I-Corps: Translation potential of an Artificial Intelligence (AI)-driven agricultural intelligence system for controlled-environment agriculture

Award ID: 2627948

This I-Corps project is based on the development of a farm management system that uses artificial intelligence (AI) to control growing conditions such as light, temperature, humidity, and nutrients within enclosed structures like greenhouses or vertical farms. Currently, one of the primary contributors to food waste in controlled growing conditions is crop loss due to late detection of pests, diseases, or nutrient deficiencies. In addition, energy demands associated with using controlled conditions have raised concerns about long-term sustainability. This technology is designed to monitor images for problems, uses sensors to monitor the environment, employs machine learning models that evolve continuously and adjust to changing environmental conditions, and requires human approval before executing a task to identify plant stress, nutrient imbalance, disease risk, and energy related inefficiencies earlier and with greater consistency than conventional manual methods. The AI-based detection and early intervention capabilities reduce the likelihood of catastrophic crop failure, which may improve yield reliability and reduce inputs such as water or nutrients. The system also lowers electricity consumption by enabling operation only when necessary. This technology may extend beyond individual farm operations to address systemic challenges in sustainable food production, making local food production more economically viable. This I-Corps project utilizes experiential learning coupled with first-hand investigation of the industry ecosystem to assess the translation potential of an artificial intelligence (AI)-based platform for controlled environment agriculture such as hydroponic and greenhouse operations. At its core, the technology functions as an intelligent supervisory control system that combines computer vision, multi-modal sensors, and adaptive machine learning to perform continuous monitoring, decision-making, and actuation across nutrient delivery, pest and disease detection, and fertigation scheduling. Unlike conventional systems that rely on static schedules or always-on automation, this system uses AI to determine when action is necessary, allowing pumps, dosing units, lighting, and auxiliary systems to operate only when required. This allows the system to minimize runtime of energy-intensive components; addressing a fundamental economic bottleneck in controlled environment agriculture that existing vertical farming systems have failed to overcome. Information may be interpreted more effectively through this integrated AI framework to improve plant health, environmental conditions, energy use, and decision making. 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.

NSF Program Director: Ruth Shuman