Design, control and user evaluations of a self-aligning knee exoskeleton

Warning The system is temporarily closed to updates for reporting purpose.

Irmak, Çağatay (2022) Design, control and user evaluations of a self-aligning knee exoskeleton. [Thesis]

Full text not available from this repository. (Request a copy)

Abstract

The demand for the treatment and rehabilitation of lower and upper extremities is expected to increase due to various reasons, such as the growing world population and increasing neurological disorders. Robot-assisted rehabilitation is preferred for the administration of physical therapies, as repetitive and physically demanding rehabilitation treatments can be delivered with high intensity and accuracy, while simultaneously ensuring the safety of patients. Furthermore, these devices can provide quantitative measurements of patient progress thanks to their integrated sensors and administer customized assistance to satisfy the individual needs of each patient. In this thesis, we present a self-aligning robot-assisted knee exoskeleton to improve the mobility of patients. The knee exoskeleton features self-alignment based on a compliant mechanism to ensure the ergonomic movements of the knee. Thanks to this self-aligning property, the knee exoskeleton enables the translational movements of the knee joint in the sagittal plane, in addition to the main flexion/extension movements. Furthermore, a Bowden cable-based series elastic actuator is utilized for the remote placement of the actuators while enabling high-fidelity interaction control. As a novel feature, a large stroke XY-stage compliant mechanism is proposed to address the misalignment problem between the human knee joint and the exoskeleton. A compliant mechanism with high rotational and low translational stiffness levels is designed such that the self-alignment mechanism can be implemented in a compact and light-weight form, while featuring no backlash and friction. Through the compliant alignment mechanism, a close match between the human knee joint and the exoskeleton joint axes is achieved by automatically aligning the joint axis of the exoskeleton, such that parasitic interaction forces induced on the user are minimized. iii A Bowden cable-driven series elastic actuator is utilized for the knee exoskeleton to actively control the interaction torques applied to the knee joint, such that assistance torques can be provided to the user. Bowden cable transmission enables the actuator and harmonic reduction units to be placed at a remote location, enabling the exoskeleton to feature low mass and inertia and relieving the user from the burden of carrying this weight. Thanks to its series elastic actuation, the knee exoskeleton displays high-fidelity torque tracking performance, while also featuring active back-driveability within the control bandwidth and passive compliance for excitations above this bandwidth. We also present a comprehensive set of human subject experiments designed to evaluate the ergonomic improvements provided by the self-alignment feature of the knee exoskeleton. The results of these human subject experiments provide evidence that the self-alignment feature of the knee exoskeleton decreases the parasitic interaction forces due to misalignment and ensures that the exoskeleton does not significantly interfere with the knee kinematics.
Item Type: Thesis
Uncontrolled Keywords: Physical human-robot interaction. -- rehabilitation robotics. -- interaction control. -- series elastic actuation. -- force-feedback exoskeletons. -- self-alignment. -- compliant mechanisms. -- human subject experiments. -- Fiziksel insan-robot etkileşimi. -- rehabilitasyon robotları. -- etkileşim kontrolü. -- seri elastik eyleme. -- kuvvet geri-beslemeli dış iskelet. -- kendinden hizalama. -- Bowden kablolu güç iletimi. -- esnek mekanizmalar. -- insanlı deneyler.
Subjects: T Technology > TJ Mechanical engineering and machinery > TJ163.12 Mechatronics
Divisions: Faculty of Engineering and Natural Sciences > Academic programs > Mechatronics
Faculty of Engineering and Natural Sciences
Depositing User: Dila Günay
Date Deposited: 24 Apr 2023 11:30
Last Modified: 10 Jul 2023 10:16
URI: https://research.sabanciuniv.edu/id/eprint/47044

Actions (login required)

View Item
View Item