SBIR Phase I: A Precision Autonomous Fluid Planting System for Pre-Germinated Vegetable, Flower and Specialty Crops

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project is to provide economic benefits to farmers and consumers as well as broader societal benefits in the U.S. and globally for many different farm operations (e.g., vegetable, flower, and herb growers as well as greenhouse operations) and farm customers. Higher farm productivity directly benefits society by expanding the availability of nutritious, affordable food. Vegetables are vital U.S. crops in terms of food sources and farm profitability. One of the main challenges for vegetable farmers that limits productivity and total yield is the need to grow starter plants for the vegetables in greenhouses (or to purchase them) and then transplant them into tilled soil, which is highly labor-intensive and expensive. This system replaces that process with a system that plants pre-germinated seeds directly into the field with precision at a projected cost of 26% of the incumbent methods. The technology is an all-electric system that reduces the use of fossil fuels used in planting 100% and is 75% lighter weight than incumbent transplanting solutions using diesel tractors. The technology also addresses the increasing scarcity of farm labor with an autonomous planting system using automatic guidance and machine learning. This Small Business Innovation Research (SBIR) Phase I project will work to solve the problem of separating pre-germinated seeds in a batch of thixotropic gel such that one and only one pre-germinated seed is planted at each desired location in real-time under actual field planting conditions. Plant scientists have proven in numerous scientific studies that planting pre-germinated seeds increases production and lowers costs. Reliable field planters have not been successful, however. The primary research objective is to prove that an autonomous seed-singulation system combining flow controls, seed detection, active machine learning, singulation, and extrusion in real-time can meet the requirements of the field to prove commercial viability. The project research will begin by using discrete event simulation to model the singulation process used to obtain optimal parameters for the system. The model’s output will be used to develop control algorithms and programs. A prototype autonomous system will be developed to test the algorithms in actual field conditions. The anticipated result of this research is that the system can reliably plant one and only one seed at each desired location at least 90% of the time. 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.