Sistema de detección de anomalías para una huerto inteligente

Abstract

In the educational context, school gardens are a pedagogical tool that promotes technical learning and sustainability. However, their implementation is hampered by barriers such as lack of maintenance, resistance to change, and lack of institutional support, among others. Based on this, the need arose to offer a tool that adapts to the constant changes demanded by today’s digital society. During project implementation, a previously installed sensor network was used. An additional node operating under the ESP-NOW communication protocol was added to this network to mitigate limitations in terms of data storage and processing. A monitoring system was designed to monitor the care of the plants assigned to each node and monitor data in real time. The components used include HD-38 and DS18b20 sensors, an ESP8266 microcontroller, and servers for data storage. A machine learning model based on a Random Forest algorithm was also implemented to detect abnormal conditions. The results demonstrate that the implemented system detects anomalous conditions with high accuracy. The sensor network continuously collects data. The monitoring system converts complex data into an intuitive experience for students. It is concluded that smart gardens are a tool with great educational, economic, and social potential. Keywords: School gardens, Integrative Project, anomaly detection, sustainability, sensor network