Thursday September 13, 2018

Farmers to Use Machine Learning to Boost Yields and Collect Crop Data

Washington University and The Climate Corporation are working on revolutionizing the farming profession. Farmers in the next few years will be using sensors in their fields and farm equipment that collects data through a Seed Advisor program to be digitally analyzed by machine learning (ML) to boost yields of crops. The data generated from the sensors will allow sophisticated ML models to predict which hybrid seeds and fertilizer have the best chance to produce maximum yields in an environment. In trials against conventional soil sampling, the ML recommendations have an 80% win rate. The $8 million project is funded by The U.S. Department of Energy's ARPA-E TERRA program and includes multiple universities and other collaborators.

News Image

In 2017, Seed Advisor demonstrated an average advantage of 6 bushels per acre with a nearly 80 percent win rate in farmer field trials. 2018 testing is underway on more than 100,000 acres, with expanded pre-commercial testing planned for the 2019 growing season. The pre-commercial test, called the Climate FieldView Innovators program, is targeting approximately 200 farmer participants.