In this episode of "Let's Talk Farm to Fork," we're joined by Brandon Contino from Four Growers, who we'll be talking to about how their autonomous harvesting robots are a labour shortage solution that is helping increase 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 Brandon Contino from Four Growers, who I'll be talking to about how their autonomous harvesting robots are a labour shortage solution that is helping increase yields.

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

Well, good morning, Brandon. How are you?

[00:00:25] Brandon Contino: Hi, doing well. How about yourself?

[00:00:27] Mitchell Denton: I'm doing pretty good thanks. Before we get into it though, I just wanna ask you if you could tell us a little bit about yourself and what you do, and maybe a fun fact about yourself.
[00:00:37] Brandon Contino: Yeah, sure thing. So my background's in electrical engineering, but spent a lot of my time in the software side. So computer vision, motion planning algorithms, uh, based here out of Pittsburgh, Pennsylvania in the US. And today I'm the co-founder and CEO of a company called Four Growers. And we're building, uh, automated harvesting and AI analytics, uh, robots for starting in the greenhouse industry.
I guess a fun fact about myself, let's see. I tend to change hobbies about every quarter, and I think right now the one I'm really into is a stable diffusion. So I've been playing around a lot with generating, uh, you know, the gen AI kind of custom images, really outlandish images, uh, that would take forever to, to create yourself.
[00:01:18] Mitchell Denton: Yeah. Wow, well, before we get bogged down on that too much let's talk a little bit more about Four Growers. So can you tell us a little bit about the history of Four Growers and how your autonomous harvesting technology works?
[00:01:31] Brandon Contino: Yeah, definitely. We, our history's a little bit, uh, we'll say windy. We didn't always set out to build a robot from the start. Uh, me and my co-founder were really interested in how can we help improve the food system. And we really were exploring how can you grow food more sustainably? And one of my big focuses was really around water scarcity.
So, how can we do it with a lot more water efficiency? And so my co-founder had actually grown hydroponic, uh, lettuce for local food banks here. And we were actually looking at building a vertical farm, and we were gonna become growers here in Pittsburgh. Uh, except we couldn't really build full conviction once we started looking at the business model, at least in the near term, that you could really build something at scale that would have commodity-level pricing.
And so we kind of took a step back, to be honest, six years ago, didn't even know greenhouses existed. And, uh, realised how amazing they are, they're over 90% more water efficient, they're 30 times more land efficient. You can, with their climate and variant, you can grow wherever. Basically, all the benefits of vertical farming exist in greenhouses without that really big energy intensivity and kind of the part that makes it really challenging to be successful.
And uh, from there, we just started picking up the phone and calling, uh different greenhouse owners, cold calling, trying to get access to them. Half the time we'd get hung up on by the receptionist, half the time we'd get through and we would just always ask them, you know, what is the biggest pain point? Why aren't there more greenhouses? How can we help you scale?
And everybody always said labour. And, uh, so that's, that was the very beginning. Uh, then my background in computer vision, motion planning, and my co-founder, mechanical engineering designing custom grippers, we realised we had the perfect skillset to attempt to solve this problem. And we basically went and just moved into the greenhouse.
Uh, and so the very beginning days, we were literally just living in the greenhouse pretty much every day working. You know, we'd be there at 1:00 AM we'd be there at 4:00 PM when it's over 120 Fahrenheit in the greenhouse. That was a really great, uh, kind of beginning experience. And now, now we're much bigger, but it was a kind of fun start.
Our autonomous harvesting technology, we're a little unique in how we go about doing is we've developed this, uh, what we call like a soft pick, uh, vacuum-based technology. And so we use a robotic arm. That with our custom AI perception models and the whole data sets we built to train it and our own motion planning, is able to reach out into a plant, uh, reach around other stems and other obstacles, uh, pick the tomatoes off the plant using vacuum and then also deposit it into this cart that can hold 24 different crates.
And so we can do entire rows autonomously without any need for any human intervention, and we're able to now on our best harvest, even do it faster than what a person might be able to do.
[00:04:07] Mitchell Denton: What would you say separates the GR-100 from other competing technology and methods of picking produce, beyond what you were listing before?
[00:04:17] Brandon Contino: Yeah, I think that the biggest thing by far is definitely the speed. Uh, I'm not sure if you've seen a lot of, uh, robot harvesting videos, especially in the specialty crop area. Uh, but a lot of times you'll kind of see it, it starts to move and all of a sudden it's 10 seconds and it's almost to the tomato, and then it's 20 seconds and it finally gets in and it comes back.
And it's really cool when you see it, you know, like, pick that one tomato. And then you talk to grow and the growers just kind of like scowling or laughing and it's like, this is never gonna work. Uh, never gonna work for me. And so I think that's really the, the, one of the big things, uh, that's different about ours is we have that speed to really make it, uh, economical and feasible, uh, for growers.
I think that's kind of the, the number one thing. And then after that, I think the next challenge that we learned from being in the greenhouse so often is even though harvesting is the hardest part, unless you solve the logistics of, you know, what do you do with all these tomatoes once you picked them?
Because you can't always stack 'em very high because they'll damage themselves. Is, how do you handle that so that way you don't need a person to interact with the robot every 10 minutes? And so that's where this whole secondary system of the packing cart is what we call it, comes into play. So I think really it's that, that super fast picking, which is a mixture of the vacuum technology and our perception models and our AI and our motion planning, and then the logistics of being able to pack it and really integrate with the existing greenhouse workflow.
[00:05:38] Mitchell Denton: It's funny that you should mention speed because now that I think about it, a lot of like the tech highlight videos you'll see on YouTube and things like that. More often than not they're playing at like four times speed. So it's kind of funny to, to really ponder on that and it's like, yeah, they probably are operating at a very painful rate.
[00:05:54] Brandon Contino: Yeah, and even those four times speeds, sometimes still it's like, kind of slow and you're like, okay, like that's not too bad. It's just a little slow. And then you see the four X, you're like, oh, okay that's actually really bad.
[00:06:02] Mitchell Denton: Yeah, yeah. Absolutely. So then, what's been your team's biggest challenge so far with your innovative product, and how did you overcome it?
[00:06:11] Brandon Contino: Yeah, so one of the hardest parts for us is we made a very conscious decision in the beginning. A big part of our design philosophy is we really wanna make it easy for growers to adopt their technology. We really wanna work with the growers, understand their problems. And so with that, we made a decision that we want to make a system that does not require any major infrastructural changes into the greenhouse.
Um, which I'm not sure if you, you've ever kind of done any, uh, major engineering projects or anything in the past, but it, it's really nice when you go to an engineering team and say, Hey, you have a blank paper and you can just design whatever you want. Uh, it makes things much easier. But we put those constraints of, Hey, you need to operate in the existing greenhouse workflow.
And we're gonna figure out how we're gonna make this happen. And so it creates a lot more difficulties to build something that can do that.
[00:06:57] Mitchell Denton: Yeah. So then how does Four Growers address sustainability and environmental concerns in their post harvest operations?
[00:07:06] Brandon Contino: Yeah, so one of the things that we really focus on is, part of it is just really supporting the greenhouse space. Uh, so by building the autonomous robot that we have in the analytics, uh, being able to enable this greenhouse industry to really expand. Uh, we think is gonna have very great impacts on the environment.
So we, earlier we kind of talked about the, the water efficiency, the land efficiency, the climate and variability. Uh, with that, I think it's also you start to have greenhouses being located a lot more closely, uh, to the, the end customer. And so with that, you can reduce some of the transportation miles, kinda higher quality produce.
And so I think for us, really enabling the greenhouse is, is a big part right there.
[00:07:45] Mitchell Denton: You mentioned earlier, labour issues within vertical farms and greenhouses. I was just wondering outside of that, what would you say is one of the biggest challenges that these industries face today?
[00:07:57] Brandon Contino: Yeah, so I think number one is definitely, it's it's labour, labour, labour. It's funny, I was talking to a grower a couple weeks ago and I asked him, first time I met him kinda like, what are the biggest challenges? Or like top three, and he said, labour, labour and labour, and I was trying to get something else and he just kept going for labour, like fix the labour problem for me.
Uh, but after that, I think from my experience, what I've seen is energy. Uh, it's much bigger problem in the vertical farming space and vertical farming, I might even say energy's number one before labour. Um, but energy tends to be number two. Uh, energy, just in terms of if you have any type of lights in your operation or if you're doing any type of heat, depending on your climate, energy can sometimes, uh, be a bit of a challenge.
[00:08:37] Mitchell Denton: Yeah. Okay, so then what would you say is the biggest revelation you've had about the AgTech industry while working at Four Growers?
[00:08:46] Brandon Contino: Yeah, there's been a, I've learned a lot definitely about the AgTech industry, uh, really just being able to kind of live in it, be in the greenhouse every day. One of the most fun things I've learned is, uh, do you know what a peduncle is?
[00:08:57] Mitchell Denton: No idea.
[00:08:58] Brandon Contino: Yeah, I had no idea what that was either. Uh, so, you know, uh, on tomatoes, they sometimes come on a vine. That green vine that they come on is called a peduncle. And then the, uh, the little crown thing on top of the tomato is called a calyx.
uh, so it's some of those, uh, those fun words that you end up learning about the space. Um, but I think for me, just kind of. Zooming out a little bit more to the whole AgTech industry, it's been really interesting to see. Uh, you know how the future is gonna move forward. And I think part of this is around the future of the food system and I think the changing in diets to having more higher quality, kind of healthier produce versus in the past we've really been focused on a lot of the, uh, more commodity kind of row crops.
And I think we're now at a point where the software and the automation is finally gonna make it possible where you might see some of those gains that we got for the commodity crops are finally gonna come now to our healthy fruits and vegetables.
[00:09:59] Mitchell Denton: That's cool. That's really exciting. So can you talk about any ongoing research or development projects at Four Growers?
[00:10:06] Brandon Contino: Yeah, so we've, uh, not only have we been able to pick tomatoes, but we've also now picked our second crop. I think we'll be showing a video of that in the next, uh, couple weeks. We'll release the first video. Uh, and then what's been really interesting is as we've been out there in our customer greenhouses picking, so we're deployed now across Canada and the Netherlands and kind of continual everyday customer operations.
Is we're gathering a lot of data around each individual plant. And so before a greenhouse, you don't really have a really good granularity understanding of the environment that you're in or even the plants. Um, and so what we're able to start to do is we're starting to get information down to the per metre squared level.
And being able to provide insights around kind of yield heat maps, uh, yield predictions, and over time continuing to build out some of these analytics capabilities. So not only can we help guarantee that growers can pick their tomatoes when they need to get them picked, but we'll also be able to start increasing their yields and be able to provide additional value to really help us grow even more, more produce, and to become even more efficient with the resources that we're using.
[00:11:12] Mitchell Denton: Yeah. That's fantastic. We, we've been talking a little bit about the future here, and I was just wondering where do you see the future of AgTech heading?
[00:11:21] Brandon Contino: Yeah, I think in AgTech there's a few different verticals. You kind of have the more like automation, AI perception side of AgTech. You also have more the biologicals or maybe some of the like alternative meat side of AgTech. And then you also have some of maybe like the supply chain pieces of AgTech.
So I think there's a, uh, quite a few different areas. I think there's others that didn't even cover right there. Um, For us we're, we're obviously a little bit biased in the automation kind of AI software kind of right in the farm, making improvements, and I think what we're gonna see is that this concept or this dream that people have been talking about of like automated harvesting for your fresh fruits and vegetables is not really a far out dream anymore.
I think it's really becoming a reality and I think what we're gonna see is, as we start getting more of these units in the field, I think we don't even fully understand what the value of the data is gonna be that we collect. And so for example, we've never really been able to have this really high granularity across our entire food system.
But imagine like all the, all the USDA reporting, all the kind of governmental stuff reporting. What if you start to have more of that information, you can start optimising your supply chain better. And what if there's ways that we can actually minimise food waste, because we have a much better understanding of what food is where, and kind of when it was picked and how long ago it was picked.
And I think there's gonna be almost this whole next level of applications that are gonna be built on top of all the data that is now going to be available.
[00:12:51] Mitchell Denton: Yeah. Oh, that's really exciting. On that note, is there a particular group or innovation within the industry that you're excitedly keeping a watchful eye on?
[00:13:00] Brandon Contino: Yeah. One thing for us, uh, is the concept of almost like the next generation sensors. And so a lot of our sensors today, uh, you know, you kind of have your RGB, you have your depth, you have your temperature, your humidity, uh, things like that. I think there's some really interesting research around, non-invasive hyperspectral sensors that are at a lower cost. I think that's really interesting for us because what if in the future today, uh, we pretty much pick all of our, our fruit based off of colour, because that's really one of the only things that you can see and do very quickly.
But with these hyperspectral, what if you could start picking your fruit based off of the sugar content? And so you can really start to get to the next level of really ripeness filtering, making sure you always have that best quality product and picking it right when it's at its peak ripeness. And so that's something that we're really interested in and something that we're, uh, very excited and I think will be coming in the next few years.
[00:13:54] Mitchell Denton: What advice would you give to other entrepreneurs who are interested in starting in the AgTech space?
[00:14:01] Brandon Contino: Yeah, I think on that, I, I would say the number one thing is really to just get in the field. Um, I think, uh, there's been so many companies in AgTech that have started with a cool idea, uh, that have built something and then have gone to a farmer. And the farmer has just said, this doesn't solve any of my problems.
And you know, farmers are very smart. They really know their business and they are, really one of, you know, many of them been around for a long time and they don't always have the biggest margin, so they're very smart and they're very focused on what they do and they're always very busy, and so they don't really have time to have something that doesn't solve a problem and they're not gonna spend money on something that doesn't solve a problem.
And so I think really getting in the field and just talking to the growers or the farmers and understanding their pain points, uh, is one of the, in my opinion, one of the best first things you can do. So that way you build something that is a real solution to them and not just something that creates more noise.
[00:15:00] Mitchell Denton: Absolutely. So Brandon, 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?
[00:15:11] Brandon Contino: Yeah, I think, uh, in regards to, uh, some of our greenhouse grown vegetables today, uh, kind of going back to what we were talking about earlier of automated harvesting being this dream or this thing that's coming in the future, It's really here today. Uh, we've made it to the point where I guess in Australia, if anyone's listening there, hasn't had one of the tomatoes that's been picked by a robot.
But if you live in the Netherlands, or if you live in the US or in Canada, there is a very real possibility that you have now actually eaten a cherry or grape tomato that has been harvested by a robot. And I think that's one of those, uh, really cool things to think about.
[00:15:47] Mitchell Denton: No, that's, that's really cool. And I mean, we're just the doorstep. I feel like Australia's very much pushing innovation in a lot of different areas of the food industry. And I feel like robot harvesting is just around the corner for us, so...
[00:16:00] Brandon Contino: Yeah, I fully agree. I actually was just talking with a grower from Australia a couple weeks ago at a conference and very excited to, uh, to see if there might be an opportunity for us to, uh, have some deployments there in the near future.
[00:16:10] Mitchell Denton: But That'd be awesome. Well, that's all for today's episode of "Let's Talk Farm to Fork". Thanks for listening, and thank you, Brandon, for joining me today.
[00:16:18] Brandon Contino: My pleasure. Thanks for having me.
[00:16:20] Mitchell Denton: If you'd like to know more about Brandon and Four Growers, 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 write a review and share with your friends.
Until next time, you've been listening to Let's Talk Farm to Fork, a PostHarvest podcast.