Towards sustainable cropping: AI-driven precision agriculture for optimal water and pesticide use via drones and soil sensors

Abstract

Artificial intelligence (AI) combined with drones and smart soil sensors is transforming the field of precision agriculture to an uncharted level where optimal water and pesticide applications have never been realized before. This article provides a detailed analysis as well as a simulation-based validation of an AI- enabled precision agriculture framework for efficient use of water and pesticides. We test the integration framework of drone remote sensing, IoT soil sensors, and machine learning algorithms in a closed-loop cyber-physical system (CPS) by quantitatively evaluating it with a 100ha farm applicable discrete-event simulation model. Our simulations show that using this AI-empowered approach for irrigation results in 35% reduced water consumption and 80% less pesticide being used, while also increasing crop yield by 5-8%. The simulation also shows a 30% decrease in operation costs and a 25% return on investment with technology pay-back after 2.3 growing seasons. Critical to this performance is the combined data fusion of spatial drone imagery with temporal soil sensor data, which supports high-confidence diagnostics and directed interventions. The simulation model also uncovered a positive feedback loop between system dynamics and improvement across time, in which execution data drives AI prediction guidance for several seasons. But despite barriers to the ability to achieve cost penetration, the simulation-validated analysis of its economic and environmental dividends makes a strong case for the role that AI-powered systems can play in facilitating sustainable agricultural intensification.