Interaction models between humans and lower-limbs exoskeletons applied to robotic neurorehabilitation

• Ano de defesa: 2020
• Autor(a) principal: Pereira, Denis César Mosconi
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Biblioteca Digitais de Teses e Dissertações da USP
• Palavras-chave em Português: acidente vascular cerebral; human-robot interaction; interação humano-robô; OpenSim; robot-assisted therapy; stroke; terapia assistida por robô
• Link de acesso: https://www.teses.usp.br/teses/disponiveis/18/18162/tde-12042023-111154/

The number of strokes has grown steadily, causing thousands of victims around the world. Approximately 90% of stroke survivors remain with some disability, requiring physical therapy of rehabilitation in order to recover the skills of carrying out the activities of daily living. The use of robots in post-stroke rehabilitation therapies has been shown to be a promising alternative to increase the efficacy of the treatment. The assurance of a human-robot interaction safe for the patient and useful for treatment has been widely studied in the field of rehabilitation engineering, focusing on the development of human-robot interaction controls and biomimetic robots, such as exoskeletons. But the validation and test of this resources is still a challenge: How to do this with reduced cost, low time consumption and without putting patients at risk? The objective of this work was to develop a computational human-exoskeleton interaction model and a forward dynamics-based simulation environment, capable of being applied in the development of interaction controls used in the rehabilitation of lower limbs. The interaction model was developed using computational neuromusculoskeletal systems from OpenSim, with virtual models of the robot actuators coupled to it. The simulation environment was developed in MATLAB, using the OpenSim API. Both the interaction model and the simulation environment were validated using data from physical experiments. Four fully computational simulations were performed with different interaction and movement controls. The results obtained proved that the model and the simulation environment are feasible and useful for the development and simulation of interaction controls between humans and lower limb exoskeletons, saving money, time and ensuring the safety of the user and the equipment used.