Resumen
The present study describes a system for recognition of a group of the most functionally useful hand movements and gestures that are commonly implemented on bionic prosthetic hands in order recover some of the physical abilities that a person has lost as consequence of an amputation. This is done using a proposed hardware trough a working prototype of a small, wireless, low-cost sensor that places on the skin surface of the subject forearm to detect patterns on muscular vibrations that occur during contraction, a technique called mechanomyography (MMG). As transducers, a piezoelectric contact sensor and a triaxial accelerometer is used, the first one can detect only the muscle low-frequency oscillations, the second one, apart from adding information to the MMG record also can provide inertial data like macro movements in order to introduce context into the detected pattern. The main advantages of this design compared to reviewed alternatives are: first, it is a non-invasive device that occupies less skin area, second, minimizes the number of input channels by introducing more flexibility to distribute sensors through the remaining portion of the limb or even the rest of the body, and third, its reduced cost makes it an attractive alternative when thinking about mass adoption of human machine interfaces for prosthetic devices control. The proposed sensor has demonstrated great potential to be used as an input device for monitoring muscular activity and body kinematics during initial experimentation where data recorded during activities has been introduced to a convolutional neural network.
Idioma original | Inglés |
---|---|
Título de la publicación alojada | Proceedings of the Future Technologies Conference (FTC) 2023, Volume 2 |
Editores | Kohei Arai |
Editorial | Springer Science and Business Media Deutschland GmbH |
Páginas | 86-97 |
Número de páginas | 12 |
ISBN (versión impresa) | 9783031474507 |
DOI | |
Estado | Publicada - 2023 |
Evento | 8th Future Technologies Conference, FTC 2023 - San Francisco, Estados Unidos Duración: 2 nov. 2023 → 3 nov. 2023 |
Serie de la publicación
Nombre | Lecture Notes in Networks and Systems |
---|---|
Volumen | 814 LNNS |
ISSN (versión impresa) | 2367-3370 |
ISSN (versión digital) | 2367-3389 |
Conferencia
Conferencia | 8th Future Technologies Conference, FTC 2023 |
---|---|
País/Territorio | Estados Unidos |
Ciudad | San Francisco |
Período | 2/11/23 → 3/11/23 |
Nota bibliográfica
Publisher Copyright:© 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.