Oyster Mushroom Cultivation Monitoring and Control with Size Quality Prediction Algorithm via Adaptive Neuro-Fuzzy Inference System (ANFIS)

Cultivating oyster mushrooms is prevalent in tropical countries and is considered one of the marketable and edible types of mushrooms. Using sensors and actuators, controlling their ideal environmental conditions could produce better quality mushrooms. As a result, this cultivation and growth monitoring is possible using temperature, humidity, carbon dioxide (CO2), and lux sensors with actuators such as exhaust fans, misting systems, LEDs, and speakers with thunderstorm audio. Previous studies used Fuzzy Logic Control; however, this research utilized Adaptive Neuro-Fuzzy Inference System (ANFIS). The study aims to monitor and control the environmental conditions inside the mushroom using the components mentioned above and predict the size classification and measurement of the oyster mushrooms. The system was built on a wireless communication medium using a WI-FI module embedded on Arduino Mega and Raspberry Pi. The gathered data was trained in MATLAB using the FIS application via hybrid propagation. Data training and checking produced MSE of 1.78 and 0.93, respectively. During the actual testing, the comparison of the actual and predicted sizes obtained an average of 96.02% of accuracy. Furthermore, the proponent did a two-sample t-test to verify the accuracy of the prediction, which resulted in the actual field test with no significant difference with the predicted values hence, validating the results of this study.