SBIR Phase I: Advanced bioeconomic forecasting enabled by next-generation crop monitoring

Arable Labs, Inc., Princeton NJ

Investigators

Abstract

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project will be to empower farmers to capture a greater share of revenue from the marketing of their crops. Agriculture is a significant engine to the U.S. economy, and farming itself is vital to creating economically vibrant rural areas. Farmers are often at a disadvantage when it comes to capturing good prices from their crops because there are significant information asymmetries in the marketing supply chain. This project develops a combination of hardware and analytics that greatly improves crop forecasts at dramatically more accessible prices, which allows farmers and their trusted buyers to make more informed marketing decisions. An addition to the narrow application of sensing hardware and analytics for forecasting, the data collected by the platform can also be used by growers to make decisions that improve operational performance of complex agribusinesses and improve the agronomy of the farm. These tools make it easier to compare performance of crops to improve yields and reduce resource costs. Together this technology continues to raise productivity and profitability per farmer. This Small Business Innovation Research (SBIR) Phase I project integrates a novel plant and weather sensing platform with analytics that synthesizes data into actionable forms that can drive agribusiness decisions. The project bundles a suite of capabilities into a single hardware unit that includes sensing, communications, GPS, mounting, and solar power, which dramatically reduces the cost and increases the simplicity of collecting agricultural data. These data are uniquely designed to monitor crop performance and its sensitivity to weather and management. Data synthesis is a critical pain point in transforming raw numbers into insights for growers to act upon. By creating an integrated hardware platform, the data is poised to provide useful advice that allow a farmer to act on emerging situations, anticipate upcoming events, and even predict the future. A research objective will be to generate probabilistic forecasts that use the unique data from our hardware to estimate key crop growth parameters and project forward for an operational yield forecast. This coupling between highly informative quantitative in-field data and sophisticated ensemble-based parameter estimation and forecast techniques could dramatically improve marketing decisions and help farmers capture better prices for their products.