SBIR Phase II: N-SMART, a highly-scalable, robot-enabled solution for managing nitrogen fertilizer applications to maximize a corn farmer's yield, profitability, and stewardship

The broader/commercial impact of this Small Business Innovation Research (SBIR) Phase II project lies in the development and commercialization of a novel artificial intelligence-based strategy for applying nitrogen fertilizer to corn fields. Using automated, unmanned ground vehicles (robots), small test plots will be created to stress test and develop an accurate simulation of each field. These simulations will determine the optimal nitrogen fertilizer application rates based on field characteristics, rainfall, and growing conditions. Harvest data from the small plots will validate that the expected benefits are delivered to farmers. Over 10% of the energy consumed in the U.S. agricultural sector is used for producing nitrogen fertilizer for corn. By enabling United States corn farmers to optimize nitrogen fertilizer use, this technology will improve the efficiency of nitrogen fertilizer application, thus reducing the energy intensity and carbon footprint of U.S. agriculture. Efficient use of nitrogen fertilizer will lower input costs and maximize yield potential, thereby improving the profitability of U.S. farmers. Moreover, this project will reduce nitrogen fertilizer pollution, which threatens human health, degrades aquatic ecosystems, and emits greenhouse gases that contribute to climate change. This Small Business Innovation Research (SBIR) Phase II project addresses the problem of nitrogen fertilizer pollution during the production of corn by commercializing a novel approach to simulate corn fields. Farmers rely on simulations to predict how much nitrogen fertilizer will be required by their crop for the remainder of the growing season, yet there is no established method to confirm if parameters, such as percent soil organic matter, are set correctly for a particular field. Using small test plots with varying amounts of applied nitrogen, an artificial intelligence-based approach will be used to find the best simulation of the field that matches simulated and observed nitrogen stress. Research objectives include improving estimates of observed nitrogen stress using an approach based in computer vision and machine learning, outfitting unmanned ground vehicles with the necessary cameras and processing capabilities, and testing the full approach on research plots and farmer fields in the Midwest U.S. The expected technical benefits of this project include providing corn farmers a reliable method to reduce nitrogen fertilizer, thereby reducing their highest input cost after purchasing seed, while ensuring that their crop has sufficient nitrogen fertilizer to reach its yield potential. 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.