This AI startup is bringing 400-pound drones to a farm near you

About 12,000 years ago, humanity abandoned its nomadic hunting and gathering lifestyle and began establishing colonies with more reliable food supplies during what is known as the Neolithic Revolution. We have been farming ever since.

The number of American farms peaked in 1935 at 6.8 million. Today, that number is about 1.9 million, with an average farm size of 464 acres, according to the U.S. Department of Agriculture.

Over the years, agricultural methods have changed. In 1837, a young blacksmith named John Deere designed a steel plow for sticky soils, allowing farmers to no longer have to stop to scrape the earth while plowing, for example.

If you guessed that the artificial intelligence revolution is once again a game changer for farmers, you would be right. In fact, the agricultural company that bears this blacksmith's name estimates that we could see fully self-sustaining corn and soy farms by 2030.

Just as AI is creating opportunities at work, at home and while traveling, it is also opening doors within agriculture.

Take Hylio. This 9-year-old startup applies AI and drones for agricultural purposes, such as applying fertilizers and pesticides, in what CEO Arthur Erickson calls “precision agriculture.”

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

“With a tractor or a plane, you take a holistic approach, but with drones you can actually reduce that chemical input into your crops… [and] you can go very, very far,” Erickson said. “It's not only good from an economic standpoint, but also you don't have as negative an impact on your farm.”

Other benefits include reducing water and fuel requirements and alleviating labor shortages.

To harness the power of AI and drones to target crops, weeds and insects, the system must learn 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 sometimes be difficult to discern, machine learning models must first be well trained with human input.

“In the past, you did [scouting] by hand, so you would ask an agronomist to go out into the field and do some surveys, maybe he would set up a grid, he would set up some traps to see if he could catch any insects, and then you could count to see how many insect pressure you get per acre and so on,” Erickson said.

Now, thanks to AI, machine learning and high-resolution cameras or sensors installed on drones, farmers can explore a field in minutes, analyze the results via the algorithm and identify areas to focus on .

Learn more: I deployed a fleet of lawn robots to save me over 65 hours of work this summer

A typical crop reconnaissance drone can cover 300 to 400 acres per day while application drones – or those that carry and release a “payload” like fertilizer or pesticide – can fly around 50 acres per hour and cover 400 acres in eight hours. .

These drones are much larger than the leisure or package delivery drones we are used to. They are not as big as a car, but more like a motorcycle.

Hylio's largest drone is a 200-pound helicopter-style drone that can carry up to 200 pounds.

“It can carry a full-grown adult male,” Erickson said. “Actually, there are videos of people hanging on to drones, which is really stupid and you shouldn't do that, but people do it.”

But these drones cannot fly very far or for very long.

“The thing is, you carry this really heavy payload, you throw it on the crop in about eight or nine minutes, you come back home, you fill the payload, then you swap the batteries,” a- he added.

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

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

“If you had three drones, you would need six people, which is of course counter-intuitive to drone logic because drones are autonomous, so you would want to force multiply what one person can do with a quote. an army of robots beneath them,” Erickson said.

To use these drones, farmers need two licenses from the FAA: a general drone pilot license and an agricultural applicator license, which tells the FAA that you are going to apply agricultural inputs by air.

Hylio received angel funding in 2019 and completed a fundraising round through crowdfunding platform StartEngine, but is not currently seeking capital.

The company is building a new production facility for its drones, approximately five times larger than its current facility. Erickson hopes to be able to make 2,000 to 3,000 drones per year within two to three years – it currently makes between 500 and 1,000 per year. Erickson estimates that it takes about 12 to 24 hours of work to build each drone.

This is part of a series of short profiles of AI startups, to help you get a sense of the current artificial intelligence business landscape. To learn more about AI, check out our new AI Atlas hub, which includes product reviews, news, tips, and explainers.

Editor's note: CNET used an AI engine to create several dozen stories, labeled accordingly. The note you are reading is attached to articles that cover the topic of AI in depth, but are created entirely by our expert editors and editors. To find out more, see our AI Policy.