Data-driven Reduction Of Pesticide Spread in high volume orchard treatments with smart sprayers.

In fruit tree crops, which are vital to EU agriculture, efficient pesticide use is crucial for maintaining high yields and assuring market quality. However, traditional spraying methods often result in significant pesticide losses through drift and runoff, posing environmental and health risks. Although new smart sprayers equipped with pulse width modulation (PWM) valves are designed to control flow rates and minimise off-target losses, the droplet size variations caused by pressure fluctuations when spraying high volumes to control pests and diseases are still poorly understood, and often lead to uncontrolled pesticide dispersion to the environment. The DROPS project will optimise pesticide application in orchards by developing a novel methodology for predicting pesticide dispersion and improving spraying practices. Specifically, DROPS will directly measure droplet size in orchards and integrate computational fluid dynamics (CFD) models with machine learning to create a Decision Support System (DSS) that guides farmers in optimising pesticide applications. The project will adapt and validate droplet size measurement techniques for dynamic outdoor orchard conditions, using PWM-based sprayers. This will provide accurate data on droplet behaviour under varying environmental conditions and sprayer settings. The data will inform predictive models of droplet size variation, enhanced by machine learning algorithms for greater accuracy. These models will then be integrated into CFD simulations to predict pesticide dispersion in different orchard environments. In the final phase, the DSS will be developed and field-validated to provide real-time recommendations for minimising off-target pesticide dispersion. The system will be implemented in smart sprayers and tested with an industry partner to ensure practical application under real-world conditions.