Nathan Boyd, a University of Florida Institute of Food and Agricultural Sciences (IFAS) professor of horticultural sciences, says about 80% to 90% of his research now centers around artificial intelligence (AI). While precision ag is nothing new, he says how AI will help integrate into precision agriculture shows so much promise.
“If you go back and you can almost hear in the articles this desire by the authors that they wish they could do certain things, but they did not have the tools to do it,” he says. “And AI is what, all of a sudden, gave us the capability to do all these things that they had wanted to do for decades but couldn’t.”
Boyd says he’s paired his field trial expertise with Arnold Schumann, a IFAS professor of soil, water and ecosystem sciences, to unlock the future of agriculture.
“I think it’s the next agricultural revolution,” he says, “I think it’s going to completely transform how we grow food. But I think, we are only at the very beginning. We have not yet seen what this is going to be.”
In Boyd’s recently published findings, a research team compared the use of AI technology with herbicide applications on weeds in tomatoes in three locations. The research team compared results of herbicide applications on weeds growing in bare soil between beds covered with plastic mulch; herbicide applications in transparent holes on top of the plastic mulch beds; and targeted herbicide applications on nutsedge, which punctures plastics, with the use of machine vision to locate the weeds.
The results, Boyd says, show targeted spray methods can work effectively for vegetable crops and lower input costs related to weed management by 75%. He says there’s also ongoing projects in strawberries, tomatoes, peppers, watermelon and more.
But part of the motivation to study tomatoes is that there are herbicides registered in tomatoes to treat nutsedge. One thing Boyd says is that the industry needs to understand that AI is a tool that works in tandem with herbicides that target weeds.
“AI is not going to solve everything,” he says. “It’s a way to make things more efficient, but it doesn’t remove the need.”
Boyd says his research team has had the support Artificial Intelligence Academic Initiative Center in its research and has worked with growers to develop the techniques.
“We said, ‘OK, we need to talk to growers from the very beginning and then so that when we design it, it’s actually designed in a way that they could use it,’” he says.
The other thing Boyd says was important with his study was to show that the AI technology can integrate into a grower’s existing system without having to modify growing systems.
“We wanted to design it in a way that could be used in current systems,” he says. “We’re also trying to design it so it’s adaptable. We have one basic AI program and system that we’re designing to use across multiple crops.”
Boyd says growers also want to be able to maintain their equipment.
“We have to design it as a modular approach,” he says. “So, you could kind of plug and play, so you can remove things and replace them if they need to be fixed.”
Boyd also says growers could see benefits beyond just a reduction in inputs. Targeted sprays can reduce the chance for crop damage from herbicides. Plus, AI systems with vision and mapping techniques can help growers better monitor what’s going in in their fields. And this goes beyond just scouting and treating weeds; AI systems will also help growers catch the presence of diseases earlier.
“Most scouts look at less than 1% of the plants in the field, just because of time restraints, but this machine going through would be looking at every single one,” he says. “That’s a big difference and much earlier detection means not only do you save because you’re not only spraying where it’s needed, but you also save because you’re able to spray things earlier.”
