An IOT Irrigation System using machine Learning
An IOT Irrigation System using machine Learning
Date
2024
Auteurs
Belharizi Nesrine
Nom de la revue
ISSN de la revue
Titre du volume
Éditeur
Résumé
This master thesis presents a smart irrigation system for small farmer communities (SFCs)
designed to optimize water use in agriculture through advanced sensor technologies and machine
learning (ML) techniques. The system employs Internet of Things (IoT) components, including
low-cost sensors such as the capacitive gravity SEN0308, water tension irrometer watermark,
and DHT22. An Arduino Uno collects sensor data and transmits it via LoRa communication to
an Arduino Mega 2560, which controls a valve and water flow sensor. ML integration enables
real-time analysis and decision-making to prevent over-irrigation by considering soil moisture,
temperature, humidity, and historical patterns. The system enhances water efficiency, reduces
labor costs, and improves crop yields, thereby contributing to sustainable agricultural practices.
Description
Mots-clés
Internet of Things, Irrigation, Low-Cost sensors, Machine Learning, Smart Farming