How the Center for Applied Artificial Intelligence Is Already Changing the Face of Florida Agriculture
by REBEKAH PIERCE
Could artificial intelligence be used to grow the perfect tomato? Nathan Boyd, Associate Director for the Center for Applied Artificial Intelligence in Agriculture at the University of Florida, certainly thinks so.
The idea came to him after visiting different applied robotics facilities, including one in Sydney, Australia.
“I was impressed by the approach that some of these facilities [took] and how they were addressing problems,” he says. “I knew all along that automation of labor and the cost of labor are key issues. I already had some familiarity with AI, but thought there could be a bigger vision.”
Upon his return to the university, Boyd created an informational video highlighting what he thought his vision could look like in real life. “I gave it to my supervisor, who gave it to other people, and then we went around the state talking to people and organizations that are ag-related.”
The pitch was a resounding success, partly due to the fact that Boyd wanted to use AI to improve specialty crop production in Florida.
“One of the biggest issues is the cost associated with labor compared to what other countries have to pay for labor,” he explains. “The solution in my mind was automation.”
The university has received $30 million from the state of Florida to construct the facility, which is currently in the design stages with an architectural firm.
With an anticipated completion date in 2027, the center in Balm is already collaborating with other universities worldwide.
“We want to be able to look everywhere we can to find solutions for the growers that are here, as well as solutions in other places,” Boyd says.
The team focuses on technology that can be commercialized within a relatively short time frame.
“We aren’t taking anything off the table,” says Boyd. “We are in the midst of defining how we can bring a novel technology out to investors, or out to equipment manufacturers that might be interested. The goal is to get tools to growers. We don’t build and sell equipment — that’s not our role. We are trying to create connections and a collaborative environment.”
The center started working with three crops — tomatoes, strawberries, and watermelon — and plans to add peppers as a fourth area of research next year.
The priorities differ some between crops. For strawberries, as an example, many of the biggest challenges are in planting, which is all done by hand. Because of this, using AI to develop solutions for transplanting strawberry crops could be groundbreaking.
That said, some priorities hold common ground among crops. Targeted pest management technologies are just one example, with the center’s researchers thinking carefully about detecting, identifying, and treating problems more efficiently. “This reduces input costs and is better for the environment,” adds Boyd.
“Each crop is a bit different, but we try to think of it as a system,” he explains, calling out the need for solutions that can be applied throughout the lifespan of a crop. Strawberries produce runners that grow off the main plant and need to be trimmed, something that’s currently all done by hand.
Rethinking Entire Systems of Production
AI isn’t just finding solutions to streamline tasks; it’s finding solutions to streamline entire systems.
“AI is just a tool,” says Boyd. “It doesn’t solve everything, and it has to be integrated into the overall system or it’s not effective. This isn’t necessarily a limitation of AI, it’s a limitation of the adoption of AI.” After all, if you create a technology that’s too cumbersome or difficult to use, the likelihood of adoption is small.
Focusing on solutions to mechanize that labor is smart, but, adds Boyd, if, “I figure out how to solve the runner cutter problem but still have to do the transplanting by hand, I haven’t really helped [the farmer] because they still have to have all the people to do the transplanting. If we think of it as a system, we can reduce the whole burden on that system.”
Thinking about how AI fits into an entire production system is the main goal of the Center, says Boyd. It’s about finding ways to work smarter, not harder.
“We’re good at making things overly complex,” he said. “But we don’t need that complexity. What AI does that other things cannot do is its ability to extract from an image, a point of interest, data, and patterns…it gives us capabilities that we could not do even a short time ago and allows for real-time decision-making.”
Rebuilding Florida’s Workforce, One (AI) Step at a Time
We’ve all seen our fair share of sci-fi movies with robots taking over the planet. While Boyd admits that a common question he gets is, “Doesn’t this mean you’re losing jobs?” he isn’t concerned.
“Actually, no,” he says. “We don’t have a domestic workforce that does this stuff. You’re just changing the type of job, going from a low-tech to a high-tech position. In fact, there’s been instances where automation led to increased employment because the industry became more competitive, and it expanded.”
As the workforce continues to age — the average age of the Florida farmer is nearly 60 — the solutions that the center proposes couldn’t possibly come at a better time.
“With this new technology, we’re seeing a younger population [in the workforce],” Boyd concludes. “It’s a way to bring people back into agriculture.”
