Evaluación de desempeño de algoritmos de inteligencia artificial embebida en microcontroladores
Cargando...
Fecha
2024
Autores
Correa Polo, Alejandro
De La Hoz Salas, Xavier
Gil Guzmán, Jaider José
Romero Domínguez, Vanessa Paola
Tovio Caballero, Justin
Título de la revista
ISSN de la revista
Título del volumen
Editor
Ediciones Universidad Simón Bolívar
Facultad de Ingenierías
Facultad de Ingenierías
Resumen
La Integración de la Inteligencia Artificial con el Internet de las Cosas (AIoT) implica combinar la capacidad de procesamiento inteligente de la IA con la conectividad y los datos generados por los dispositivos IoT, lo que potencia la automatización, el análisis de datos en tiempo real y la toma de decisiones autónoma en diversas áreas como el hogar inteligente, la salud, la agricultura y la manufactura. En el ámbito de la salud, por ejemplo, la AIoT permite monitorear la salud de los pacientes y prever problemas médicos. En la agricultura, optimiza el riego y la fertilización de los cultivos para aumentar los rendimientos. La fabricación se beneficia con el mantenimiento predictivo de equipos y la optimización de la cadena de suministro.
Este artículo aborda los desafíos y propone enfoques para la integración efectiva de redes neuronales (RNA) en microcontroladores, como el ESP32. A pesar de los desafíos por las limitaciones del ESP32, se superaron mediante la utilización de materiales y herramientas apropiadas, implementando algoritmos de RNA y estableciendo una metodología para medir los tiempos de ejecución en el dispositivo
The Integration of Artificial Intelligence with the Internet of Things (AIoT) involves combining the intelligent processing capacity of AI with the connectivity and data generated by IoT devices, which enhances automation, real-time data analysis and autonomous decision making in various areas such as smart home, healthcare, agriculture and manufacturing. In the field of health, for example, AIoT makes it possible to monitor the health of patients and predict medical problems. In agriculture, it optimizes irrigation and fertilization of crops to increase yields. Manufacturing benefits from predictive equipment maintenance and supply chain optimization. This article addresses the challenges and proposes approaches for effective integration of neural networks (ANNs) into microcontrollers, such as the ESP32. Despite the challenges due to the limitations of the ESP32, they were overcome by using appropriate materials and tools, implementing ANN algorithms, and establishing a methodology to measure execution times on the device.
The Integration of Artificial Intelligence with the Internet of Things (AIoT) involves combining the intelligent processing capacity of AI with the connectivity and data generated by IoT devices, which enhances automation, real-time data analysis and autonomous decision making in various areas such as smart home, healthcare, agriculture and manufacturing. In the field of health, for example, AIoT makes it possible to monitor the health of patients and predict medical problems. In agriculture, it optimizes irrigation and fertilization of crops to increase yields. Manufacturing benefits from predictive equipment maintenance and supply chain optimization. This article addresses the challenges and proposes approaches for effective integration of neural networks (ANNs) into microcontrollers, such as the ESP32. Despite the challenges due to the limitations of the ESP32, they were overcome by using appropriate materials and tools, implementing ANN algorithms, and establishing a methodology to measure execution times on the device.
Descripción
Palabras clave
Inteligencia Artificial, Dispositivos IoT, Redes Neuronales Artificiales, Perceptrones multicapa, ESP32
