimization techniques
- 일반주제명
- Planning
- 일반주제명
- Graph representations
- 일반주제명
- Robots
- 일반주제명
- Attitudes
- 일반주제명
- Visualization
- 일반주제명
- Robotics
- 일반주제명
- Aerospace engineering
- 일반주제명
- Industrial engineering
- 기본자료저록
- Dissertations Abstracts International. 87-05B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260203s2023 us c eng d■001000017365963
■00520260209102904
■006m o d
■007cr#unu||||||||
■020 ▼a9798265400222
■035 ▼a(MiAaPQ)AAI32315602
■035 ▼a(MiAaPQ)GeorgiaTech73138
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a629.13
■1001 ▼aKing-Smith, Matthew.
■24510▼aGraph Optimization and Dual Quaternions for Spacecraft Autonomy During Close Proximity Operations
■260 ▼a[Sl]▼bGeorgia Institute of Technology▼c2023
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2023
■300 ▼a168 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-05, Section: B.
■500 ▼aAdvisor: Tsiotras, Panagiotis;Dellaert, Frank.
■5021 ▼aThesis (Ph.D.)--Georgia Institute of Technology, 2023.
■520 ▼aOver the last decade, the space servicing industry has become an increasingly profitable and tantalizing domain for both the private and public market sectors. For instance, the on-orbit satellite servicing market size is projected to grow from $2.4 billion to $5.1 billion from 2023 to 2030, pushing the demand for dedicated satellite servicing Mission Extension Vehicles (MEVs) and projects. As the scope and ambitions of spacecraft servicing missions grow beyond basic module-exchanging servicers for both lower Earth and geostationary orbits, the need to enable the next generation of robotic servicing technology has become increasingly relevant.Although the mechanical, electrical, dexterous, and sensing capabilities of MEVs have reached a technology readiness level for appropriately servicing both new and heritage flightproven architectures during proximity operations, the outstanding needed core infrastructure for bridging the interactions between all of these different components is that of machine intelligence or autonomy.In this dissertation, graph optimization and dual quaternion modeling and control techniques for enabling the next generation of spacecraft autonomy for future space missions during close proximity operations are investigated. Specifically, this work contributes to enabling next-generation spacecraft autonomy by advancing two key areas of intelligent machine decision-making for space systems: a) task performance, or the capacity to model and control one's self to execute predefined tasks or goals, and b) navigation, or the capacity to model, localize, and navigate environments.The first main contribution of this dissertation is to show that dual quaternions enable task performance for multibody systems via a compact 6-Degree-Of-Freedom (DOF) formulation of position and attitude or pose for modeling and control of both ground-base and SpacecraftMounted Robotic Systems (SMRSs). In particular, this thesis presents results for both a multibody robust hybrid global dual quaternion controller for simultaneous pose-tracking of a spacecraft base and end-effector of SMRSs, along with a novel allocation technique to mitigate problems such as system singularities in a simulation environment, and hardware verification of dual quaternion kinematics on a ground-based manipulator system in the Dynamics and Control Systems Laboratory (DCSL).The other significant contribution of this thesis consists of work about graph-based optimization, notably, a deterministic approach, i.e., A* search, and a probabilistic factor graph optimization technique, for online trajectory generation, optimal control, state estimation, collision avoidance, and object detection for spacecraft navigation. More specifically, leveraging the dual quaternion algebra with the A* search algorithm, an attitude-constrained, collision-avoiding, path-planning approach for 6-DOF spacecraft navigation in an environment with moving objects is developed and verified in simulation. Additionally, we present a new algorithm for spacecraft navigation, Simultaneous Control And Trajectory Estimation (SCATE), which concurrently solves probabilistic optimal control, and sensor fusion for state estimation and object localization problems using a factor graph-based optimization approach. The SCATE algorithm's flexibility for constrained motion planning is demonstrated online to solve an optimal control problem constrained by two-norm minimized control input threshold-limiting, attitude-pointing, collision avoidance, and waypoint navigation, as well as the algorithm's capacity for object localization and state estimation of spacecraft via air-bearing robotic platform, with their respective onboard sensors, in the DCSL.
■590 ▼aSchool code: 0078.
■650 4▼aAeronautics
■650 4▼aKinematics
■650 4▼aCoordinate transformations
■650 4▼aOptimization techniques
■650 4▼aPlanning
■650 4▼aGraph representations
■650 4▼aRobots
■650 4▼aAttitudes
■650 4▼aVisualization
■650 4▼aRobotics
■650 4▼aAerospace engineering
■650 4▼aIndustrial engineering
■690 ▼a0771
■690 ▼a0538
■690 ▼a0546
■71020▼aGeorgia Institute of Technology.
■7730 ▼tDissertations Abstracts International▼g87-05B.
■790 ▼a0078
■791 ▼aPh.D.
■792 ▼a2023
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17365963▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.


