AI_Card on the edge: for teaching and learning neural networks and AI
DOI:
https://doi.org/10.26507/rei.v21n42.1444Keywords:
neural network, Generative artificial intelligence, educacion, LLM, AI on the EngeAbstract
This article reports a proof of concept carried out with two purposive samples —six eleventh-grade students with no prior programming background and eighteen professionals from various fields— and states three contributions: (1) the design and construction of low-cost educational hardware integrating an ESP32-S3 microcontroller, DVP camera, Wi-Fi access point, microSD and thirteen GPIO outputs in a single module; (2) D_T_I_AI.py, unified software that runs the complete Data–Training–Inference pipeline locally and without an Internet connection, automatically trains and compares three architectures (MLP, CNN and CNN with advanced regularisation), and embeds a multimodal analysis module using multimodal language models (LMM) served by Ollama; and (3) preliminary findings suggesting that participants without prior programming background can complete the full cycle of dataset construction, training, inference-based contrast, and physical actuator control within bounded working sessions.
Author Biographies
Ferney Cortés Garnica, Secretaría de educación de Bogotá
Licenciado en Electrónica de la Universidad Pedagógica Nacional de Colombia (2006), Especialista en Matemática Aplicada de la Universidad Sergio Arboleda (2009), Magíster en Didáctica de las Ciencias de la Universidad Autónoma de Colombia (2016) y Doctor en Educación de la Universidad de Baja California, México (2022). Se desempeña como docente investigador en la Secretaría de Educación de Bogotá y en PIXIE.
Kevin Daniel Martínez Ospino, Colegio Manuel Cepeda Vargas IED
Técnico en sistemas informáticos del SENA (2024) y Tecnólogo en sistemas informáticos del Politécnico Santafé de Bogotá (2026) actualmente se desempeña como asistente de investigación en la Institución Educativa Manuel Cepeda Vargas IED.
References
UNESCO, “Consenso de Beijing sobre la inteligencia artificial y la educación”, 2019, pp. 27-38 https://unesdoc.unesco.org/ark:/48223/pf0000368303
UNESCO, “Inteligencia artificial y educación. Guía para las personas a cargo de formular políticas” 2021. [En línea]. Disponible en: https://unesdoc.unesco.org/ark:/48223/pf0000379376
UNESCO, “Marco de competencias para docentes en materia de IA”, 2025 https://unesdoc.unesco.org/ark:/48223/pf0000393813?locale=es
UNESCO, “Marco de competencias para estudiantes en materia de IA”, 2025 https://unesdoc.unesco.org/ark:/48223/pf0000391105
Departamento Nacional de Planeación, “Documento CONPES 4144”, 2025, https://colaboracion.dnp.gov.co/CDT/Conpes/Econ%C3%B3micos/4144.pdf
D. Situnayake, & J. Plunkett, “AI at the edge: Solving real-world problems with embedded machine learning”. O’Reilly Media, 2023.
P. Warden y D. Situnayake, “TinyML: Machine learning with TensorFlow Lite on Arduino and ultra-low-power microcontrollers”. O’Reilly Media, 2020.
K. Sharma et al., “Revolutionizing education with Edge AI”, En “Edge Artificial Intelligence,” Cap. 26, 2025 https://doi.org/10.1002/9781394355037.ch26
V. Janapa Reddi et al., "Widening access to applied machine learning with TinyML", arXiv:2106.04008, 2021. https://arxiv.org/pdf/2106.04008
B. Plancher y V. Janapa Reddi, "TinyMLedu: The tiny machine learning open education initiative", en Proc. 53rd ACM Technical Symposium on Computer Science Education (SIGCSE 2022), vol. 2, 2022, p. 1159. https://doi.org/10.1145/3478432.3499093
B. Plancher et al., "TinyML4D: Scaling embedded machine learning education in the developing world", Proceedings of the AAAI Symposium Series, vol. 3, n.º 1, 2024. https://ojs.aaai.org/index.php/AAAI-SS/article/view/31265/33425
S. Hymel et al., "Edge Impulse: An MLOps platform for tiny machine learning", en Proceedings of Machine Learning and Systems (MLSys), 2023. https://arxiv.org/pdf/2212.03332
I. T. Sanusi, S. S. Oyelere, H. Vartiainen, J. Suhonen y M. Tukiainen, "A systematic review of teaching and learning machine learning in K-12 education," Education and Information Technologies, vol. 28, n.º 5, pp. 5967–5997, 2023. https://link.springer.com/article/10.1007/s10639-022-11416-7
O. Spantidi, "Theory is cool but so are microcontrollers: Computer science student reactions to Arduino TinyML", en Proc. 57th ACM Technical Symposium on Computer Science Education (SIGCSE TS 2026), 2026. https://doi.org/10.1145/3770762.3772499
[15] V. Caravella, S. Yassine y S. N. Kadry, "Test the capability of Arduino TinyML for machine learning", en EAI ROSENET 2023, Springer, 2024, pp. 161-175. https://doi.org/10.1007/978-3-031-64495-5_12
J. Lin, L. Zhu, W.-M. Chen, W.-C. Wang y S. Han, "Tiny machine learning: Progress and futures", IEEE Circuits and Systems Magazine, 2023. https://arxiv.org/pdf/2403.19076
[17] Congreso de la República de Colombia, “Ley 1581 de 2012, por la cual se dictan disposiciones generales para la protección de datos personales”, 2012, https://www.funcionpublica.gov.co/eva/gestornormativo/norma_pdf.php?i=49981
Ministerio de Educación Nacional de Colombia, “Orientaciones curriculares para el área de Tecnología e Informática en educación básica y media”, 2022, https://www.colombiaaprende.edu.co/sites/default/files/files_public/2022-11/Orientaciones_Curricures_Tecnologia.pdf
How to Cite
Downloads
Downloads
Published
Issue
Section
License
Total or partial reproduction of the documents published in the journal is authorized only when the source and author are cited.
| Article metrics | |
|---|---|
| Abstract views | |
| Galley vies | |
| PDF Views | |
| HTML views | |
| Other views | |
