서브메뉴
검색
The Development and Evaluation of Biomimetic Control for Robotic Lower-Limb Prostheses: Advancing Toward Real-World Clinical Use
The Development and Evaluation of Biomimetic Control for Robotic Lower-Limb Prostheses: Advancing Toward Real-World Clinical Use
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202105217
- ISBN
- 9798291565827
- DDC
- 610
- 서명/저자
- The Development and Evaluation of Biomimetic Control for Robotic Lower-Limb Prostheses: Advancing Toward Real-World Clinical Use
- 발행사항
- [Sl] : University of Michigan, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 208 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-03, Section: B.
- 주기사항
- Advisor: Gregg, Robert D.;Rouse, Elliott J.
- 학위논문주기
- Thesis (Ph.D.)--University of Michigan, 2025.
- 초록/해제
- 요약Robotic prostheses with motorized joints have the potential to improve the mobility and quality of life for millions of people worldwide who live with lower-limb amputations by more fully replacing the functions of the amputated limb. However, the translation of these potentially transformative devices from research contexts into mainstream clinical practice has been limited, despite their many promised benefits. Commercialized robotic knee prostheses have seen limited clinical adoption, and no combined robotic knee-ankle prosthesis has been brought to market. This limited translation into clinical practice is, in large part, due to the absence of clinically viable control methods to effectively manage robotic prosthesis behavior across the highly variable conditions encountered in real-world ambulation.To address this need, Part I of this dissertation proposes a novel data-driven control method for robotic prostheses that can automatically adapt to variable walking speeds and ground inclines. By creating biomimetic (i.e. human-like) joint behaviors across different terrains without requiring manual tuning, our approach provides users with meaningful assistance while maintaining clinical viability. Furthermore, we propose new methods to accommodate non-negligible drive-train dynamics, which are common in commercialized robotic prosthesis designs. Our impedance compensator, which uses a simple physics-based model of the prothesis's drivetrain dynamics to improve joint impedance control accuracy, expands the applicability of our data-driven walking controller and enables us to deploy it on the second-generation Open-Source Leg and the commercially available Ossur Power Knee™.Next, in Part II, we evaluate the potential clinical benefits of robotic prostheses utilizing the data-driven control strategies developed in Part I. Through experiments with both prototype and commercialized prosthetic hardware, we demonstrate that robotic knee and knee-ankle prostheses can provide meaningful clinical benefits over passive microprocessor knee prostheses. We further show that, in many cases, our data-driven controller enhances these beneficial effects compared to traditional control methods. Together, these initial findings lay the foundation for future larger-scale clinical trials investigating the long-term health implications of robotic prosthesis use under various control policies, potentially motivating their wider inclusion in insurance policy coverage.Finally, in Part III, we investigate the effects of varying a prothesis's mechanical joint impedance (i.e. the relationship between small joint position perturbations and the corresponding torque responses). We first show that in a robotic prosthesis, the joint impedance can be selected independently from the desired nominal kinematic and kinetic behavior, making it a free design choice. In walking experiments that include momentary treadmill belt speed perturbations, we then report how different choices of joint impedance, including biomimetic, constant, and random trajectories, affect a user's nominal walking behavior as well as their ability to recover from perturbations. Our results provide initial guidance on which joint impedance characteristics should be used in a robotic prosthesis and motivate further research on the relationship between joint impedance and users' walking stability.As a whole, these contributions help to advance the field's understanding of lower-limb robotic prosthesis control in ways that are applicable to real-world clinical use. By demonstrating user-centric clinical benefits with a biomimetic control approach that adapts to variable terrain, we hope to help accelerate robotic knee-ankle prostheses out of the research labs and into mainstream clinical practice, ultimately helping to improve users' quality of life.
- 일반주제명
- Biomedical engineering
- 일반주제명
- Robotics
- 일반주제명
- Mechanical engineering
- 키워드
- Prosthetic legs
- 키워드
- Optimal control
- 기타저자
- University of Michigan Robotics
- 기본자료저록
- Dissertations Abstracts International. 87-03B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017359804
■00520260202105217
■006m o d
■007cr#unu||||||||
■020 ▼a9798291565827
■035 ▼a(MiAaPQ)AAI32271764
■035 ▼a(MiAaPQ)umichrackham006498
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a610
■1001 ▼aBest, Thomas Kevin.
■24510▼aThe Development and Evaluation of Biomimetic Control for Robotic Lower-Limb Prostheses: Advancing Toward Real-World Clinical Use
■260 ▼a[Sl]▼bUniversity of Michigan▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a208 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-03, Section: B.
■500 ▼aAdvisor: Gregg, Robert D.;Rouse, Elliott J.
■5021 ▼aThesis (Ph.D.)--University of Michigan, 2025.
■520 ▼aRobotic prostheses with motorized joints have the potential to improve the mobility and quality of life for millions of people worldwide who live with lower-limb amputations by more fully replacing the functions of the amputated limb. However, the translation of these potentially transformative devices from research contexts into mainstream clinical practice has been limited, despite their many promised benefits. Commercialized robotic knee prostheses have seen limited clinical adoption, and no combined robotic knee-ankle prosthesis has been brought to market. This limited translation into clinical practice is, in large part, due to the absence of clinically viable control methods to effectively manage robotic prosthesis behavior across the highly variable conditions encountered in real-world ambulation.To address this need, Part I of this dissertation proposes a novel data-driven control method for robotic prostheses that can automatically adapt to variable walking speeds and ground inclines. By creating biomimetic (i.e. human-like) joint behaviors across different terrains without requiring manual tuning, our approach provides users with meaningful assistance while maintaining clinical viability. Furthermore, we propose new methods to accommodate non-negligible drive-train dynamics, which are common in commercialized robotic prosthesis designs. Our impedance compensator, which uses a simple physics-based model of the prothesis's drivetrain dynamics to improve joint impedance control accuracy, expands the applicability of our data-driven walking controller and enables us to deploy it on the second-generation Open-Source Leg and the commercially available Ossur Power Knee™.Next, in Part II, we evaluate the potential clinical benefits of robotic prostheses utilizing the data-driven control strategies developed in Part I. Through experiments with both prototype and commercialized prosthetic hardware, we demonstrate that robotic knee and knee-ankle prostheses can provide meaningful clinical benefits over passive microprocessor knee prostheses. We further show that, in many cases, our data-driven controller enhances these beneficial effects compared to traditional control methods. Together, these initial findings lay the foundation for future larger-scale clinical trials investigating the long-term health implications of robotic prosthesis use under various control policies, potentially motivating their wider inclusion in insurance policy coverage.Finally, in Part III, we investigate the effects of varying a prothesis's mechanical joint impedance (i.e. the relationship between small joint position perturbations and the corresponding torque responses). We first show that in a robotic prosthesis, the joint impedance can be selected independently from the desired nominal kinematic and kinetic behavior, making it a free design choice. In walking experiments that include momentary treadmill belt speed perturbations, we then report how different choices of joint impedance, including biomimetic, constant, and random trajectories, affect a user's nominal walking behavior as well as their ability to recover from perturbations. Our results provide initial guidance on which joint impedance characteristics should be used in a robotic prosthesis and motivate further research on the relationship between joint impedance and users' walking stability.As a whole, these contributions help to advance the field's understanding of lower-limb robotic prosthesis control in ways that are applicable to real-world clinical use. By demonstrating user-centric clinical benefits with a biomimetic control approach that adapts to variable terrain, we hope to help accelerate robotic knee-ankle prostheses out of the research labs and into mainstream clinical practice, ultimately helping to improve users' quality of life.
■590 ▼aSchool code: 0127.
■650 4▼aBiomedical engineering
■650 4▼aRobotics
■650 4▼aMechanical engineering
■653 ▼aProsthetic legs
■653 ▼aTransfemoral amputees
■653 ▼aOptimal control
■653 ▼aMechanical impedance
■653 ▼aTransfemoral prostheses
■690 ▼a0771
■690 ▼a0541
■690 ▼a0800
■690 ▼a0548
■71020▼aUniversity of Michigan▼bRobotics.
■7730 ▼tDissertations Abstracts International▼g87-03B.
■790 ▼a0127
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17359804▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.


