In this episode of "Let's Talk Farm to Fork," we're joined by James Paterson from Aerobotics, who we'll be talking to about how their aerial imagery and machine learning are helping suppliers increase their yields.

Transcript

[00:00:00] Mitchell Denton: Hi there, and welcome to "Let's Talk Farm to Fork". The PostHarvest podcast that interviews people of interest across the food supply chain. Today on our show, I'm joined by James Paterson from Aerobotics, who I'll be talking to about how their aerial imagery and machine learning are helping suppliers increase their yields.

So, with no further delays, let's get started.

Well, hello James. Thanks for joining us on the podcast today.

[00:00:26] James Paterson: Hi, Mitch. Very good to be here.

[00:00:28] Mitchell Denton: Before we get into it though, would you mind telling us a little bit about yourself and what you do and maybe a fun fact about yourself?
[00:00:35] James Paterson: Yeah, so I am, James Patterson. I'm the CEO of Aerobotics, and one of the two co-founders. So, myself and, uh, Benji Meltzer actually started Aerobotics eight years ago. I am an aeronautical engineer by training. I got my masters at MIT in Boston, where I worked a lot on drone systems and especially analysing the data collected from drone systems, um, at the time in the context of search and rescue.
So that's where the, the drones and and AI side of things comes from. But, um, yeah, a fact about myself, I'm actually, um, come from a fruit farming background. So, my dad is a fruit farmer. We grow, uh, citrus, avocados and stone fruit in South Africa, and that's where the love of agriculture comes from.
So, um, once I completed my masters, joined up with Benji and decided to bring this, uh, technology back to the farm.
and yeah, I started with my dad there and delivering the new technology and building technology for farming.
So he's still a customer of ours and we grew it out from there out to, to the neighbours.
Shifted our technology into Australia and then into Europe and the USA.
[00:01:54] Mitchell Denton: Wow, that sounds like it was a match made in heaven. So then continuing on from you telling us what you do, would you mind telling us a little more about the history of Aerobotics and how your innovative technology works?
[00:02:06] James Paterson: Yeah. So as I mentioned, we started the business eight years ago. Um, and if you go back eight years, Drone technology was still very new then. Um, the AgTech space was even newer than, than what it is now, and we had to build everything ourselves. So we started building our, our own drones, our own camera systems.
The software to fly the drones was, was written by us. And then of course, the most important was the extraction of information from the data that, that these drones were collecting. And wrapping that into a product where, where growers can make production decisions. So the way, the way the technology works, we're a computer vision company.
We've progressed to collect imagery from satellites, drones, and actually smartphones out in the field where customers can take photos of, of trees. And what, what we do for customers in the agriculture space is help them monitor and optimise their production.
So we'll start by measuring each tree, its performance, how it's growing over time, and the health and transpiration of that tree. So that's used to optimise that tree which we see as the factory that's producing the fruit. And then we take that a step further.
We use higher resolution imagery to actually, uh, identify and track the fruit through the season. Um, so tracking the fruit size development and predicting through to harvest, what that size structure is going to look like.
[00:03:37] Mitchell Denton: Yeah. That's fantastic. You mentioned your father before and his farm as a case example. I was just wondering, can you share any success stories or case studies from farmers who have used your platform to improve their crop management?
[00:03:51] James Paterson: Yeah, it's, um, we're making quite a, uh, a large difference to growers and especially, uh, vertical groups that don't just grow the crop, but also pack and market the fruit. So as an example of how customers would, would use this fruit information, early season, the production groups would use this for, um, setting the crop, thinning fruit, for example, using fruit count and fruit size, um, as well as setting, um, irrigation and fertiliser practices for the season.
Um, mid-season, the vertical groups, the sales teams actually need an early estimate of what fruit's going to be coming in and what the size structure looks. So they, they use that to optimise the logistics, supply chain and marketing programs.
And then, uh, a big one is at, at harvest time, analysing, you can imagine these big groups have hundreds of, of blocks. Um, some of our customers have 40, 50 farms in their operation and just knowing where the fruit is ready to harvest, what percentages at the right size that they wanna harvest, really helps them, you know, make their processes more efficient and increase the utilisation percentage of, of the fruit.
So, yeah, I'm, I'm very happy with the solution we're providing to the, these, um, large farming operations and the vertical groups where they can optimise processes right from production through to the sales process.
[00:05:17] Mitchell Denton: Sounds like you've really helped some farmers optimise their operations. I see that Aerobotics is not only concerned with helping farmers optimise their crop management, but also with providing a more efficient crop insurance solution. Would you mind explaining to listeners how Aerobotics precision crop insurance is more beneficial than traditional methods?
[00:05:37] James Paterson: Yes. So, our, um, if you think about what we do, we are measuring the trees and the fruit. And if you think about what that means for a, a grower, those are essentially the assets of that operation. A tree is a long-term asset. It's gonna be producing a crop for 20 plus years. Um, and the fruit is the current asset.
That's the revenue that they're going to be generating that season. So naturally that's a good fit in financial services, both on the crop insurance and on the financing side. And what we've decided to do there is to enter the crop insurance base first. Um, we started in the USA and we act as a licensed broker, helping growers set their crop insurance policies according to what the crop is and how the trees are performing.
So that is the, the, um, solution. It's running in, as I mentioned in the US at the moment, and we are investigating, additional markets. We've spoken to a few of our customers in, uh, Australia and in South Africa about a solution there as well.
[00:06:44] Mitchell Denton: That's really exciting. I'm sure the farmers out here in Australia would really appreciate those services. I'm curious, whether it be internal or external, what would be the biggest challenge your team have encountered so far with your technology?
[00:06:57] James Paterson: Yeah, I think the, the agricultural space is challenging. One of the challenges we faced, early in our, in, in building the business was, is what we call the long tail in agriculture. So if you take one crop family, for example, citrus, that has quite a lot of variety within it, you've got, uh, you know, even at a broad level, you've got navels, oranges and mandarins, and then you can go down within that to different varieties of Mandarin.
And another step further is the same variety of Mandarin will grow differently in different regions. Um, so when you're doing your prediction of, of fruit, it's challenging that, that the fruit grows differently in different regions and different farming practices. So I guess dealing with that, that long tail was, uh, an initial challenge for us.
And we did struggle with that in, in the early days. And what we did was to build the technology to actually localise to the region that it's operating in. So as one of our customers uses this, this technology through a season, it'll actually, um, the following season use that data of that farming operation of the varieties that they've got to improve the following year's prediction models.
That's worked quite well for us because it's, the previous problem of trying to solve all of this variation in agriculture, uh, was, was non-tractable. And using customer data to update their own models has, has been great for us.
[00:08:31] Mitchell Denton: Yeah, I'd imagine putting together prediction models early on would be quite a challenge. So I'm, I'm glad you've managed to iron those details out. So then, what are some of the most exciting developments happening in the AgTech industry right now, and how do you see Aerobotics fitting into that landscape?
[00:08:47] James Paterson: Firstly, I think, AgTech is an extremely exciting space. It's a very challenging space because, traditionally there's been quite low adoption of, of technology in agriculture. But I think it's a space that's got, um, one of the biggest potentials to make a difference to the world, um, both from food production and climate change.
So I think the space is extremely exciting. It's still extremely young. Um, so there's so much still to come here. I think. Um, some of the things I see as exciting at the moment, um, the robotics and automation. That's just a, a personal love for me as well coming, coming from the engineering side. But I do think that, uh, labor and efficiency is a, is a challenge in, in many regions we operate in, so I like that space.
Technology that's, that's, um, making inputs more efficient. So, you know, irrigation technology companies, uh, fertiliser tech. Um, new biologicals. I think, think those are very important. Trying to reduce things like pesticide use out in the field. And then I believe the big one for me is, is data and, and sensors that can, uh, collect this, this data out in the field.
And that's where I think both of our companies, uh, operate in. And yes, I, I believe that with, with better data, growers are are experts in farming, right? So with better data, they can just scale that knowledge across the operation, become a lot more efficient in producing the crop and then especially in the supply chain, better data there for, um, yeah, managing optimising processes, making those more efficient and reducing food waste.
[00:10:34] Mitchell Denton: I 100% agree. I feel like all those areas you mentioned are all really interesting and we've seen some great innovation happening, but at the end of the day, I think data collection is the real big one that's really pushing things forward, so that's really cool.
Continuing that thought, how do you see the role of drones and other aerial imaging technology evolving in agriculture in the years to come?
[00:10:57] James Paterson: Yeah, I think it's, uh, it's interesting because it's already evolved quite, uh, quite a lot. If you go back six years or so, which is just after we started, there was a huge hype globally about drones. And then they were gonna be able to do everything for you, that drones are going to be walking your dog.
Um, you know, so there, there was a little bit, um, I, I guess over hype in that industry in particular, and things have settled now to more realistic use cases of, of these drones out in agriculture. So currently, um, obviously they're very good at collecting data across large areas and getting that perspective from above at scale.
What I see happening now is also use of drones for applications. So I see, uh, the spraying space picking up quite nicely. The volumes that are able to put down, um, still need to increase, but certain applications like, uh, fruit fly baiting or things like that, um, they're quite a good fit there.
And then the beneficial insect release space. So use of beneficial insects in agriculture is picking up quite a bit as a replacement for, for pesticide use. Um, and releasing those efficiently across large areas, uh, by drone is a good, uh, is a good use case there.
I, I do want to say here that, that our company in particular is also, you know, analysed where drones perform well and and where they don't perform well. And as I mentioned, for us collecting imagery at large, large scale across the crop, uh, we use drones for, and then for actually, um, measuring and tracking the fruit, we've brought that technology onto, uh, smartphones and infield cameras, that our customers can actually use themselves whenever they want to collect fruit information.
So, yeah, we, the, the space is evolving and also the understanding of where the, where these, um, machines are strong and where they need to be augmented with other imagery types is, uh, is always evolving.
[00:13:05] Mitchell Denton: Yeah. Okay. That's exciting. So then, besides the obvious problems on the front end of the food supply chain that you are looking to be a solution to, what would you identify as being one of the biggest pain points in the food industry?
[00:13:19] James Paterson: Yeah, as I, as I mentioned earlier, I think access to good data for the entire supply chain. Is is a big pain point at the moment. It's very hard to estimate what the crop is out in the field. Um, when you do make that estimate, things can change. The, the weather can change and the crop might grow slightly differently.
And the whole supply chain relies on this information, the packhouse, the harvesting process, the packhouse, the logistics chain, right through to the retailer who's selling the fruit. So there's a lot of wastage happening there at the moment due to, um, insufficient data and that's, that's really a, a place that I think both of our, both of our companies are, are working into to help optimise and make that that chain more efficient.
[00:14:05] Mitchell Denton: Absolutely. Yeah. Traceability and transparency, they're the two big factors that we're definitely trying to kind of tackle. So is there a particular group or innovation within the industry that you're excitedly keeping a watchful eye on?
[00:14:19] James Paterson: Yeah, we, we obviously track the other AgTech groups, um, and what they're, what they're building and developing. If I look at the, the other company solutions and especially customer feedback, um, when, when I speak to customers, I think, uh, groups like Semios are doing a good job. Um, out in the field, they have a, a solution to reduce pesticide use and, and manage pest and disease outbreaks automatically for growers.
They're also extending their solution into things like irrigation. Um, I think, uh, Phytech is is doing a, a pretty good job. We see them quite a lot in California and Washington and, and also in Australia.
So I think that's, that's a great solution. And then we mentioned robotics earlier on, uh, Burro is, is, um, also expanding quite nicely, helping initially with, with harvest processes, but um, yeah, it's essentially a little, little robot that can, um, carry things out in the field and, and also potentially mount, uh, camera systems on it to collect data out in field.
[00:15:28] Mitchell Denton: Yeah, we've actually had Charlie from Burro, uh, on the podcast previously as a guest, and yeah, it's really cool what they're doing. It's actually a very simple solution, but very effective and they're, they're doing a great job, so it's, it's really cool.
[00:15:44] James Paterson: I think that's, that's one of the keys in, in agriculture technology is the solution needs to be simple and effective as, as you've described now, um, it doesn't need to be, um, In, you know, uh, how agriculture's going to be in 10, 20 years time. Um, we need to take, um, small, simple steps forward here in, in this industry, and I think they're a good example of starting with a simple solution and then, uh, they can progress from there.
[00:16:15] Mitchell Denton: Yeah, I, I totally agree. So then what advice would you give to entrepreneurs looking to start a company in the AgTech space?
[00:16:24] James Paterson: Yeah. Um, look, I, I always advise people to enter the startup space in general. I think it is, from an education and, and personal growth side of things. There's no better way to, to learn than entering the startup space. Much better than any, any PhD that, that you'll get out there. So I really encourage people thinking about making the leap to make that leap.
The risks aren't as high. You can always go back to a, a regular job if things don't work out. So yeah, definitely enter the, the startup space and build something if, if you are interested in that.
I would say in the AgTech space in particular, um, two things. The one is, I believe that you should only enter that space if you have some connection to agriculture or some background to agriculture.
It's quite a different space to, to the you know, traditional markets, FinTech, et cetera. So I believe you need some sort of connection or, or knowledge of the industry to, to increase success levels of starting there. And then the second thing is, is the problem you're solving needs to be extremely clear and simple and easy to use in the customer's operations.
That's very important. Um, it needs to easily tie into how farmers currently farm or how they currently run their processes.
[00:17:50] Mitchell Denton: That's actually really practical advice. I'm, I'm so used to most guests getting this question and telling people, just don't do it. Just don't bother getting involved, so.
[00:18:01] James Paterson: It's not easy, right? There's there a lot of highs and a lot of lows, but generally there are more highs than lows. And, um, if you can, can stay positive and if you enjoy it, then, then I think the chances of success are, are decent to high.
[00:18:17] Mitchell Denton: Absolutely. So James, we are coming to a close, but before we do, I just wanted to ask you, what is the main point you really want the listeners to take away from this episode?
[00:18:27] James Paterson: I think that the takeaway from, from what we've spoken so far is that. You know, technology currently is, is in quite a difficult space, AgTech in particular seen as a, a tough sector to crack and a tough sector to invest into, but um, for me it is really the space that can make the biggest difference for us at the moment.
So I think both agriculture itself, and then AgTech supporting agriculture can have a huge impact on food supply and climate change. So make, making these changes for the globe overall will require, um, investing into the space and it will require founders entering the space and building companies to solve some of these problems.
So I think that's the key takeaway here is it's, it's a challenging space. But I think one of the most important ones and, uh, an industry that needs support and needs people entering the, the space and building solutions.
[00:19:28] Mitchell Denton: Absolutely. I couldn't have said it better myself. Well, that's all for today's episode of "Let's Talk Farm to Fork". Thanks for listening and thank you James for joining me today.
[00:19:37] James Paterson: Thank you, Mitch. I really enjoyed the conversation and hope to speak to you soon.
[00:19:42] Mitchell Denton: If you'd like to know more about James and Aerobotics, check out the link in the description of this episode. Make sure to subscribe to the podcast so that you never miss an episode, and don't forget to leave a review and share with your friends.
Until next time, you've been listening to "Let's Talk Farm to Fork", a PostHarvest podcast.