Podcast: Robot Farmers Bring Back Biodiversity

author avatar
Podcast: Robot Farmers Bring Back Biodiversity

In this episode, we talk about how researchers from the Technical University of Denmark are tackling the issue of monoculture farming by utilizing robotics and next-gen image analysis.

In this episode, we talk about how researchers from the Technical University of Denmark are tackling the issue of monoculture farming by utilizing robotics and next-gen image analysis.

This podcast is sponsored by Mouser Electronics


(3:15) - Agricultural Robots Can Help Improve Biodiversity

This episode was brought to you by Mouser, our favorite place to get electronics parts for any project, whether it be a hobby at home or a prototype for work. Click HERE to learn more about how energy harvesting robots can make robotic applications in remote environments - like farming - viable!


Howdy folks, welcome back to the Next Byte podcast. And today we're talking all about agriculture, specifically about how we gotta bring biodiversity back or we all doomed. Hashtag save the bees. Let's get into it.

I'm Daniel, and I'm Farbod. And this is the NextByte Podcast. Every week, we explore interesting and impactful tech and engineering content from Wevolver.com and deliver it to you in bite sized episodes that are easy to understand, regardless of your background. 

Farbod: All right, people, we're gonna be talking about agricultural robotics, but before we do so, let's talk about an ad from today's sponsor, Mouser Electronics. Now, you know we love Mouser. They're one of the world's biggest electronic suppliers, and by being one of the world's biggest electronic suppliers, they have a lot of connections. They have connections to industry, they have connections to academia, and they have a lot of insight because of those connections. Sometimes they write about them, and they wrote this really cool article about energy harvesting robots. Now, you might be wondering, well, how's that related to today's episode about agriculture? Well, let me tell you. These folks are talking about why right now we have this approach of charging robots or EVs or whatever that requires them to come back to a base station, plugging them in, and then they have this like, let's say one-hour, two-hour, however many hours of downtime. Well, if they could benefit, if they could use their surroundings to charge themselves back up, then they could stay in operation for much, much longer. A great use case of this would be something like an agricultural robot, which is just working over and over and over again and able to feel itself from its surroundings, a.k.a. the sun or the wind and keep on going.

Daniel: Well, and it's, it's all about agility, right? You're no longer constrained by battery life or time to charge or location of charger or how long your extension cord is, right? If you've got the ability to do this in a manner that's feasible, right? Harvest energy from the environment around you to power these robots, you've got the flexibility to say, put a bunch of agricultural robots out in a field and you never need to go plug them in again or you don't have to plug them in for a very long time because they're able to use the environment to power them, make them more flexible, give them more uptime, et cetera.

Farbod: Yeah. So, if you guys are interested in learning about this, definitely check out the technical resource we've linked in the show notes. This is one example. They actually go into multiple examples. The shortcomings, yeah, they don't just talk about the positives, they talk about the downsides as well. And, yeah, if you liked it, you should definitely go give it a look.

Daniel: Well, and we always say, Mouser does a great job of providing you with not only the recipe, but they also give you the ingredients to cook the secret sauce. This is another one where they provide actual parts that you could use to help build your own energy harvesting robot that surrounds, it uses the surroundings to out power it. So again, huge hat tip to Mouser, the reason why we love them so much. And it's just exemplified in this technical resource, which we're gonna link into the show notes for today.

Farbod: Yes, sir. That said, let's move on to today's article. Now this one's coming from the Technical University of Denmark, and we're gonna be talking about robotic farmers. Before we do that though, we need a little trip down memory lane. We gotta have a little history lesson.

Daniel: I love history.

Farbod: I do too. I love a good story. Now, apparently before World War II, the approach to farming was that the farmer would have a piece of land and they would grow multiple crops on it. When World War II came around, and industrialization was obviously booming, we really started aiming for optimizing our yields and going for economies of scale.

Daniel: Well, and really what we did right was industrialized agriculture.

Farbod: Yes.

Daniel: It went from this thing where farmers, like you said, could cover a diverse swath of crops and it's focused on producing crops to sell and to trade to some extent, but also I think to some extent it was also based on subsistence and when we turned that into an industrial machine where it's like, let's churn as much dollars per square foot of acre of land that I have, right? That changed the way that agriculture works and the way that farmers were incentivized and encouraged to interact with the land. And it turned it from, hey, let me build, or let me grow four or five different crops that all grow really well in this region to let me focus on one in the United States, the financial incentives align that to be corn.

Farbod: That is true.

Daniel: I spent a lot of time in the Midwest for work and I'm looking at like, fields and fields and fields of corn. And what that's called when you have just one species that you're growing, that's called a monoculture. And there's a lot of conversation about how since we walked down this historical path. Since World War two, this monoculture industrial farming type society that we've built, not just in the U S but around the world. And you know, this is a University in Denmark that's commenting on this has led to a lot of lack of biodiversity, has led to a lot of decrease of health in the environment, so to speak, especially around agricultural areas. So essentially turning everything into the super industrial machine that we have just to produce a bunch of crops is actually damaging the soil, our ecosystem, and our relationship with the earth in the long run.

Farbod: Another crazy part about this, about the monoculture, is that, like you said, because we were maximizing for yield, like how do we feed the most amount of people for the least amount of money and maximize our usage of the land. We went to the one crop system, but the one crop system was also kind of a requirement for this new efficient method of farming that relied so heavily on these huge farming machines. Like those farming machines are great at extracting one type of crop or planting one type of seed. So, the land needs to be made so that it can comply with the requirements of the machine as well.

Daniel: Well, and you know, whether it was the chicken or the egg, we don't know, but because we're only using one crop, it allowed us to build these huge machines because you're not, you don't have to be agile. You can plow through tons and tons of land and plant the exact same crop row, row, row after row after row. So, we choose the plants, select a monoculture, this creates big machines. And because there are big machines, we can't plant different types of crops. And it's a paradox that keeps perpetuating. And some of the impacts of this monoculture and lack of biodiversity includes reduced pollinator populations that think that this is part of the contributing factors of why we have such a decreasing population of bees.

Farbod: Hashtag save the bees.

Daniel: Yeah, save the bees, man. Also increased vulnerability to pests and diseases, which then causes us to have greater reliance on chemical pesticides, which further harm the environment. And people like Monsanto will say it doesn't harm us as the consumers, but I've got some sub skepticism around that too, right? A greater reliance on chemical pesticides and herbicides. It can't be good for humans as well, right? Whether we're getting it through the drinking water or directly through the crops that we're consuming, if it's for produce. Basically, lack of biodiversity is part of what's causing such issues with agriculture as it stands today.

Farbod: And the funny thing is that the monoculture lowered the biodiversity, which lowered the number of predators that would be going after these pests. So now we're using these herbicides and chemicals to make sure that our farms can keep on chugging away. All that to say, it's not just the Technical University of Denmark, there's a report from the UN Food and Agriculture Organization that is confirming that exact same idea, essentially saying, hey, this approach isn't good. We need to bring back biodiversity somehow, right? Now we've talked about why biodiversity has been kind of ruled out. For the most part, the developed world benefits from this monocultural system where we can get a lot of yield, we can get a lot of food to a lot of people.

Daniel: Use huge combine machines to do a lot of the manual labor instead of humans, right?

Farbod: So, this logic of like, do we just go back to the way things were before World War II? That doesn't really make sense either. Like we can't sustain our population with that, nevertheless grow it. So, what are we to do? Well, the Technical University of Denmark was like, well, robotics might be a good solution. And in fact, they're saying this from a good perspective. They're saying this from a good stance because they have experience doing this. I didn't know this, but in 2021, they came up with the, I hope I don't butcher this name, the Galirumi robot, which works in organic farms to remove weeds mechanically. So, they've already had some reps in setting up robotic farmers, and they were like, we think there's actually an approach here with robots that could kind of give us best of both worlds.

Daniel: And again, the competitive advantage that designing new small agile robots provides is we will have the flexibility to use different parts of our land for producing different types of crops. We don't need to create these acres wide rows of just one crop to be able to use our machines to help assist with that, which is kind of a requirement when you're using something like a combine machine. You can use these small autonomous robots to do things like sowing, like weeding, like checking for pests, watering, spraying, harvesting, et cetera. Using all the advanced sensors and instruments that are applied to it. But they're also they're going to have a smaller footprint.

Farbod: Absolutely.

Daniel: So, you're able to use different types of crops because it's got a bunch of different sensors and a bunch of different functionalities on there. But also, because they're small, you don't need to plant giant wide rows of these crops. You can actually do a couple rows of corn and then a couple rows of wheat and then a couple rows of potatoes and a couple rows of beets and tomatoes and jalapeno peppers, make it look more like your backyard garden, but offer that same flexibility to farmers who do that as their job.

Farbod: Yeah. And you can actually start making a more efficient use of the land space as well because you're no longer constrained to I need to make this farmland viable for a massive farm equipment to go through like they were talking about how they modified the landscape to make sure that it's ready for a big combine machine to come go through. In addition to that, something that I wasn't taking into account either, but the article thankfully shouts out, is if you were to have some sort of infestation within a modern farm right now, like let's say a potato farm or a corn farm, because it's all one crop, it's very likely that that pest is gonna spread very quickly across the board, right? But by having that diversity, let's say you have a row of corn and then a row of, I don't know, blueberries and a row of strawberries, whatever. Those pests are likely not gonna target different types of fruits. They're gonna attack the one corn row, you're gonna be able to catch it, get rid of it, and then you prevent the spread from getting absolutely terrible.

Daniel: And it's basically like a farmer's version of hedging their bets, right? They've got now the ability to, instead of going nuclear and using giant machines and then using and applying tons of chemicals to help prevent these risks from occurring. You can use a small robot that's agile enough that has the different functionality to try and address these pests or address these weeds, et cetera. In addition, like you're saying, if you've got a vast spread, a diverse set of a bunch of different types of crops, like you're saying, if you've got one type of weed or fungus or pest that particularly affects your strawberries, you're not gonna lose your entire crop because now you've got the flexibility to use one eighth of your land for strawberries and the other seven eighths for other different types of produce that you're interested in producing.

Farbod: Yeah, yeah. Now, we've talked about the general part of the secret sauce, right? Like they're gonna use a robotic system that has all these advantages, but these researchers were like, hey, we know robotics could be good, but we're still some time away from getting fully autonomous robots. And in fact, a robotics system that's not fully autonomous doesn't really have much value at it because the farmer's still gonna have to supervise it and we're just kind of back to square one. So, they were like, what would it take for us to get there? And obviously, pretty much like every other autonomous system developer, including the EV manufacturers like Tesla, they focused on the sensors. They were like, there's lidar, there's cameras, that's usually how we ingest the data about our surroundings. What's missing there for us to react as fast as we possibly can and make decisions about our surroundings? They really zoned in on the cameras. Now, they have, they did this comparison of, we have your traditional shutter cameras, like the DSLRs, right? That can capture 120 frames a second, and at maximum, let's say you process that in an instant on your computer and make decisions about what to do next. They were like, what if we look at event cameras? Do you know about event cameras? Because I didn't before this.

Daniel: No, I didn't either. And I also saw them call them neuromorphic cameras until we did research for this episode. I had no idea what they were.

Farbod: It's crazy. I've also seen it called Silicon Retinas. And as all those names kind of imply, they are supposed to mimic like how human beings essentially perceive the world. At their core, they are, instead of capturing an event constantly as things are happening. They only capture an event when something new changes in the environment. So, within this silicon every pixel is like its own camera and every brightness measurement that it's taking is where it's registering an event? So, if the brightness right now is a hundred and someone flashes a light at that pixel and it goes to 120 that registers as an event. That reference changes so now it knows every time he changes from 120, that's a new event.

Daniel: Well, and in comparison, to what you're saying with DSLR cameras where you've got the best ones of these can capture 120, maybe 240 frames per second. With these neuromorphic cameras or these event cameras using this algorithmic processing of only capturing frames when something changes, it allows it to great massively increase the overall frames per second it can capture, right? Hundreds of thousands, maybe up to a million frames per second.

Farbod: Yes, up to a million frames per second.

Daniel: Which is incredible.

Farbod: That is a logarithmic increase in the amount of response time that you're decreasing, right? So, by being able to capture data much quicker, you're able to respond to that data, process it much, much quicker as well. So, if you're concerned about deploying these robots at scale within a farm, because you don't want them to do the wrong thing, you don't want them to hurt people, you want them to be able to respond to their surroundings as ideally fast as a human being could, if not faster. This piece of technology will allow them to do that. And that is, I would say, I don't know, the spicy pepper. I forgot what that pepper was called in Drake and Josh, that one in the, you know what I'm talking about, the…

Daniel: I can't relate here, unfortunately.

Farbod: Okay, nevermind. Well, this is the secret on the secret sauce, the little garnish that is making it all come together.

Daniel: Yeah, I'm with you, man. And I'm honestly compelled by this, and like you said, understanding all the different fields where these neuromorphic cameras could be applied. I'm sure there's folks in the other autonomous vehicle space, like such as cars, such as drones, etc., that are looking at this type of camera to help improve response times and controls across all different types of these autonomous vehicles, not just small farming vehicles, right?

Farbod: Yeah. I don't know, I'm pretty stoked about this. The future of it seems pretty promising. The so what, generally, I feel like we've touched on it, but imagine farmers with smaller pieces of land, they can now leverage farming tools that maybe just did not exist to them or were not effective enough to use. These robotic systems allows you to get some of that going, even in the smallest pieces of land. We make better use, we get to bring back that biodiversity, which is greatly impacting our environment and our sustainability kind of like a win all around because we're not giving anything up. We're just, I don't know, expanding on this privilege of being able to automate our farming resources, but in a way that helps us and helps the environment.

Daniel: That's what I was going to mention is I think the incentives for something like this, if and when it gets developed and hits the market, which they say could be as far as 10 years out, just as a caveat there, you can't go buy one of these what do they call it? SAVA (Safe Autonomous Vehicles for Agriculture), right away.

Farbod: So, I can't make Farbod farm happen overnight.

Daniel: I mean, but they're saying the technology is there. We need to industrialize it and there needs to be a demand for that in the market. And that's what they think will take around 10 years to become ubiquitous. So, I think the incentives are there for farmers to do that on their own, right? Because you can talk about increased yields, you can talk about this decreased use of certain chemicals. The fact that it will make their soil more rich. One of the things that the UN agency is the FAO (Farming and Agriculture Organization) mentioned is that focusing on biodiversity can massively increase soil health, crop resilience. These are things that farmers care about to increase their yields and to make sure that they aren't basically wasting money planting crops that aren't gonna survive. That being said, I would love to see your cultural shift as well to where it becomes one of the marketing things that farmers can use, you would go look for produce at the store that has the little biodiverse sticker on it, the way that some people look for organic stick.

Farbod: Or GMO or pastries or whatever.

Daniel: I would go shop for a biodiverse farm type produce if I know that it's going to be crucial to achieving our ecological balances, health of plants, health of all the animals, health of our water supply. One of the things they call it, they call it ecosystem services. So that’s all the wastes that the ecosystem serves us humans, right? Monoculture is massively detrimental to the ecosystem services. So, I would love to see a cultural shift where biodiversity becomes an important selling point for certain types of produce. From everything from this helps reduce the amount of chemicals we use, this makes agriculture more sustainable, this can help save the bees, et cetera. I would love to see that become something that consumers are aware of, not just something that happens in the background because it works well for farmers.

Farbod: Well, that's a great point. I love that. I think it's a good point throughout the episode.

Daniel: Yeah, can you wrap us up?

Farbod: Let's try. Alright, so biodiversity is a big problem. Essentially after World War II, we try to maximize the yield that we were getting from farms. So, we planted one crop that was great for big machines to plow through and give us the most amount of product. Well that actually kind of started leading to some problems because the lack of biodiversity meant that we didn't have enough predators to go after all these pests, so we started using more pesticides, we started to use more chemicals, and we're getting less and less nutritious rich soil, which means, according to the United Nations Food and Agriculture Organization, we need to go back. We need to figure out how to solve this problem. And these folks at the Technical University of Denmark have an idea. Let's use robotics. Robots can do all these specialized tasks, they don't need massive spaces of monoculture, like one type of food, one type of crop. They can do it all. They can do it in less space. So, we can still get these amazing yields that we need to sustain our populations or even grow it, but be a little bit nicer to the environment and just be nicer to the, be more sustainable.

Daniel: Yeah, dig it, man. I love the part about being nicer to the environment, right? That's really what, that's the luxury that this allows us is farmers, you can still use automation, you can be more agile, you can have better crop yields, et cetera, but also do it in a way that allows us to interact more kindly with the environment and not be putting our future in jeopardy just to produce tons and tons of corn today.

Farbod: Yes, sir, that's the goal. All right, folks, thank you so much for listening. And as always, we'll catch you in the next one.

Daniel: Peace.

As always, you can find these and other interesting & impactful engineering articles on Wevolver.com.

To learn more about this show, please visit our shows page. By following the page, you will get automatic updates by email when a new show is published. Be sure to give us a follow and review on Apple podcasts, Spotify, and most of your favorite podcast platforms!


The Next Byte: We're two engineers on a mission to simplify complex science & technology, making it easy to understand. In each episode of our show, we dive into world-changing tech (such as AI, robotics, 3D printing, IoT, & much more), all while keeping it entertaining & engaging along the way.


The Next Byte Newsletter

Fuel your tech-savvy curiosity with “byte” sized digests of tech breakthroughs.

More by The Next Byte

The Next Byte Podcast is hosted by two young engineers - Daniel and Farbod - who select the most interesting tech/engineering content on Wevolver.com and deliver it in bite-sized episodes that are easy to understand regardless of your background. If you'd like to stay up to date with our latest ep...