This AI Startup Is Bringing 400-Pound Drones to a Farm Near You

Old MacDonald had a drone, A-I-A-I-O.

Lisa Lacy Lead AI Writer
Lisa joined CNET after more than 20 years as a reporter and editor. Career highlights include a 2020 story about problematic brand mascots, which preceded historic name changes, and going viral in 2021 after daring to ask, "Why are cans of cranberry sauce labeled upside-down?" She has interviewed celebrities like Serena Williams, Brian Cox and Tracee Ellis Ross. Anna Kendrick said her name sounds like a character from Beverly Hills, 90210. Rick Astley asked if she knew what Rickrolling was. She lives outside Atlanta with her son, two golden retrievers and two cats.
Expertise Technology | AI | Advertising | Retail
Lisa Lacy
4 min read
An AI-guided drone sprays a specific area of a farm.

About 12,000 years ago, humanity shifted from a nomadic lifestyle of hunting and gathering and began establishing settlements with more dependable food supplies in what is known as the Neolithic Revolution. We've been farming ever since.

The number of US farms peaked in 1935 at 6.8 million. Today, that figure is around 1.9 million, with an average farm size of 464 acres, per the US Department of Agriculture.

Over the years, farming methods have changed. In 1837, a young blacksmith by the name of John Deere designed a steel plow for sticky soil, which meant farmers no longer had to stop to scrape off dirt while plowing, for example.

If you guessed that the artificial intelligence revolution is once again changing the game for farmers, you'd be correct. In fact, the agricultural company named after that blacksmith believes we may see fully autonomous corn and soybean farms by 2030.

AI Atlas art badge tag

Just as AI is creating opportunities at work, home and in travel, it's also opening doors within farming. 

Take Hylio. This 9-year-old startup is applying AI and drones for agricultural use, such as applying fertilizer and pesticide, in what CEO Arthur Erickson calls "precision agriculture."

That means if there is a weed infestation or a fungal outbreak on a farm, a farmer can target the affected area rather than having to spray the entire farm.

"With a tractor or an airplane, you're taking a blanket approach, but with the drones, you can actually reduce that chemical input to your crops … [and] you can get way, way, way down," Erickson said. "That's not only good from an economic standpoint, but also you are not as negatively impacting your farm."

Other benefits include reducing water and fuel needs, and easing labor shortages.

To tap into the power of AI and drones to target crops, weeds and bugs, the system has to learn how to distinguish between crops and weeds so the drone knows what and where to spray. But because the difference between a weed and a crop can be hard to discern sometimes, the machine-learning models have to be well-trained with human input first.

"In the old days, you would do [scouting] by hand, so you'd have an agronomist go out to the field and they would do surveys, maybe they put out a grid, they put traps out to see if they could catch insects, and then you could count to see how much insect pressure you get per acre and whatnot," Erickson said.

Now, with AI, machine learning and high-resolution cameras or sensors on drones, farmers can scout a field in minutes, run findings through the algorithm and identify the areas to focus on.

Read moreI Deployed a Fleet of Lawn Robots to Save Me More Than 65 Hours of Work This Summer

A typical crop-scouting drone can cover 300 to 400 acres per day while the application drones — or those that carry and release "payload" like fertilizer or pesticide — can fly about 50 acres per hour and cover 400 acres in an eight-hour day.

These drones are much bigger than the hobby or parcel delivery drones we're used to. They're not quite as big as a car — more like a motorcycle.

Hylio's biggest drone is a 200-pound helicopter-style drone that carries up to 200 pounds.

"It can carry a full-grown adult male," Erickson said. "In fact, there's videos of people hanging onto the drones, which is really dumb and you shouldn't do that, but people do it."

But these drones can't fly very far or for very long.

"The trick is, you carry this really heavy payload, you dump … the payload over the crop in about eight or nine minutes, come back home, you refill payload and then you swap batteries," he added.

In February, the FAA gave Hylio the OK for one operator to fly up to three drones above the 55-pound weight class limit.

Prior to that, the FAA required two people per drone in the field.

"If you had three drones, you'd have to have six people, which is, of course, counterintuitive to the logic of drones because drones are autonomous, so you'd want to force multiply what one person can do with a quote-unquote army of robots underneath them," Erickson said.

To operate these drones, farmers need two licenses from the FAA: a general drone pilot's license and an agricultural applicator's license, which tells the FAA you're going to be aerially applying crop inputs.

Hylio received some angel funding in 2019 and did an equity raise through crowdfunding platform StartEngine, but is not currently seeking capital.

The company is building a new production facility for its drones, which is about five times bigger than its current facility. Erickson hopes to be able to manufacture 2,000 to 3,000 drones per year in two to three years -- it currently makes between 500 and 1,000 annually. Erickson estimates it takes about 12 to 24 hours of labor time to build each drone.

This is one of a series of short profiles of AI startups, to help you get a handle on the landscape of artificial intelligence activity going on. For more on AI, see our new AI Atlas hub, which includes product reviews, news, tips and explainers.

Editors' note: CNET used an AI engine to help create several dozen stories, which are labeled accordingly. The note you're reading is attached to articles that deal substantively with the topic of AI but are created entirely by our expert editors and writers. For more, see our AI policy.