Research within the field of human motion assistive device development, with the purpose of reducing the metabolic cost of daily activities, is seeing the benefits of the exclusive use of passive actuators to store and release energy during the gait cycle. Designs of novel exoskeletons at the University of Ottawa implement the Pneumatic Artificial Muscle (PAM) as the primary method of nonlinear, passive actuation. The PAM is proven as a superior actuator for these devices when compared to the linear mechanical springs used by other researchers. There are, however, challenges regarding PAM pressure loss and the limitation of PAM elongation that have been identified. This thesis aims to develop a hyperelastic tubular soft composite that replicates the distinctive mechanical behaviour of the PAM without the need for internal pressurization. The final soft composite solution was achieved by impregnating a prefabricated polyethylene terephthalate braided sleeve, held at a high initial fibre angle, with a silicone prepolymer. A comprehensive experimental evaluation was performed on numerous prototypes for a variety of customizable design parameters including: initial fibre angle, silicone stiffness, and braided sleeve style. Moreover, two separate analytical models were formulated based on incompressible finite elasticity theory using either a structural model of Holzapfel’s type, or a phenomenological model of Fung’s type. Both models were in good agreement with the experimental data that were collected through a modified extension-inflation test. This research has successfully developed, tested, and validated an innovative soft composite that can achieve specific mechanical properties, such as contraction distance and nonlinear stiffness, for optimal use in human motion assistive devices.

Mobility is essential to healthy aging and is closely related to health and overall quality of life. Mobility impairment is an early predictor of disability, and is intimately linked to falling injuries, loss of independence, hospitalization, and mortality. Mobility Assistive Devices (MADs), such as canes, walkers, and wheelchairs, provide support to older adults to improve their balance, coordination, and strength. Despite the acknowledged benefits of MADs for the older adult population research shows that as many as 75% of older adults are non-adherent to prescribed MADs. As non-adherence is a contributing factor to declines in mobility and overall quality of life, it is important to consider the reasoning behind it. Therefore, this doctoral project took a psychosocial perspective and investigated the psychosocial factors that predict non-adherence to MADs among older adults. This study considered a sample of older adult MADs users from long-term care units at the Perley Health Centre in Ottawa with different types of mobility limitations. The study examined the role of psychosocial variables using the Psychosocial Impact of Assistive Devices Scale (PIADS), social support using the Medical Outcomes Study Social Support Survey (MOS-SS), and the demographic variable of sex in the prediction of MADs non-adherence. Predictor variables that were associated with non-adherence in a univariate regression analysis were subsequently entered into a multiple regression analysis. Of the 96 residents invited to participate in the study, 49 gave their consent to participate, and out of this number, 48 completed the study, for a response rate of 51%. The data from the 48 residents (26 females and 22 males), with a mean age of 86.8 (Standard Deviation (SD) = 10.2, age range= 66 - 101), were therefore available for analysis. The most common reported mobility limitations were due to balance problems and leg weakness (29.17% for each). The most common device used was a walker (47.92%), followed by a manual wheelchair (33.33%). No statistical difference was detected between the groups of sexes regarding any of the study variables (P≥0.05). In the univariate regression analysis, the three PIADS subscales, namely, Competence, Adaptability, and Self-esteem, were significantly correlated with non-adherence (p < 0.001). Sex was an insignificant variable, while social support was significantly correlated with Competence, Adaptability, Self-esteem, and non-adherence. In the multiple regression analyses, only Self-esteem showed significant associations (p < 0.05), and the Self-esteem multivariate model explained 43.5 - 54.3% of the variance in non-adherence. This study revealed that the Self-esteem construct, which includes several concepts related to psychological wellbeing, was the only significant predictor of non-adherence among the studied sample of older adults. The theoretical and clinical implications of the findings are subsequently discussed.