Deep learning based motion recognition and adaptive sliding mode controlled prosthetic foot design with IMU data

This research sets out with the aim of developing a user-controlled prosthetic foot that will not make the absence of the missing limb felt for individuals who have undergone below-knee amputation due to stroke, neurological diseases, orthopedic issues, or trauma. Designed to enable users to comfortably carry out their daily life activities, this prosthetic foot with two degrees of freedom mimics a natural gait, providing users with freedom of movement and natural balance. The developed prosthetic foot possesses critical features such as preserving symmetry, mimicking natural movements, and reflecting the natural dynamics of walking. Comprised of various components including actuators, controllers, sensors, connection elements, and power units, this prosthetic foot is designed in accordance with human anatomy, offering users a wide range of motion. The data-driven approach underlying the research enables the creation of customized prosthetic foot designs tailored to the needs of each individual. This aims to enhance the quality of life for amputees, thereby promoting their more active and independent participation in society. Additionally, it aims to contribute to the advancement of scientific research in this field, thereby supporting the development of prosthetic technology.