Achieving Near-Natural Locomotion in Transfemoral Amputees - A Control Theoretic Approach
Authors
Kibria, Zunaed
Issue Date
2024
Type
Dissertation
Language
Keywords
Control System , Gait Asymmetry , Neural Network Learning , Neuro Dynamic Programming , Prosthetic Leg , Transfemoral Amputation
Alternative Title
Abstract
Amputation of the lower limb is prescribed to address conditions such as trauma, vascular issues, tumors, neuropathy, frostbite, and complications from diabetes. Post-surgery, the individual has to be fitted with a prosthetic limb to regain mobility. While a good-fitting, well-designed prosthetic device can help support body weight and locomotion, the loss of lower limb muscles profoundly impacts body support, and stability and often results in asymmetrical gait patterns. Traditionally, prosthetic limbs were designed primarily to support weight and mimic the appearance of natural limbs. However, modern prosthetic devices aim for active functionality, striving to replicate near-natural human locomotion. However, achieving such functionality is not an easy task as the control of the prosthetic limb has to determine the user's intent while adapting to unknown nonlinear dynamics, and varying terrain and environmental conditions. Among lower limb amputations, transtibial amputations are the most common. The loss of the ankle joint for transtibial amputees causes several constraints during gait, such as lower walking speed, increased metabolic energy expenditure, and reduced power in the amputated limb during the push-off phase. For above-knee amputations, where the individual loses both the knee and ankle joints, the effects are more severe. Losing multiple joints leads to significant asymmetry in gait, which can affect musculoskeletal health in the long run.