Host: Dr. Ping Yu, UGA Department of Horticulture Guests: Dr. James “Jim” Robbins, Extension Specialist and Professor (retired), University of Arkansas Division of Agriculture; Dr. Joe Maja, Director, Center for Applied AI for Sustainable Agriculture, South Carolina State University (joining from Japan) Producer: Rich Braman Episode: S02E04 — Blooms and Beyond, Season 2 (special two-parter, Part 2) Duration: 29:18 Air Date: June 28th, 2026
Welcome Back to the Drone Conversation
Ping Yu: Hello everyone, and welcome to the Blooms and Beyond Podcast, a podcast that uncovers plant history, culture, and management through the lens of science. I’m your host, Ping. How’s everyone doing today? I’m doing great because, uh, I’m going to pick up the conversation where we left off last week with, of course, two of my favorite people, Dr. James Robbins and Dr. Joe Maja, on the drones.
Last week, in the first part of our conversation, we got into some basics about drones in the nursery and greenhouse, types, licensing, and many others. This week in part two, we’re going into more drone applied research, including using RFID tags for inventory and, of course, more. So without further ado, let’s just jump right into it. Here is a conversation with James and Joe in part two. I hope you enjoy it.
Inside the RFID Inventory Project
Ping Yu: Let’s circle back to one of the projects that you guys are doing right now. It’s the RFID project, and I know for that, there’s a lot of, like, labeling. Um, that kind of– the labeling plays a significant role in terms of the, like, RFID. Can you tell us a little bit about the RFID project and what role does the labeling play in terms of, like, for a commercial nursery and in terms of the drone application in general?
Jim Robbins: So let’s go back to the original, again, our evolution, because Joe and I have been involved in trying to find an automated way to improve the inventory process in nurseries for many years, and we tried ground-based. When we first kind of thought about– for many years, we’ve thought about the RFID tags. So just think about it: it has to have line of sight. And so containers are jammed together, that’s just… Until you, you know, put them on a conveyor belt or something where you can get them all… The brilliance– the advantage is that it doesn’t require line of, line of sight. And if we could attach the plant, whether that… And then just fly over the top and simply count these things, we thought this was brilliant.
And then we also envisioned a more complete system. So let’s say you’re a wholesale producer and you put– at the time of canning, you put a tag on the container, the beds, the container beds. And again, you’re keeping in… A simple antenna at the back of the truck. And so as it goes onto the truck, probably on a conveyor, it would be read, and a very accurate inventory would occur. Maybe as a double check for, you know, manual methods, but touchless– so touchless inventory. That truck then goes to a garden center, comes off and, and recode that same tag. So they’re gonna then… Some information. That’d be their inventory number or their price, et cetera. And at checkout, they would use, again, a simple, either handheld scanner like we currently use for barcode, or they would use an antenna in a portal. So as they walk out, it’s again– like some of these stores, Amazon or whoever has these stores where there’s no workers there. You just walk out. It’s the same premise. So the tag, in theory, could be used on the production side, but also at the retailer side.
We also, I– oh, Joe? Oh, well, the other thing was that we don’t wanna just exclusively say you have to use a drone, because let’s say you’re a small nursery or you just wanna get a small block or something. You can just use a handheld reader. Uh, so it could be an either-or type, you know, situation. And we also extended it to a, uh, entry just into, uh, getting to work with this. One of the nurseries that we work at– is with the corner of a block. And so as they drive in a vehicle down the road, they would just shoot a handheld gun out the truck window. And on their laptop, which is already there in their cab, they would look at the crop history so that this tag be directly linked to their inventory system. And they would know, “Hey, we fertilized that crop three months ago.” I could see that on the screen. But I called up– I made a link to our production database using an RFID tag. So that was kind of a, a lower, uh, step into this process. Instead of labeling every plant, we would just label, like, a block and then database. So even that’s something.
And then I wanna give credit. A big– Joe is just exceptional in that, again, so he’s always working on what he calls dashboards. And so the, the, you know, it’s useless if we just collect this information, but we need to generate that, put it in some kind of a very graphically usable form.
Ping Yu: So, you know, he’s great at developing these– so, James has described the full version of the RFID system, but, uh, from the technical side, what kind of RFID tags are you using? Uh, what are the read range that you are achieving with the drone-mounted reader versus a handheld? Uh, what does the data pipeline from those readers to the dashboard look like?
Joe Maja: Yeah. So from a, uh, technical perspective, we’re primarily using passive, not active– passive UHF RFID tags. So these are cost-effective and don’t require a power source, which makes them practical for large-scale deployment. So the read range depends on several factors, including the reader configuration, tag orientation, and environmental conditions. In fact, we have a publication for this. With a drone-mounted system, that directly influences flight planning, altitude, speed, and coverage pattern. So handheld readers provide more controlled scanning, but require manual operation. Now, the, uh, data pipeline is also important. So once the tag is read, that information is transmitted to a backend system where it’s processed and visualized through a dashboard. Uh, that’s what turns raw data into something actionable.
RFID vs. Traditional Tags — and Lessons from the Field
Why Switch from Traditional Tags?
Ping Yu: One of the things that I wanna ask you guys is, like, what’s the difference of, uh, like, a traditional normal tag or, uh, normal label versus an RFID label, and why growers need to use an RFID label instead of the traditional label? Well, you kind of touched base a little bit on there. But– and also, if people want to decide using the RFID label, I know there are companies that they can work– uh, they can work with. Do you have any suggestions on how they can, uh, properly use that in their nursery to help with managing their, uh, inventory? What, what would be… Yeah. Thank you, Joe, for bringing that up because that’s one of the questions I wanna ask you. And if people want to do it, why they should do it, and how much of the price difference between the RFID tag versus the normal tag, because at the end of the day, they’re gonna make a decision: “Okay, is this worth doing it? How much money that I’m putting in there?” and all that sort of things.
Joe Maja: Yeah. Let me start with the simplest way to think about it. So traditional plant tags are essentially visual. They rely on line of sight. You or your staff have to physically see the tag, read it, or scan it one at a time. So that works, but it’s labor-intensive and doesn’t scale well, especially as operations get larger.
Now, RFID changes that completely. So with RFID, you don’t need line of sight. You can read tags wirelessly, in bulk, and almost instantly. That means instead of scanning one plant at a time, you can scan hundreds or even thousands of plants in a matter of seconds. And that shift from manual one-by-one interaction to automated bulk data capture is really the core value of RFID.
Now, why should a grower consider switching? So it really comes down to efficiency, accuracy, and visibility. With RFID, you’re not just tracking inventory, you’re creating a live, dynamic system. You know where your plants are, how many you have, and how they’re moving through your operation in near real time. That’s something traditional tags simply cannot provide. So another important aspect is traceability. So as the industry moves more toward data-driven decision-making and supply chain transparency, RFID gives you a foundation for that, so you can track plants from propagation all the way to retail, and even integrate that information into a point-of-sale system. So it’s not just about inventory, it’s about connecting the entire life cycle.
Now, in terms of getting started, I usually recommend that, rather than trying to implement everything at once, it’s very important to start small. Maybe choose a single block or focus on higher-value crops where the return on investment is easier to justify. And that allows you to test the system, understand how it fits into your workflow, and identify any challenges before scaling out. Now, on the vendor side, there are a number of RFID providers and integrators out there. The key is to work with someone who understands ag applications, because this isn’t the same as retail or logistics. So things like moisture, uh, plant density, and container materials can affect performance. So you want a solution that’s been adapted for that environment.
Now, uh, let’s talk about cost, because that’s always the big question, right? RFID tags are more expensive than traditional tags. There’s no way around that. But the way I encourage growers to think about it is not just the upfront cost, but the overall system value. So when you factor in labor savings, uh, reduced errors, uh, improved inventory accuracy, and better decision-making, the return on investment can actually be quite compelling. And the other thing to keep in mind is that costs are coming down. So as the technology matures and adoption increases, we’re seeing more affordable options and better performance. So what might have been cost-prohibitive five or ten years ago is becoming much more accessible today. So at the end of the day, RFID isn’t just a replacement for a traditional tag, it’s a shift toward a more automated, data-driven way of managing nursery operations.
What Works, and What Can Be Improved
Ping Yu: So over the years, ‘cause, you know, I know you guys have been working on the drone application for, well, 20, almost 20 years or over 20 years. Have you… What have you learned over the years in terms of what works with a commercial nursery on drone application and what doesn’t work? I wouldn’t say what doesn’t work– what can be improved? Uh, what, what… Now, over the years, with your experience working with drone application, what have you learned? Like, what works and what doesn’t?
Joe Maja: Yeah. So one of the biggest lessons I think we’ve learned over the years is that, uh, simplicity almost always wins, right? Simple is better. You can design a very sophisticated system in a lab or controlled environment, but if they’re too complex to operate or maintain in the field, they’re not going to be adopted. So the systems that work best are the ones that integrate smoothly into existing workflows. Growers already have established processes, and if a new technology requires them to completely change how they operate, it creates a problem. So we’ve learned to design systems that complement what they’re already doing rather than replacing it entirely.
Another important lesson is that agriculture is a very harsh environment for technology. You’re dealing with dust, heat, humidity, wind, and sometimes unpredictable conditions, weather. What works perfectly indoors or in a test setting can fail quickly in the field if it’s not designed with those conditions in mind. So we’ve also seen that reliability is often more important than precision. A system that works consistently, even if it’s not perfect, is more valuable than one that’s highly precise but unreliable. So growers need tools they can depend on day in and day out.
So on the flip side, some of the things that don’t work as well are over-engineered solutions. There’s a tendency, especially from an engineering perspective, to try to solve every possible problem at once. But that can lead to systems that are difficult to use, difficult to maintain, and ultimately not practical. So over time, we’ve shifted toward more modular, scalable approaches– start with a core functionality that delivers value, and then build on that as needed. And I think the final lesson is the importance of collaboration. So the most successful projects we’ve worked on have been the ones where engineers, horticulturists, and growers are all involved in the process. So that feedback loop is critical for developing solutions that actually work in the real world.
How Grower Interest Has Shifted
Ping Yu: So Joe, I have seen, uh, increasing interest from growers in drones. Uh, over the past five years, has grower interest in drones grown, or shifted in what they’re asking about, from your experience?
Joe Maja: Yes. So we’re, uh, definitely seeing the trend, and it’s become much more noticeable over the past several years. So early on, when we would talk to growers about drones, most of the questions were very general. Things like, “What can this technology actually do?” Or, “Is this something that’s practical for my operation?” So it was really more curiosity than anything else.
Now, that conversation has changed quite a bit. The growers are coming in with much more specific questions. They’re asking about how to implement the technology, what type of system fits their operation, how to handle data, and what kind of return they can expect. So it’s moved from curiosity into a much more practical decision-making stage. Another big driver is labor availability. As labor becomes more limited and more expensive, growers are actively looking for ways to maintain productivity, and that’s where drones and related technologies are becoming much more attractive. So overall, I would say, Ping, the interest has clearly shifted from “What is this?” to “How do I use this?” And that’s usually a strong sign that the technology is moving toward mainstream adoption.
Where Growers Can Learn More
Ping Yu: And what are the resources that’s available for, um, a commercial grower who wants to get more into the drone application or adopt more advanced technology into their operation? Are there any– ‘cause I know you guys have put out a lot of, like, extension fact sheets and, um, magazine, trade, uh, magazine papers out there, but are there any additional resources that’s available for them to learn more about it?
Mm. Yeah. Um, I don’t… Well, I heard there might be some, like, a group of researchers who are doing drone, um, um, work. ‘Cause I think I came across a researcher from the University of Maine, and they have a group of people who are doing the drone work. And they have– I think they have some sort of drone training program that you have to sign up for. But besides that, I don’t have any other additional resources. And I think you have to pay for that too, but I could be wrong.
Joe Maja: Well, a good starting point is university extension programs. So many land-grant institutions like, you know, UGA, Clemson, South Carolina State, and others are actively working on drone applications and publishing fact sheets, guides, and case studies. So these are usually free and designed specifically for growers, so they’re very practical.
Beyond that, there are also training programs. Some are short workshops, others are more structured courses. Some of these are offered through universities and others through private organizations. So they can be helpful, especially for understanding regulations like, you know, Part 107, and getting hands-on experience.
Now, industry conferences– uh, and I’ve experienced this one– and trade shows are another valuable resource. They give growers a chance to see the technology in action, talk directly with vendors, and learn from other growers who are already using the system.
For those who prefer a more self-directed approach, I think there’s also a growing amount of content online. You can find video tutorials and forums, even on YouTube, where people share their experiences. So the key is to be selective and focus on credible sources. What I usually recommend is starting with extension materials to build the foundation and then moving into more hands-on training or vendor demonstrations once you have a clear idea of what you want to implement.
What’s Next for Drones in the Green Industry
Ping Yu: And from, like, what I would like to know with your experience– and, um, I know you guys are still looking for new ways to advance the technology or the drone work in nurseries– what are some of the next steps that you’re excited about in the drone work? Um, and what do you see this technology going forward being incorporated into the green industry? What is your vision for it?
Joe Maja: Yeah. So looking ahead, what really excites me is the integration of multiple technologies into a unified system. So we’re already seeing the pieces– drones, RFID, robotics, AI– but they’re often used independently. So in the future, I think those systems will work together seamlessly, similar to what we’re doing with Jim. They collaborated; technology collaborated as well. And for example, a drone could be used for monitoring crop health, an RFID system could track inventory in real time, and a ground robot could assist with transport or targeted application. All of that data would then feed into an AI-driven platform that helps growers make decisions. So instead of isolated tools, you have an interconnected system that supports the entire operation.
I also think we’ll see more automation in routine tasks. So things like inventory tracking, crop monitoring, and even certain types of application could become more automated, which is, um, especially important given the labor challenges the industry is facing. Another area of growth is in data analytics. So as we collect more data, the ability to interpret that data and turn it into actionable insights becomes critical. So that’s where AI and machine learning will play a major role.
But I think it’s important to keep expectations realistic. You know, not everything will be fully automated overnight, and adoption will happen gradually, and it will vary depending on the size and type of operation. Still, the direction is clear. Now we’re moving toward more connected, data-driven, and efficient systems in agriculture.
Ping Yu: Do you think that in the next fifteen years or twenty years– do you think that, uh, how many of those nurseries or commercial nurseries do you think they are gonna adopt, uh, this technology? Just give a, give a guess. Fifty percent?
Joe Maja: Yeah. That’s always a, uh, a challenging question, because adoption doesn’t happen all at once or in the same way across the industry. But if I had to give a realistic estimate, I would say somewhere in the range of about fifty to seventy percent over the next fifteen to twenty years.
That said, I think it’s important to break that down a little bit. Not all applications will be adopted at the same rate. So for example, using drones for basic imaging or monitoring is already becoming quite common. And that type of application will likely reach higher adoption levels sooner, because the barrier to entry is relatively low. Now, on the other hand, more advanced applications like automated spraying systems, fully integrated RFID-based inventory tracking, or AI-driven decision support may take longer to scale, and those require more investment, more infrastructure, and sometimes a shift in how operations are managed.
So another factor is the size and structure of the operation. Larger commercial nurseries are generally in a better position to adopt these technologies early, because they can justify the cost through efficiency gains and labor savings, right? The smaller operations may adopt more gradually or focus on specific use cases where the benefit is clear. So when I say fifty to seventy percent, I’m really thinking about a blended adoption across different technologies and different types of operations. Overall, I do think adoption will be significant, but it will happen in stages, not all at once.
Ping Yu: Okay. Yeah. I, I think it’s going to be, it, it has to be a game changer for– to deal with the, uh, issue that we’re facing right now with the labor and the safety issues and all, all in all. But, um, yeah, I think that’s one of the beauties that I see from the work that you guys have done, because, uh, safety is one of the top issues that a lot of people, uh, at least in agriculture, are facing right now.
Final Thoughts and Goodbyes
Ping Yu: The last question that I wanna ask you guys: are there any other last comments that you wanna leave for the audience? Are there any suggestions you wanna give for people in horticulture in general, with your experience in horticulture for how many years?
Joe Maja: Yeah. So I think one of the biggest takeaways from everything we’ve talked about is really the importance of collaboration. So this kind of work sits at the, um, intersection of multiple disciplines. Engineering, agriculture, data science– and no single area can really solve these challenges on its own. So the most successful efforts we’ve been part of are the ones where those perspectives come together in a meaningful way.
So for growers, um, I would say the key is to start with a clear objective. There’s a lot of technology out there, and it can be overwhelming oftentimes. But instead of trying to adopt everything at once, it’s much more effective to identify a specific problem– whether it’s inventory tracking, crop monitoring, or improving application efficiency– and then look at how these tools can help address that.
It’s also important to approach this with a mindset of gradual adoption. You don’t have to transform your entire operation overnight. So start small. Test what works, learn from it, and then build from there. So that approach tends to lead to more sustainable and successful outcomes.
And, uh, for students and young professionals, I would say this is really an exciting space to be in right now. Ag is evolving rapidly, and, uh, there’s a growing need for people who can bridge technology and applied systems– you know, whether your background is in engineering, uh, plant science or even data analytics, there are opportunities to contribute in meaningful ways. And for the industry as a whole, I think we’re at a point where these technologies are moving from being experimental or research-focused into something that’s becoming more practical and accessible– they’re not just concepts anymore, they’re tools that can help address real challenges, especially around labor efficiency and sustainability.
Yeah. Now, so I think overall, I’m very optimistic. Uh, there’s a lot of work to be done, but the direction is clear, and I think we’re going to see continued progress as these technologies become more integrated into everyday operations.
Ping Yu: So Joe and James, with that in mind, if people want to find more about your work, where do you recommend they, uh, go look for more? I can put, I can put the resource or link, whatever, in there if you have any, like, source people can, can find more of your work. Yeah, ’cause I know Joe has a Google Scholar page, and then he also has a, like, a university profile page. Uh, I can, I can just easily put a link in there and put it in my show notes.
Joe Maja: The easiest way to find my work is to search my name along, you know, with South Carolina State University or Google Scholar, which will take you to my university profile page, where you can see an overview of what we’re working on. You can also find, yeah, many of my research publications, again, on Google Scholar, which is probably the most comprehensive place to follow our work over time. Um, we’re also in the process of developing a center website that will highlight our projects and outreach activities. So that should be available soon as well.
Ping Yu: Uh, thank you. Thank you so much, Joe. [laughs] Joe, Joe is– Joe is recording from Japan. Uh, and thank you for joining me today, and, uh, it’s, it’s great to see you guys, and, uh, like I said, I really appreciate it, and you guys set a great example of how collaboration works and how collaboration matters. So thank you guys so much for joining me today.
Jim Robbins: Thank you.
Closing and Credits
Ping Yu: That wraps up today’s episode. How exciting. All those wonderful information– that’s, it’s just incredible. So this is actually a very– a great example of the, uh, future-shaping technology that is happening right now, uh, right in, in horticulture. So I thank Dr. James Robbins and, uh, Joe Maja for taking time out of their busy schedule, and Dr. Joe Maja is actually joining us from Japan. It’s very late today, but I thank them for, uh, taking their time to talk with us.
And, uh, always go check the show notes to learn more about this topic and other topics that we feature on this show. Uh, make sure you hit that subscription button. Consider giving a review for the podcast and donating to the podcast to support the show. Again, thank you for listening. Till the next time, stay healthy and go plants. Adios.
Transcript produced for Blooms and Beyond / UGA Center for Urban Agriculture