Development of an Integrated Mushroom Cultivation System Based on Sensor Array and Machine Learning

The cultivation of oyster mushrooms (Pleurotus ostreatus) in Indonesia is growing rapidly and has strong economic and environmental potential, yet farmers struggle to meet increasing demand due to inadequate infrastructure and limited environmental control technologies. This study develops an Integrated Mushroom Cultivation System that combines a multi-point sensor array with Machine Learning to optimize microclimate regulation in a controlled chamber. Environmental parameters—including temperature, humidity, and light intensity—are continuously recorded in a big-data structure, enabling multi-sensor evaluation and more accurate, robust estimation of the cultivation conditions. A Neural-Network–Assisted PID controller is employed to process these sensor readings and generate automated control actions, particularly for misting and ventilation, when conditions deviate from optimal thresholds. Field data from 1–25 August 2025 show relatively stable conditions, with average temperatures around 24 °C and humidity within the optimal fruiting range at 80.44%, while light-intensity anomalies are attributed to direct sensor exposure to the light source. The findings highlight the importance of sensor calibration and anomaly detection to ensure reliable automated decision-making and suggest that implementing these mechanisms can enhance environmental stability, improve production quality, and support more adaptive precision agriculture systems for oyster mushroom cultivation.