Exoesqueleto cooperativo mecatrónico para tratamiento muscular de zonas inferiores basado en la implementación de técnicas de edge computing
This degree project focuses on the use of exoskeletons as devices for interactive therapies, implementing deep learning technologies embedded in ESP32 (TinyML). The research covers the capture and processing of physical variables through sensors, improving accuracy by means of a Kalman filter, with...
- Autores:
-
Alvarez Vanegas, Daniel Alejandro
- Tipo de recurso:
- Trabajo de grado de pregrado
- Fecha de publicación:
- 2024
- Institución:
- Universidad de los Andes
- Repositorio:
- Séneca: repositorio Uniandes
- Idioma:
- spa
- OAI Identifier:
- oai:repositorio.uniandes.edu.co:1992/74698
- Acceso en línea:
- https://hdl.handle.net/1992/74698
- Palabra clave:
- Exoesqueleto
Cinemática
Atrofía muscular
Filtro de Kalman
TinyML
IMU
Stepper
OpenPose
Edge computing
ROS2
Ingeniería
- Rights
- openAccess
- License
- Attribution 4.0 International
| Summary: | This degree project focuses on the use of exoskeletons as devices for interactive therapies, implementing deep learning technologies embedded in ESP32 (TinyML). The research covers the capture and processing of physical variables through sensors, improving accuracy by means of a Kalman filter, with error corrections ranging from 20 % to 54 %. These data, after being processed in real time, are transmitted to a Raspberry Pi type computer, which monitors and guides the actions of a specifically designed and manufactured exoskeleton. Communication between the components is done through UDP and serial protocols, thus contributing to the improvement of walking in individuals with muscular atrophy. The entire system is integrated into the ROS2 middleware for efficient and modular infrastructure management. |
|---|