Solarid AR, an Arkansas-based agtech company, received $981,168 from the National Science Foundation’s Small Business Innovation Research program to continue development of a smart insect control system. Roughly $340,000 of this award will go to researchers at the U of A who are incorporating a camera — powered by artificial intelligence — into Solarid’s existing insect trap system to identify and count a range of insects.
Ashley Dowling, a researcher in the U of A System Division of Agriculture and professor of entomology and plant pathology, and Khoa Luu, an assistant professor of computer science and computer engineering, are leading development of the smarter insect trap.
The system is designed to simplify monitoring of plant phenology and pest populations by processing the appropriate data in real time through a central dashboard that can be accessed with a mobile app. Day or night, growers will be able to find out which pests and in what concentrations are in their fields during the different stages of plant growth through a season. This information will be indispensable for early, accurate identification of pests and for timely responses to reduce inputs and crop loss.
Ongoing work has focused on building the database of insects the system needs to recognize and improving the reliability of detection within a species. To assist with this the team will work with the University of Georgia Center for Invasive Species and Ecosystem Health to add EDDMapS, which is the largest invasive species and pest database in North America. They will also integrate the center’s digital applications to create a robust framework for disseminating information to site managers, municipalities, organizations and the public.
The goal is to miniaturize and weatherize the system so that it can be easily deployed to run off the solar cells that currently power Solarid’s insect traps.
This next phase of the work will shift to California crops, where the team will partner with industry experts at University of California Agricultural and Natural Resources and the USDA Agricultural Resources Services. The team will train AI algorithms to identify fall armyworms, which eat rice, and navel orangeworms, which, despite their name, are a major threat to almonds, though they also menace pistachio, fig and pomegranate production as well.
Randy J. Sasaki, managing partner of SolaRid AR, noted that both the company and the scientists were extremely proud of the work that led to the latest round of NSF funding. “It is a significant milestone after five years of development,” he said. “It is also an endorsement by authorities that the AI technology has been developed and of the importance of commercializing the technology that is intended to make America more competitive.”
Dowling added: “It’s exciting to know we have the funding to take the next step toward producing a marketable, field-ready unit and that a few years from now our invention could be in fields across the U.S. helping farmers monitor and protect their crops.”
Producing a compact, weather secure and affordable smart trap for use across a range of cropping systems anywhere in the world is the final goal of this project.
Source : uark.edu