AI-ENGAGE: DEVA: Disease Detection and Effective Control using Versatile UAVs and UGVs in Apple Orchards

Apples are one of the most important fruits worldwide, supporting millions of livelihoods and economies around the globe. However, apple orchards face critical challenges from diseases such as apple scab, fire blight, and powdery mildew, which significantly reduce fruit quality and yield. Farmers traditionally rely on widespread applications of pesticides across entire fields, which is costly, labor-intensive, and environmentally harmful. This project develops a new approach that combines drones (unmanned aerial vehicles, UAVs) and ground robots (unmanned ground vehicles, UGVs) with smart sensors and artificial intelligence (AI) to detect and treat diseases only where needed, using precise amounts of pesticides. The system aims to help farmers reduce chemical use, save labor, and improve economic longevity by identifying early signs of disease and applying treatments in a precise, targeted manner. By working with apple growers in diverse locations, the project also ensures that the technology is practical and accessible for real-world farming. This platform also supports workforce development educational and training activities in robotics and AI. This project addresses the critical challenge of efficiently detecting and controlling diseases in apple orchards, such as apple scab, fire blight, and powdery mildew, which significantly impact crop health and productivity. The research aims to develop a novel, integrated robotic platform combining UAVs and UGVs with advanced sensing and AI technologies. UAVs equipped with multi-spectral and thermal cameras systematically, globally monitor orchard health from above, detecting subtle indicators of plant diseases before they become severe. Concurrently, UGVs operating on the ground use local, detailed sensor feedback to perform precise, site-specific pesticide spraying. Central to this approach is a learning-based model predictive control (MPC) framework, which adapts spray patterns in real time by continuously interpreting sensor data, environmental variables, and disease severity. Additionally, the project establishes a cloud-based IoT platform to integrate and analyze data collected by UAVs and UGVs, providing orchard managers with actionable insights and recommendations through user-friendly dashboards. The project plans to validate the entire system across multiple geographically diverse orchards to ensure broad applicability and robustness. Ultimately, the project aims to significantly enhance disease management efficiency, reduce chemical inputs, and advance precision agriculture practices globally. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.