Technology is rapidly advancing in many different industries, but what about agriculture technology? Drones? Artificial intelligence? How does this benefit farmers and does it benefit shoppers?
UF/IFAS experts Kati Migliaccio and Nathan Boyd sit down to discuss the latest and greatest in farming technology. We discuss how advances in farming technology not only help farmers but how those benefits trickle down to the options shoppers have in the grocery store. (And if you’re scared of robots and artificial intelligence taking over the world – this one is for you, too. We clear that up as well.)
Artificial intelligence at UF/IFAS
More on HiPerGator
AI for precision agriculture
To learn more about UF/IFAS and how food IS our middle name, visit: ifas.ufl.edu/food/
Welcome to the Food is Our Middle name podcast. I'm your host, Tory Moore. And today we'll ask what is the deal with farm technology? We're talking drones, artificial intelligence and more.
When you think of farming, what comes to mind? Maybe a tractor or livestock of some sort. But what about drones or groundbreaking technology like artificial intelligence? And did you know there are even robots created to help farmers? Technological advancements in farming are not only fascinating, but important tools for farmers to feed more people with fewer land, farmers, and other resources.
Here to discuss the latest and greatest in farm tech are UF/IFAS experts Dr. Kati Migliaccio and Dr. Nathan Boyd. Welcome, Kati and Nathan. I'm excited to hear about some fascinating ag technology today.
So, before we get into the interview here on the podcast, we have a little game that we play to kick off each episode. Katie, I'm going to nominate you to play the game with us. And I'm going to ask you a series of fun food questions. There are no wrong answers, and the goal is to answer as many questions as you can before we run out of time. Okay. Are you ready?
Okay. Let me get my timer started. So, you'll have 15 seconds to answer as many as possible. Here we go. Favorite food?
Coffee or tea?
Food you hate?
Sweet or spicy?
Favorite meal of the day?
And what's your ice cream order?
Aha. Perfect. That's out a time. So, let's get into our interview. But first, I want to tell the audience a little bit about your roles here at the University of Florida and the type of work that you do in the ag-tech arena. So, Kati, let's keep going with you and tell me a little bit about your role here at UF/IFAS.
Thanks, Tory. So, my role here primarily is to serve the Agricultural and Biological Engineering Department as Chair. I'm also a faculty member as a professor in the department, and my research was in Water Conservation, Irrigation, Hydrology. Right now, I also have a role with the University of Florida Provost Office as Co-chair of the Quality Enhancement Plan that is focused on AI across the curriculum for the university.
Okay. And Nathan, what about you?
So, I serve as the Associate Center Director at the Gulf Coast Research and Education Center, which is just outside of Tampa. And I'm also a weed scientist by training. And my research program is focused on weed management and specialty crops, so fruit and vegetables. And we develop precision technology for weed pest control.
Wonderful. And we're going to talk about some of those technologies later in the podcast. But thanks for that. So, we know that technology is changing rapidly in many different industries, not just agriculture. But Kati, can you tell me about how technology has changed in the ag industry in recent years?
Sure. There's been lots of different changes.
You know, I'm from the water, the water side. And some of the changes on the water side have had to do with the kind of data we have available to make decisions. And that has, it has flowed into other aspects of agriculture as well. There's more data for almost every decision you can make on the field and AI is a powerful tool when you start talking about these types of data. And so that's one change has happened, is that the availability of data, the amount of data, the data that is, is available not only in terms of and time, but also space. So having all of this information available to us to make decisions helps us make better agriculture decisions using technology.
And some of the other changes that I've seen is movement into controlled-environment agriculture or urban, you might hear urban ag, where we're looking at production in different ways, and then we have this Walkley and also looking at how production of ecosystem services can go together.
So, there's a lot of great ideas that we can embed into agriculture because of the technologies we have and that we didn't have 20 years ago. And it's not just having the technology, it's the technology being affordable. And so, we're reaching a point where some things maybe we knew how to do it, but it wasn't a viable option at the time. It's becoming more viable in agriculture now.
Right. Okay, and when you say ecosystem services, can you break that down for us a little bit?
So, the world was in nature, right? And we came in and we built houses and we started cultivating farms and different things. And then we kind of moved nature aside. But nature has a purpose. And nature is completely circular system. It uses all of its waste. It thins for itself and it survives, right? And so, there are components of that that we can embed into agriculture, where it becomes a system that feeds into itself. And so, the waste can be fed into itself.
We can use it as storage for water. We could use it as a source for carbon for different things that actually have a greater impact overall instead of perhaps just growing one crop in one location, that that became a very, probably still is one of the most popular ways of growing crops. Then when you embed ecosystem services into that, you even get a greater value because you're getting more sustainability of the production, you're using your resources more efficiently and, and you're also helping the world outside of that agriculture production because you're providing a service that's greater.
That's great! Thank you so much for clarifying that. So, and then you talked about controlled-environment production. So, are we talking greenhouses when we say that? Is that an example?
That's an example. You may hear vertical farms or vertical ag. And you may hear containers. You're seeing a lot of people using containers for controlled-environment production like a, like an 18-wheeler, if you think about a container on a truck, things like that.
Oh, wow. Okay.
There's all kinds of ways to do it. And, you know, interestingly enough, COVID almost pushed us a little more into that world because we started seeing some of those pressures on supply chain and our food supply, where it's coming from, how we can move it around and food safety. And so -- good or bad, probably bad -- COVID did help push that a little further. But also, this technology we're talking about with AI had pushed it further because it's allowing us to do things in those controlled spaces that we were not able to do before in terms of being viable economically.
That was actually my next question was, is this something that's happened in more recent years, these advancements and these changes, or is this been something that's been going on for a long time? And it sounds like both.
Yeah. I mean, I think it's been going on for a very long time, but only certain crops probably.
You know, if you ever been to Disney, they've been growing lettuce a long time at Disney, right? So, there's, there are crops have been grown for a long time in these systems that we're now moving toward other crops, are looking at breeding for those systems, we're looking at nutrition. So, can we produce a higher – a crop that has higher nutritional value for the same amount of inputs of water and nutrients?
And there's also – the, the biggest thing when you talk about controlled environments is energy. So, thinking about how, what kind of energy you're going to need, where you're going to get that energy and how you can make it – is, as -- how you can make it use the least amount of energy possible, because that's one of your biggest economic barriers when you start talking about controlled environments.
Okay. Okay, that's so interesting. So, I want to talk about why technological advancement is so critical to farming. Nathan, can you talk to us about the importance of these advancements?
Sure. There's a lot of reasons why it's important. One of the ones we hear the most about and recently is labor, especially crops in particular require a lot of hand labor. And so, there's a lot of push because there's issues with getting labor to do that type of work. So, there's a lot of push to try to automate things and develop technology that can help with that process. But that’s just one, you could name many things.
Another thing to think about is just markets. So, if you had a better way to predict when you're going to harvest a particular crop and how much of it, that crop you were going to harvest, you might be able to market your, what you're producing more effectively and increase your bottom line.
Right, because we've talked about in the podcast before the seasonality of crops, especially here in Florida. You know, we're growing for different markets. We're growing different times of year, right? So, having more data and understanding how you can better serve that market and what pricing and everything could be really helpful. So, that's another aspect that ties into something we've already talked about on the podcast.
So now I want to talk about some specific examples of ag technology that you guys have worked on or that you're aware of at UF/IFAS, so we'll start with Nathan: so talk about maybe the coolest farming technology you've seen come about in recent years that you worked on.
Oh, there's so many. (Laughter) There's so, there's so much happening at UF and so many technologies. I mean, the one that I know the most about is the one that I work on. So, I'm biased and I'm very interested in that, but it's just one amongst many fascinating technologies.
So, what we do is we use precision spraying technology. So, what I mean by that is we equip spray equipment with cameras and with deep learning models, which are just a type of artificial intelligence that can be trained to find and identify things. So, if you have -- if you put those and you integrate them into a spray system, you can, rather than spraying herbicide over an entire field as you're driving through, it finds the weeds and then only applies the herbicide where those weeds occur. And of course, the result is: you use a lot less herbicide, which is good for the environment and it's good for the growers, they save money. You reduce the amount of pesticides that are on the food that people are going to buy to eat, so fewer pesticide residues, which is good for everyone.
So, there's, there's this all of these benefits, and that's just one small example that I do in my lab. But you can think about how this technology just, just that one technology that I mentioned, the ability to find and identify something.
We've been trying to do this for decades, but it's really been the development of deep learning that made it more effective and more cost effective, as Kati mentioned earlier. So now you can find diseases, you can find insects, and if you can find them, you can usually do something about them. And that that ability really affects pest management and agriculture.
So, I want to hone in on the concept of deep learning. And I know it's probably really technical, you know, to really dig into how it works, but can you give the 10,000 foot view of what deep learning is and how AI learns, so to speak.
So, you can kind of think deep learning was kind of developed around the idea of trying to simulate what happens with neurons in the brain. Of course, much, much simpler.
So, deep learning, you can think of it as a computer program that processes and looks for patterns in images, or if we're talking about image recognition.
So, in the past we had machine vision, which you could tell a machine to find something. But now, instead for deep learning, it's looking for the patterns that within, within the images, you're feeding it. So, it’s -- I don't want to use the word deciding -- that the algorithm is identifying consistent patterns and then using that to determine that this is whatever you're trying to get it to identify.
So, for example, if I wanted to identify a weed species. I will feed the program thousands of pictures of a particular weed species. And in the process of doing that, it's going to find a consistent pattern that makes that weed unique from all the other images I sent it. And that's -- then it's going to be able to define. And you can do the same thing with insects and diseases and, and a whole range of things, even nutrient deficiencies and leaves.
That's fascinating. And what kind of accuracy are you seeing with something like this once it, quote unquote learns and it's out and operational? Is it, you know, I guess what is the error there?
It depends on how good your program is, how quick your images are, how many images you have. The really cool thing about this technology is you start out with, you know, we just trained a model and we had 13,000 images. That's a lot. But once you have it trained and you're running it in the field and it’s collecting video with cameras, it can continually train itself to get better and better and better.
Keep learning basically.
Basically to keep learning. So, the accuracy depends on the amount of effort you've put in and the amount of images you're giving it and all those types of variables. So, with some, we did a project a while back where we were looking at specific weed species and turf. And we were right anywhere from 90 to 95% accurate.
We've had other scenarios where the models weren't nearly as good and you're only around 50% accuracy. But the point is, is the more you train it, the better it gets.
And you don't, once you've trained it, you don't have to, you don't -- It's like kind of like riding a bike. Once you know how to ride a bike, you know how to ride a bike. And these program, once they can detect weed, they get better and better at it. But you don't have to start at point zero every time. You just keep building upon what you've already trained. So, if I train a model to detect Weed A, I'll take that same model and teach it Weed B and it's more efficient than starting at point zero.
That's fascinating. That's really, really cool. So, Katie, do you have an example of Ag tech that is being worked on here at UF/IFAS that you want to share and talk about?
Sure. First, I want to say that is the best definition of deep learning I've ever heard. So, congratulations.
And so, I think the one I will share is, is, is a simple one, but it's one that has made a lot of impact on our growers here. And I think part of what I like about it is the transition from something from a research laboratory to the field and then to, to industry to be used in. And that has to do with drones and the use of drones to take images, particularly of, of trees.
One of the big challenges some of our, our citrus growers have is counting trees, which seems very simple, but it actually is quite complicated and it's tied to insurance. And so, I'm creating a mechanism that they can use to count the trees using drone data, which is very easy to fly over. It takes much less time than someone having to count each tree.
You know, if you've ever worked on a farm, there's always more work than people. There's always more work to be done. And so if you could take some of these things out of that workload and put them into drone load and they can do a better job or as good of a job, then you created a system that's going to work better for the farm because those people that would be just walking the field, counting trees can be do something else more useful for the agriculture production system.
So, to me, some of these very simple things that have been created where we're using drone data to count trees, using drone data to look at water stress and nutrient stress to help with more precision type agriculture -- I think that’s some of the technology that is a kind of low-hanging fruit technology it’s things that we can do and can easily transfer and be useful today for growers.
One of the other things Nathan brought up was the imaging when she figured out what a mite looks like, you can use the technology to find mites. And that's another really great tool for, for AI -- is to use cameras to take pictures of leaves or to use drones to fly over and take pictures and to look at what is the, the past in the field so that we do use our chemicals better and we identify problems before they get out of control, because once again, it's really hard to have a person walk a whole field. It's much easier to have a drone fly a whole field. So, I think the integration of drones is part of the agriculture business practice. It’s one of the great technologies that we continue to work on throughout IFAS with many different types of applications.
Right. And I think sometimes, especially when we use the word learning or, you know, I think sometimes people have this perception or fear a little bit of artificial intelligence. And what does it mean? And is it like, the big brother or going to take over the world kind of situation?
Can you talk a little bit more about what AI really is and how, how it is helping? We've covered that. But, but what about people's fears? How could we calm them?
So, I would say everyone is using AI. You are using AI since you got up this morning. So, if you're worried about AI that boat has gone, that ship has sailed, whatever. You know, we use AI to find shows on TV, to buy things on Amazon. Pretty much every part of your life -- health care, everything involves AI now.
So, I think, and one of the things to understand about AI is it's just a tool. You know, there's been a lot of different things that were developed in agriculture. The tractor really changed agriculture, right? All these things, I'm sure, came with some hesitation of the unknown. But, but I would say that, especially in the state of Florida, the researchers, the Extension offices, we work really hard to help. And you heard the wonderful explanation Nathan gave of deep learning. So, you know, we all work really hard to make sure what we do is quality and that what we are providing to our producers is, is good products with AI.
And one of the, one of the underlying themes of many of the AI programs, if not all the program AI programs on campus is the ethics involved in AI. So, really looking at, you know, not just using AI to do something cool, but thinking about, is this the right tool for the job? There may be another tool that would work better than AI, but I would say if you are, if you, if you have some hesitation about using something that has an AI component, I would say talk to your specialist or talk to your extension agent and ask them questions and help them to, to fill that gap of knowledge that you're missing to really feel like you're comfortable using that technology.
I think that's one of the great things about University of Florida and IFAS and our extension offices: We are here to help and to, to fill that gap. But I would also say, rest assured that we are not just -- it's not, you know, matrix. It's, it's things that, that really do make sense the way we're doing it and we are using, you know, state of the art knowledge and technology to make sure that we are making the best decisions using them correctly.
Right. And it sounds like AI, when used correctly as we are, is used to help make lives easier and more efficient. Like even as silly as the Netflix or, you know, streaming, for example. Like, it helps me find a show that they know I'm going to like based on what I've already watched. Again, it's just using data to make more informed decisions or choices. It's -- it doesn't have to be scary.
It's exactly that. And the other thing I'll add is, like I said when we first started, we have so much data available to us as a human, it's really hard to process all of that data in your brain. It would take so much time, but by using AI or deep learning or these other methods, you can process that information much more quickly. And so, I think that's a real powerhouse, that it brings that processing ability.
The other thing it brings is it brings a little bit of expertise. So, if Nathan trained one of his new technologies to identify a weed and take it out of field and he hired a new person, it may take that new person longer to figure out which weed is what, right? But if they had a technology that helped alert them, “Hey, we think this is this,”, you know, that's really helpful. So, it's also not maybe the final decision maker, but it’s an assistant to help you come to the right decision.
Right. And I think I've heard today or previously that, you know, really great AI is the use of a combination of person and technology working together to move forward, whatever that might be. Right?
Yeah. So, Nathan, I would love to hear your perspective. You know, being boots on the ground with farmers sometimes. What do you hear from farmers about AI or new technology that your team is working on?
You get the whole gamut. Some, some farmers are, you know, they hesitate. But other farmers are extremely excited about it and just the potential of what can be achieved.
You asked earlier about, you know, what people that are afraid of AI, because, you know, they have in their mind Matrix or Terminator movie stuff, things like that. But I think it's -- one of the things I always tell people is the AI programs we're using in agriculture, they only do what we train them to do.
So like, you know, an insect detection algorithm -- Dr. Lee in Gainesville is working on one that does that. It only can detect insects. It can't, it's, it's not truly thinking on its own. It can't make decisions on its own. It's exactly what Kati said. It can find an insect that basically is, you know, highlighted as this is possibly what you're looking for, okay.
But here's what -- and I said all that to say this is what growers get so excited about, is and it's in, Kati already mentioned this is -- it takes so much time for a person to walk a whole field to find mites, are the perfect example, to spot a spider mite on strawberries. You can't even see them with the naked eye, or not, or barely.
So, you can imagine trying. You have to take leaf samples. You have to send them to allow them to count. Decide whether you should treat for mites or look for the symptoms in the leaves. Do you have a technology that's not going to get tired, that can go through the field and find those for you, and you're not having somebody having to walk for hours.
Just think of the savings. And -- even so, I'm a weed scientist. I'm supposed to know my weeds, but I have on my phone these wheat ID apps because it makes it so much easier. I see a plant I'm not familiar with. I just hold my phone over it and it says, “Oh, I think it's this.” And, and I can then narrow it down to the exact species. And, and if I do that, just think how much that helps a grower who may not know, you know, weeds, species or insects or diseases. It's such an, an awesome tool. And especially if you think of these apps that we use in the phone for pest detection or nutrient deficiencies and all those things, it's not just a tool that big growers can use. Anybody can get them. A lot of these a lot of these tools are for free or very, very minimal cost. So that means even if you're only farming two acres and you know, you may, you know, might only be part time, you could still get all these tools that right now that can help you farm more efficiently.
Right. So, that makes me think about too, I mean, some of these tools are available even to, you know, a hobby gardener. But what other benefits, even if we're talking about to maybe the larger farms, how do how does advancements in farming technology trickle down then to the consumer?
You know, we've lost farms in recent years.
Due to competition from foreign markets. So, if you think about the difference between the farms in the U.S. and the farms in the foreign markets, the difference is what they have to pay for labor. So, if we develop a technology, well, the foreign markets are probably going to buy that technology, too. But they're going to pay the same amount. So, all of a sudden, you've removed the differential. And all of us here at IFAS believe in the ability of our farmers in Florida. And we believe that given a fair footing, that they can compete internationally.
So this technology, there's number one, it's just preserving agriculture in the U.S.. That's one thing it does. Second of all, you're you know, you're producing food with less -- could be less fertilizer or less pesticides, which benefits the environment, which again, benefits everybody. And then you're also -- there isn't a lot of data on this, but we, we just finished an experiment, got little bit of preliminary data that would suggest that you also end up with less pesticide residues on food, which is, you know, repeatedly a concern of consumers. So, if there's technology that you can use in some way to focus those pesticides where they're needed but not use them where they don't, that's going to benefit people in general.
Right. And then we've talked about some of the ecosystem services and water. Can you share an AI technology or something, you know, an advancement in farming that impacts water quality and quantity in the state of Florida?
The real strength of AI is the ability to take really complicated data and find patterns that are difficult for us to find with conventional statistical methods. So, if you're looking for weather patterns, rainfall patterns, nutrient movement patterns, and how that might affect algal blooms or how that might affect the movement of nutrients within an entire ecosystem, that's really complicated data and it's really hard for us to kind of find those patterns. That's the real strength of AI. You can really use that to look at what's happening in an entire ecosystem level.
Kati, I know you work again a lot with water. We mentioned that earlier. So, share an example or two that you've seen in advancements that impact water quality and quantity for the state.
Sure. So, there's been a lot of change in how we manage water. And for a couple of reasons. One has to do with the, the, the technology. As with all things, things got cheaper so we can measure plant water stress, we can measure soil moisture in the soil, we can measure weather conditions and estimate evapotranspiration. And so, all these things can feed into what kind of water we need to apply that will be used by the plant and not be lost in the system.
And so AI as, as Nathan said, allows us to look at these huge amounts of data and to really think about that in a forecasted way. And so not just using information from what happened the last five days, but really looking at the forecast. We can forecast ET. We can look at potential rain patterns and then how much of that rainfall will be stored in soil and how much of that rainfall might be lost. And then estimate where, where do we need to irrigate to get the most use out of that water so that that water is -- that is a very valuable resource. And we want to make sure, one, that water is available to the plant that needs it. But two, it doesn't take other valuable things with it and leave.
Nutrients and pesticides, right? The other thing that's really important when we start talking about water and water use is climate change and how the change in climate is going to change the amount of water we need for these crops. You have changes in temperature. You have changes in the rainfall pattern. So, it's not just how much rainfall you get in a month, but it's how that rain comes and how much comes at a time. And then how much of that can be stored in that soil. All those things get fairly complicated when you're trying to manage a fairly large property and you're either pumping water, which costs money and surface water that you're using. And then with each of those you have intrinsic water quality issues you may need to be dealing with. And so, it becomes a very complex problem to optimize the water for that system.
So, I think AI helps with trying to find the best solution, right? There may be multiple solutions depending on what your goal is. Is your goal to make sure there's absolutely no plant stress? Is your goal to maximize rainfall and as part of the irrigation of the plant? And then you can make those decisions and then use AI to figure out how do I need to water our fields? And, you know, pretty much most of Florida is irrigated agriculture and certainly controlled environment irrigated agriculture.
So, thinking about protecting water as a resource, making sure it's available not only for ag, but for us. I mean, we need it as well. We need it to drink, we need it to bathe, we need it for washing our clothes and our dishes, but also our natural systems which are very dependent on water and the quality of that water in the state of Florida.
Right. And I think if you're from Florida, even if you're not, you know that water is a hot topic and something of not just concern, but, you know, it's important for us to protect quality and quantity, especially with a growing population, as we've mentioned already. So, we're always going to need more water.
So, another agricultural technology that we've talked about a bit in other podcast episodes is plant breeding. So, can you talk about how that is advancing and how that impacts consumers that shop at the grocery store, Kati?
So, there is a connection between breeding, phenotyping, and the food that comes from that process. And AI has a role to speed up that process and some of the things that we may look for -- when we do that for different crops – might be to reduce its, the impact of different pest and disease on the crops so that you don't have to use as much chemicals and so you grow something that has less of those residues on the crop -- the fruit or the vegetable when it gets to the store.
The other thing is you can also grow our -- you can grow crops that require less water. So, they're, they're more hardy in terms of the type, the amount of water they need to grow. And so, they can withstand different situations better. If we have some water stress situations, we could have situations where we have some salinity in the water and plants that can perform well and in salinity situations may also be an option.
There's also breeding, and I think this is one of the most interesting things to me is we have technology, and we have traditional plants, but they need to meet in the middle. So, if we're talking about creating robots that can harvest, we need to breed in a way that they can harvest. So, there's a lot of things when you think about breeding and technology that go hand in hand that will actually help with either reducing costs, reducing risk, food safety, nutrition. When that, when that fruit, that vegetable lands in the grocery store and you buy it and take it home.
And Nathan, can you give some examples, specific examples of how this is used or maybe what's happening down where you're located?
Sure. So, a perfect example, I think, is the strawberry breeding program. Anybody who's gone and bought strawberries or they're in season and you pick up that package and you get that smell of fresh, you know, strawberries and that taste.
That taste of one that we like that. So, what a lot of people don't realize is there's been a lot of effort in the last few years at the University of Florida to improve the flavor of strawberries, but also disease resistance. So that's just with classical breeding programs.
But one of the ways AI can really help is in the classical breeding program, you're looking at thousands and thousands and thousands of plants and selecting the one that's the most disease resistant or the one that has fruit earlier in the season or the one that has, you know, the uneven production of fruit over the season. So, AI can help pick out the best varieties.
And, you know, a lot of consumers are very concerned about pesticide residues on food. But the best way to reduce the use of pesticides is to breed for crops that are resistant to those pests.
And if you can use AI to do that more effectively, that means consumers are going to get products that they want and love, but may be concerned with pesticide residues, but will have less pesticide residues on it. And it's also just being able to breed ones that give a consistent supply. So, that have you ever gone to the grocery store and you want to buy strawberries to make a strawberry shortcake and there's nothing left. AI can help with that. Making sure that consistent supply because you can better predict based on the breeding programs we've developed. A plant is going to have berries throughout the season rather than peaking a couple times.
So, there's so many ways that breeding, that AI can just as an effective tool to work together with classical breeding to improve what you see in the grocery store.
All these benefits, they trickle down to the consumer and they start with the farmer.
So, we've talked a lot about some specific technology. But I think as we start to conclude, I want to take a step back and talk about maybe what's next for AI at the University of Florida or in agriculture in general. So, Nathan, maybe talk about dreaming big, maybe things down the future that you would love to see or you do see coming down the pipeline in AI?
So, I think as far as the University of Florida goes, I think what we're already seeing but it will continue is that we'll see the, the AI become part of every curriculum. And that's kind of the goal. But it also will see it take a more prominent role in overall education as a very valuable tool.
In agriculture, I, I'm one of those people, I really believe that we are in the process, process of a major agricultural revolution, that this is going to change completely how we grow all the way from autonomous vehicles that are able to fertilize and buy pesticides or cultivation, whatever the need may be, depending on the crop to very, taking very complex data that we've talked about a lot and bringing this all to a platform that gives the grower, you know, growers can get overwhelmed with huge amounts of data, but the ability to process that data and give the grower an actionable output is the term we like to mean, which just means tell them what needs to be done.
So, make a decision.
Make a decision and give you the information that you need to make that decision and give you the information based on a lot more variables than we could do in the past.
So, just to give you a real quick example, I might be able to recommend when you should apply a management technique to control weeds based on temperature. I can even do it based on temperature and rainfall. I started to get more complicated. But then if I try to bring in more variables than that, it's really complicated. But it's not -- It becomes possible with the use of AI.
So, I think we're going to see more and more where farms -- a grower is going to be able to keep track of what's happening in his fields, what decisions need to be made often from single platforms. And you're going to see the integration of these systems. So rather than machine that just controls weeds and one that can find insects and one that makes sure optimizes irrigation, we're going to start to see more and more integration of these technologies functioning together as a unit. And that is really, in my view, anyway, that's the ultimate. That's where we really want to go, where you have a systems that are integrated and functioning as a single unit to help the growers.
And especially to get adoption from growers, right? Because the more simple and straightforward it is, I'm sure the higher the adoption rate will be because they're more comfortable with it.
Absolutely. So, we've, we have made and this will continue and I see this amongst all kinds of people who are working on AI, a real emphasis on taking platforms that are really easy to use. So, our precision spray technology, the underlying technology may be complicated, but the entire thing can be run from my smartphone.
I can turn it on with a platform that I'm very accustomed to seeing with very simple. I don't need any training to use it. And that's kind of the real goal of all this technology, is just to make it really user friendly. Use things that people are used to seeing and accustomed to operate.
Right, that makes sense. Because it's only as good as it being put to use, right?
Yes, and adoption of new technologies, and I'm sure Kati would attest to this, it's always been a challenge with any new major change -- adoption of that. You know, it's always a slow kind of process as people get comfortable with it.
So, there will be some of that similar process. And that's where IFAS is really critical. And our extension programs are really critical to help people understand how these benefit them, how they can be used, what they do, you know, what they don't do sometimes it's just as important as what they do. And the amazing extension programs that we have at UF, they're going to play a really critical role in leading this process of getting that information out.
Okay, So we've talked about the future of ag technology and AI and what might be coming down the road. But what about Hyper Gator? Kati, can you tell us a bit about that?
Sure. So, Hyper Gator is an amazing resource we have at the University of Florida. It's the eighth most powerful supercomputer at a educational institution. So, we are super lucky to have HiPerGator here. It was created by a donation from NVIDIA as well as infrastructure here at the University of Florida. And what it does is it allows us to process an enormous amount of data in a short amount of time. And so, all of these different technologies require many data point images, samples, rinds, whatever it is you're doing to collect the data and to, to process them, you need something that can process them in an amount of time that makes them useful. If we use our desktop computer, the amount of time it's going to take is not going to make that answer useful. And so, having Hyper Gator really allows us to develop these technologies much faster. And that goes for across the spectrum of disciplines. It goes from medicine to law to agriculture, of course, natural resources, to the arts and the sciences it’s just – i’'s such a powerful tool.
I know that my, some of my friends, they are at other institutions. They have to farm out to other services and have this sort of computational speed. So, it truly is a benefit for all of us at the University of Florida and in fact, the whole state of Florida. So powerful tool.
And that's awesome. That's awesome. That's so exciting. We're definitely lucky to have that. And I think, like you said, it just helps cut down, quote unquote, lag time. You know, I think we all know we wish sometimes research did happen quicker, but with tools like that, it helps us move things along faster and create results. That's exciting.
So, before we go, we've talked about a lot today, but to wrap up, what is one thing that you wish people knew about technology and agriculture? And if they basically forget everything else that you said today, what do you hope sticks with them? So, Kati, I'll have you go first.
So, the one thing I would want people to know is that the technology that's being developed for agriculture truly has agriculture in mind with the idea of making it easier to make really good decisions for growers. And so that they can do their job better with, with better knowledge and they, and they have the power -- they still have the power -- to make those decisions. Those decisions are taken away by the technology.
Okay, Nathan, what about you?
I think, and this has already been mentioned, but I think the key thing is, is all of these technologies are just tools. But they’re tools to make, as Kati said, to help you make decisions with better information or easier. There are also tools to help reduce the amount of labor in some instances and to make things more automatic. There are tools to make farming more efficient, more profitable and better for both farmers, but also the citizens of Florida and the U.S.
Well, thank you both so much for sharing your knowledge today and breaking down a really technical topic that can be really hard for people to understand and just wrap their minds around. So, thank you again for coming on the show today and doing that.
Thank you, Tory.
Thank you, Tory.
So that's it for today's episode of The Food is Our Middle Name podcast. To learn more about the latest innovations in ag technology, check out the show notes and follow us on UF/IFAS Blogs.