Design and Development of an IoT-Driven Automated Plant Management System: Integrating Machine Learning and Mechanical Innovations for Optimized Growth and Irrigation

Efficient plant management is critical for sustainable agriculture and resource conservation. This study presents an innovative automated system that leverages IoT and machine learning (ML) technologies to optimize plant growth and irrigation practices. The system integrates a sun-tracking module, automated watering mechanisms, and robust ML models—linear regression, random forest, and gradient boosting regressors. Environmental factors such as sunlight exposure, temperature, humidity, soil type, and water frequency were analyzed to predict optimal watering hours. The methodology involved feature engineering, model evaluation, and system design, emphasizing minimal manual intervention. Results indicate that linear regression outperformed other models with an MSE of 0.272 and an R² of 0.880, highlighting its simplicity and efficiency. A fidget spinner-inspired water distribution mechanism and Arduino-based sun trackers were incorporated to ensure even water distribution and optimal sunlight exposure. The proposed system demonstrated significant improvements in water-use efficiency, durability, and convenience.