(light music) - [Bret] In a small rural Minnesota city, a University Research Center is pioneering designs for solar-powered robots that could revolutionize non-chemical weed control in farm fields.

- We have been doing work in robotics since 2017.

There's a company in Norway that's been doing research with these robots.

They're not quite selling 'em to anybody yet, but they were, it was something that we could purchase as a functioning robot with four-wheel steering and then it's configurable to whatever width and length you want to adjust it to.

- [Bret] The West Central Research and Outreach Center in Morris, Minnesota first developed an autonomous cowbot to remove weeds in rocky cattle pastures that mowers couldn't reach.

Then researchers moved on to design a robot that could cultivate agricultural crops.

- We listed about 30 different concepts and will those down to a half a dozen that we wanted to investigate farther.

We tried to laser, we tried electrocution, we tried a number of different things and we finally hit on the, just sawing 'em off with a saw blade or a trimmer, if you will.

- [Bret] "Prairie Sportsman" was at one of the Research Center's first trials of a robotic Weed Terminator.

- [Eric] You're really here at that early stage.

We still have a lot to learn.

- [Pete] The parts that we're using on this prototype are actually commercially available trimmers that you would purchase at the store and use use in your yard.

The two in the middle are following a straight line through the row and then the two on either side, one on either side are actually kind of bumping up against the corn plant as they go along the row.

It's mostly meant for late term corn.

- You can cultivate a crop up until a certain point.

Once the crop is a certain height, you can't take a tractor and cultivator out there anymore.

And that's when the Weed Terminator, as we termed it, coined it, takes over and can go out and do the weeding from that point on.

- So the corn is gonna be maybe two and a half, three feet tall, all the way up to the point where the corn starts tasseling and pollinating.

- [Eric] We're cutting about two inches off the ground, two to three inches off the ground.

The corn canopy is gonna close and those weeds will not go to seed.

- If we can keep the weed from going to seed, then we can keep the weed pressure down in future generations, future crop years.

(light music) - [Eric] We're making vehicles of the size that can be electric.

We can basically charge the batteries in these robots with solar energies.

- [Pete] The system that we have in place right now is GPS guided.

And so as long as the robot has electrical energy in it, we could drive day and night.

- We've got four researchers working under Dr. Volkan Isler, part of the Computer Science Department on the Twin Cities Campus.

They're all working on different aspects of the navigation and the control system.

- We are making it do what it does autonomously.

(light music) Right now what we have is something where we use GPS information.

So like what do you have on your phone when you walk around and when you drive in a car, you get the GPS information.

We similarly have a GPS module on our robot.

We use the information from the GPS to do the navigation.

- [Pete] Most of the corn in West Central Minnesota is planted with a GPS-assisted tractor.

And so they already know where the corn plants, where the corn rows are located.

And so once we know that information, we can send this robot out into that same corn field where we already know where the corn plants are and it will be able to navigate through those corn rows.

(soft twangy music) - We configured it for corn fields, for 30-inch rows, in our case.

This field that we are converting to organic.

So it has not had herbicide sprayed on it.

And as you can see, there are a lot of weeds in it.

The Weed Terminator is weeding one row at a time.

So that seems maybe pretty inefficient.

But what we kind of envision is maybe you have 30 one-row robots out in the field doing the same work that one 30-row implement pulled behind a tractor would do.

Farm equipment gets larger and larger, bigger and bigger because you're trying to increase the productivity of one driver.

And if you don't need a driver anymore, now we don't have to have stuff that's so large, we can bring this equipment back down to size where we can do a better job of weeding in one row at a time 'cause we've seen the actual plants and the weeds and we can deal with those individually.

- At Farm Fest near Redwood Falls, this past summer, we actually ran the robot up and down the rows of corn, weeding that corn and it was very well received.

Most of the people who saw the demonstrations down there were very receptive to it.

And I think mostly the manpower situation that exists in ag today, there's just not enough people to do the work sometimes.

Nobody said, "No, it won't work.

I don't wanna do this."

They were all, "How soon can we get this?

How many acres will it do?

How fast will it go?"

You know, so they're very interested in wanting to know more specifics about it.

(light music) In this particular case, it probably will be some younger farmers, farmers who could see the economic benefits of this.

Organic farmers would probably be the first ones to adopt this.

- Organic farmers have a much more difficult time controlling weeds than conventional farmers do.

Even in a conventional farmer who's maybe seen some herbicide tolerant plants like Palmer Amaranths, those aren't gonna just stay in his field.

Those weeds are gonna propagate and be all around the state and it's gonna be a problem for everybody.

The cost of building the robot was probably 60 to $100,000 all in with all the different tools and things we had to do and try and fail at.

And that's a lot of money for a one-row robot.

But again, this is the first one and this is kind of like how expensive was the first computer.

As things get manufactured more, then those prices go down.

- [Pratik] The next step that we are currently working on is how to integrate cameras on it.

The Jeep is not able to see what's coming in front of it, but the camera can see and detect weeds, can detect rocks, can detect gopher mounts, can detect a ramp.

So we are working on that technology right now.

- We've really only had the robot about a year and we're just working through some of the initial bugs and getting it to the point where we can test it in the field.

In the end, if it doesn't, if it's not cost competitive, it's not gonna be adopted.

So it's just like anything else.

I mean if EVs were always twice as expensive as a gas-powered vehicle, no one would, I mean only a few people would buy 'em, right?

So it's gonna have to, you know, you're always thinking about this five to 10 years down the road and if you just look at what say Tesla has done with EVs on the road and even autonomous driving on a highway and in traffic, and how much easier it is to have an autonomous vehicle in a corn field where you're not worried about pedestrians.

(light music) - We have another phase of this project that we are starting this next year where we're actually gonna have multiple robots and so we're gonna look at two robots out in the field at the same time.

We really had three goals with this project.

One of 'em was to eliminate herbicides, which we've done.

There are a lot of herbicide resistant weeds that Roundup won't touch at this point.

Removing the operator from the equation was one of our goals.

And then decarbonizing ag, our system right now doesn't emit any greenhouse gases.

We recharge it with a solar charging station that's got solar panels on it.

We're not using any energy off the grid to power this system.

I think once people start seeing a system like this out in the field and seeing the advantages of it, I think you're gonna get a wide variety of people to adopt it.