Interfaces e estratégias de controle baseadas em machine learning aplicadas a um exoesqueleto de braço para reabilitação motora

• Ano de defesa: 2024
• Autor(a) principal: Dias, Eduardo Antonio Fragoso
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Espírito Santo; BR; Mestrado em Engenharia Mecânica; Centro Tecnológico; UFES; Programa de Pós-Graduação em Engenharia Mecânica
• Palavras-chave em Português: Exoesqueleto; Eletromiografia; Interfaces de controle; Engenharia Mecânica
• Link de acesso: http://repositorio.ufes.br/handle/10/18155

Stroke is one of the leading causes of acquired disability worldwide, with approximately 80% of survivors living with permanent disabilities. Recovery of upper limb functionality is particularly challenging, with only about 50% of patients regaining some functional use after the event. In this context, physiotherapy and occupational therapy are essential, but robotic rehabilitation emerges as a promising alternative to enhance therapeutic outcomes. However, the effectiveness of these new therapies is directly linked to the control interfaces and strategies applied in the patient’s interaction with the robotic device. Thus, this work investigates the development and application of control strategies and interfaces in an arm exoskeleton designed for neuromotor rehabilitation of post-stroke patients. We explore different strategies using surface electromyography (sEMG) to identify the patient’s movement intention and strategies operated by the physiotherapist, applied to an upper limb exoskeleton. Additionally, a rehabilitation protocol was implemented using an interface based on assistive trajectory control. The results indicate significant improvements in rehabilitation, highlighting the efficacy of incorporating advanced robotic technologies in the neuromotor recovery process, providing a more effective and individually tailored approach to patient needs.