Optimizing Lettuce Crop Yield Prediction in an Indoor Aeroponic Vertical Farming System Using IoT-Integrated Machine Learning Regression Models.

The rise in agricultural innovation has led to the use of sustainable farming practices, such as aeroponics, which increase crop production. Aeroponics, a soil-free indoor precision farming system, cultivates crops using vertical towers, garnering global attention for its environmentally friendly and productive cultivation methods. Aeroponic systems can grow lettuce, a popular green-leafy vegetable, quickly and with minimal water usage. However, yield prediction is a tedious task in real-world scenarios. To efficiently predict lettuce yield, various scientific experiments have integrated IoT and machine-learning techniques. This research work utilized various machine-learning regression models, including linear, support vector, random forest, and XGBoost, to estimate lettuce yield based on specific growth parameters such as pH, EC, temperature, total dissolved salts (TDS), turbidity, humidity and light. After implementation, the results showed a high prediction accuracy of 93% and minimal error rates produced by the XGBoost regression model when compared with the other regression models. Further, fine-tuning the model parameters enhanced the XGBoost model's performance, enhancing its generalization capability to handle new realtime data. This indicates that optimizing the lettuce yield involves not only using indoor aeroponic farming methods but also utilizing advanced sustainable food production systems. [ABSTRACT FROM AUTHOR]