AI-ENGAGE - Smart Scout: Empowering farmers to monitor and manage soybean lodging and estimate yield using flexible AI-Enabled systems

About This Grant

Soybean farmers across the world face a common problem—plants falling over before harvest, a condition known as lodging. This not only lowers yields but also makes harvesting harder and less efficient. One major cause is a hidden pest called the soybean stem borer, which is difficult to detect. Today, most farmers rely on time-consuming, manual methods to spot problems, which are often inaccurate and inconsistent. This project brings together partners from the Quad countries to create a new tool called Smart Scout. It uses advanced cameras and artificial intelligence (AI) to help farmers quickly and accurately detect damage, estimate yields, and make better decisions during the growing season. The system is flexible—it can be used by hand or attached to farm vehicles—and provides easy-to-understand, visual insights right in the field. While it currently focuses on soybeans, Smart Scout is designed to work with many other crops, offering a scalable solution for improving food production. By helping farmers make timely, data-driven choices, this tool can enhance productivity, reduce losses, and support economically prosperous agriculture around the world. Modern agriculture increasingly demands timely, accurate insights to manage crop health, yield potential, and environmental stresses—challenges made more urgent by climate variability, labor shortages, and sustainability goals. This project proposes Smart Scout, a user-inspired, AI-enabled computer vision system designed for real-time monitoring and decision support in soybean production, with adaptability to other major crops. The system will integrate visual data on pests, plant physical traits, and yield indicators to provide standardized, georeferenced insights at the field level. Its modular design allows deployment as a handheld tool, robotic platform, or machinery-integrated system, supporting scalable, flexible adoption. A core goal is to offer intuitive, actionable dashboards that both visualize AI reasoning and build user trust in the technology. The project will leverage collaborative data from QUAD partners to strengthen model accuracy and relevance across diverse production environments, while an independent testbed will validate technical performance and user experience. This unique AI-driven approach has broad potential to inform crop planning and management decisions, advancing precision agriculture through data-enabled, transparent, and adaptive tools that can be extended to a wide range of cropping systems. 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.