MKU-1: Modular Real-Time Sensor Networks and Edge Computing for Precision Crop Management in the Eastern Economic Corridor

2026-01-09T10:00:33+07:00

This paper presents an Internet of Things (IoT)-based smart farming framework for the precision cultivation of medicinal and economic crops in Thailand’s Eastern Economic Corridor (EEC). The proposed system was experimentally validated using Capsicum annuum (chili pepper) and Ocimum tenuiflorum (holy basil). The framework integrates a multi-sensor network for continuous monitoring of key environmental and soil parameters, including air temperature, relative humidity, light intensity, soil moisture, soil temperature, electrical conductivity, soil pH, and NPK nutrient levels. Sensor data are processed locally by the MKU-1 modular control unit, which functions as a lightweight edge-enabled automated control platform. At the field node, an ESP32 microcontroller performs sensor acquisition, data validation, timestamping, threshold-based decision execution, actuator control, and cloud synchronization without requiring immediate cloud feedback. In the current prototype, computationally intensive machine-learning analysis is performed offline or at the cloud level, while the field node is responsible for real-time local control of irrigation, fertigation, and supplemental lighting. Field experiments comparing system-assisted cultivation with conventional non-system cultivation demonstrated improved crop growth and greater environmental stability. For chili pepper, the system-assisted plot achieved a 16.4% higher average plant height than the non-system plot. For holy basil, system-assisted cultivation achieved an 11.5% higher average plant height and a 24.1% higher average number of leaves across the cultivation period. The system also maintained more stable soil moisture conditions, with average soil moisture values approximately 61.8% higher for chili pepper and 31.1% higher for holy basil compared with non-system cultivation. Feature-importance analysis showed that air temperature, light intensity, and relative humidity were the most influential factors affecting cultivation performance, with importance scores of 0.41, 0.17, and 0.13, respectively. The analytical model achieved an overall accuracy of 96.55%. Economic analysis for a representative 10-rai cultivation area indicated an estimated annual water-cost saving of THB 20,000, equivalent to THB 2,000/rai/year, and an estimated return on investment of approximately 25% per year. These results confirm that the proposed MKU-1 framework enhances cultivation efficiency, improves environmental control, reduces resource variability, and supports sustainable smart agriculture practices aligned with Thailand’s BCG and EEC development strategies

Journal of Engineering Technology Access (JETA) (Online)

MKU-1: Modular Real-Time Sensor Networks and Edge Computing for Precision Crop Management in the Eastern Economic Corridor