IoT-Enabled Smart Farming System Based on Sugeno Fuzzy Logic for Land Monitoring, Automated Irrigation, and Fertilization Recommendation

Abstract

Conventional farming systems are widely practiced without considering field conditions, including temperature, soil moisture, and soil pH, leading to inaccurate irrigation, fertilization, and crop selection. This study designed and implemented an Artificial Intelligence and Internet of Things-based smart farming system using the Sugeno Fuzzy Logic method to support real-time land monitoring, fertilizer recommendations, and automatic watering control. An experimental method was applied through literature review, hardware and software design, multisensor integration, implementation, testing, and evaluation. The system integrates an ESP32 controller, a DHT11 sensor, a soil moisture sensor, a soil pH sensor, an ultrasonic water-level sensor, an ACS758 current sensor, an Android-based Blynk application, and solar panels as the energy source. The results show that the system displayed temperature, humidity, soil pH, reservoir water status, fertilization status, and watering status in real time via the Android app. Sensor testing showed reliable performance, with the DHT11 producing average differences of 0.675°C for temperature and 1.75% for humidity compared with BMKG reference data. The soil moisture sensor identified dry, normal, and wet soil conditions with average values of 23.73%, 47.50%, and 80.23%, respectively. The pH sensor recorded average values between 6.48 and 6.72, while the ultrasonic sensor achieved an average reservoir-level error of 0.62%. Solar panel testing produced average voltages of 13.89 V on the multimeter and 13.61 V on the solar charge controller.