About OpenKnowledge@NAU | For NAU Authors

The design and testing of a powered exoskeleton to reduce the metabolic cost of walking in individuals with cerebral palsy

Bair, Michael O. (2018) The design and testing of a powered exoskeleton to reduce the metabolic cost of walking in individuals with cerebral palsy. Masters thesis, Northern Arizona University.

[img] Text
Bair_M_2018_Design_and_testing_of_a_powered_exoskeleton.pdf - Published Version

Download (2MB)


Cerebral palsy (CP) is the most common form of motor impairment in children. Although CP is a non-progressive neurodevelopmental disorder, its secondary effects can lead to a decline in mobility over time. Eventually, individuals with CP may lose all ambulatory function. Interventions for CP, such as surgery and physical therapy, seek to delay the decline and improve mobility. Research into alternative interventions attempt to find improved outcomes over the traditional treatments. Robot-assistive devices are one such intervention that makes use of powered actuation, sensing, and control to help individuals with CP to retain and improve mobility. The purpose of this research was to develop and test a robot-assistive device that reduces the metabolic cost of walking in children and young adults with CP. Our device consisted of an actuator-and-control module worn on the back, and two custom ankle-foot orthotics (AFOs). The control module contained brushless DC motors, electronics to control actuation, and a power supply. Plantarflexion, or “push-off”, assistive torque was transferred from the motors to the orthotics using Bowden cables. 32-Bit ARM microcontrollers operated the system with inputs from force sensitive resistors (FSR) placed under the ball of each foot, and torque sensors aligned with each ankle joint. Proportional gain control was used to control motor output torque. We implemented a two-state Finite State Machine (FSM) to control the timing of assistance. We completed a pilot study with three participants with CP. We used indirect calorimetry to estimate metabolic cost. The study participants walked under three conditions: baseline, zero-torque, and powered assist. During the baseline condition, participants walked using their own shoes/AFOs and not our device. During the zero-torque condition, participants wore the device and the motors actively maintained a zero-torque reading at the ankle. During the assisted condition, participants wore the device while the motors gave plantarflexion assistance during late stance. Two participants had a decrease in metabolic cost between baseline and assisted conditions, an average of 16.5%. All participants showed an average reduction in the metabolic cost of walking of 40% between the zero-torque condition and the assisted condition, on average. Our youngest and lightest participant (5 years old, 16 kg) did not show a net decrease in metabolic cost between baseline and assisted walking, a result likely due to our device being a large percentage of their total mass, 12.5% as opposed to 7.5% and 4.4%. We conclude that our device shows a strong potential for clinically-relevant applications. Further studies may show that robotic-assistance can improve mobility and quality of life in individuals with CP.

Item Type: Thesis (Masters)
Publisher’s Statement: © Copyright is held by the author. Digital access to this material is made possible by the Cline Library, Northern Arizona University. Further transmission, reproduction or presentation of protected items is prohibited except with permission of the author.
Keywords: Ankle; Cerebral palsy; Exoskeleton; Physical therapy; Walking; Wearable robotics; Biomechanics; Biological sciences; Applied sciences; Health and environmental sciences
Subjects: T Technology > TJ Mechanical engineering and machinery
NAU Depositing Author Academic Status: Student
Department/Unit: Graduate College > Theses and Dissertations
College of Engineering, Forestry, and Natural Science > Mechanical Engineering
Date Deposited: 31 Oct 2018 00:44
URI: http://openknowledge.nau.edu/id/eprint/5403

Actions (login required)

IR Staff Record View IR Staff Record View


Downloads per month over past year