Precision Farming with Autonomous Drones: Real-time Data and Efficient Fertilization Techniques

The study explores the challenges farmers face due to environmental issues and the growing global population. It suggests that digital technologies, particularly drones, can improve agricultural productivity. The research focuses on developing a drone capable of autonomously fertilizing fields, providing real-time data and insights. With Raspberry Pi, multispectral sensors, GPS technology, CNN, Tiny Yolo v4, and AI-driven analytics, the drone can deliver chemicals to targeted areas and return to its base station, significantly reducing farmers' efforts. This model conducted four tests to evaluate the drone's accuracy in a 60 m2 cropland area. It utilized a single drone, although cooperative operation with multiple drones could enhance efficiency. As a result, the findings show that optimal results were achieved with a load of 500 grams at a speed of 28 km/h, achieving an overall precision fertilization accuracy rate of 98.3%. This payload is sufficient for small-scale operations, but its efficiency depends on crop density and the required fertilization level. The drone can operate continuously for 20 minutes at full capacity. In the test period, with a fly time of 3.86 minutes, the drone covered approximately 59 m2 cropland area. Compared to traditional human labor, the drone demonstrates higher precision and reduces environmental impact by minimizing chemical waste. The experiment was conducted in a controlled agricultural field to simulate realistic farming conditions. The use of drones in field inspection and pesticide application is examined in this study, emphasizing the methodology, controllers, and hardware/software components. It illustrates how AI and machine learning improve agriculture's technical efficiency and environmental sustainability.