Development of a tendon-driven, voice-controlled soft robotic hand exoskeleton
Author(s)
Tran, Phillip
Advisor(s)
Editor(s)
Collections
Supplementary to:
Permanent Link
Abstract
Functional hand movement is an important component of many activities of daily living, such as using a phone or eating. Cervical spinal cord injury (SCI) can severely impact hand motor and sensory function, and accordingly, patients with SCI are often unable to complete basic everyday tasks without assistance. This thesis concerns the design and development of a tendon-driven, voice-controlled soft robotic assistive hand exoskeleton (FLEXotendon Glove) with the purpose of providing active assistance to users during grasping and pinching motions, which is intended to improve hand and finger functionality during the performance of everyday tasks in individuals with hand dysfunction. The FLEXotendon Glove underwent several iterations to reach the final design, allowing for successive refinement of materials, fabrication processes, and control methods. The final FLEXotendon Glove system comprises four separate subsystems: a soft exoskeleton glove, tendon tension sensors, actuation units, and a smartphone. The soft exoskeleton glove is composed of 3D-printed plastic tendon routing units and force-sensitive resistors incorporated into a high consistency rubber silicone glove. The fabrication process for the construction of the soft exoskeleton glove is presented and describes a novel method for the creation of a durable, flexible, lightweight, and patient-specific soft exoskeleton glove. Characterization of the force-sensitive resistors and the developed circuit are presented. The tendon tension sensors were refined over two versions from photoresistive sensors to piezoresistive sensors, and are used to provide force feedback for the developed admittance controller. The design and characterization of the tendon tension sensors, derivation and modeling of the admittance controller, and validation of the admittance controller are presented. The actuation units of the FLEXotendon Glove system utilize ultra-high molecular weight polyethylene threads and incorporate bio-inspired tendon routing for active finger flexion and extension. The design, development, and evaluation of the different iterations of the actuation units in the FLEXotendon Glove system are presented. A smartphone app was developed as the front-end user interface for the exoskeleton system to provide an equitable platform for the FLEXotendon Glove owing to the ubiquitous use of smartphones. The development and operational scheme of a voice control system for exoskeleton operation coupled with a smartphone-based button interface for exoskeleton calibration is presented. To explore alternate strategies for the manipulation of objects while using the FLEXotendon Glove system, a 3D-printed self-sealing suction cup and self-charging pneumatic circuit was developed and integrated into the exoskeleton system. The design of the suction cup was optimized to be 3D-printed in a single print and incorporates a gimbal mechanism to accommodate a wide range of object contact angles. The design, fabrication process, material selection process, modeling of the pneumatic circuit, and characterization and evaluation of the suction cup are presented. The FLEXotendon Glove system as well as the self-sealing suction cups were assessed with impaired and healthy human participants to evaluate the performance of the systems in a clinically relevant setting and from a general efficacy viewpoint, respectively. The FLEXotendon Glove system was evaluated with both adult and pediatric subjects with hand impairments resulting from SCI or traumatic brain injury. Standardized hand function tests were used as primary outcome measures and standard clinical questionnaires were used as secondary outcome measures for the evaluation of the exoskeleton system, and the results from these assessments are presented. Electromyographical (EMG) signals from the finger flexor and extensor muscles were used to compare the effort required for manipulating objects with and without the exoskeleton and/or the suction cups. The EMG signals from healthy participants were collected to analyze the general performance of the exoskeleton and the results are presented.
Sponsor
Date
2022-12-21
Extent
Resource Type
Text
Resource Subtype
Dissertation