Autonomous technology has grown deep roots in the farming sector

Quiz time. Among the hundreds of companies offering grand visions of self-driving and connected-vehicle futures, name the one that's had both technologies deployed in commercial operations for nearly a decade.

The answer isn't one of the tech giants from Silicon Valley that pursued self-driving developments following the end of the DARPA challenges. Nor is it one of the traditional automakers. The answer is John Deere.

Farming and mining have served as formative proving grounds for technologies still waiting to make their mark in the transportation realm. With deployment of automated-driving systems in 2002 and embedded modems in tractors in 2011, John Deere has been at the forefront in determining where the rubber meets the road. Or where the combine meets the corn.

John Deere says farmers using its AutoTrac systems and JD Link connected-data platform see tangible returns of $5 to $15 per acre. For a farmer with 5,000 acres, "those become real numbers real fast," says Joel Hergenreter, machine automation strategy lead for the company.

Using GPS, the company's automated driving systems have delivered sub-inch-level accuracy since 2003. That helps farmers avoid running over their own crops or leaving fertile soil unplanted.

Connected-vehicle technology helps tractors talk to each other, not only conveying safety information that keeps them from bumping into each other, but data on which fields have been planted or harvested. Data stored in John Deere's cloud on when seed was planted, how deep it was planted, which varieties were used, when herbicides were sprayed, can be used to make smarter decisions.

"Farmers have about 40 chances at building a crop," says Deanna Kovar, global director of John Deere's Ag & Turf Division. "It's not an infinite number. You don't have as quick of an ability to learn — it might take a whole year to learn and understand. So data can help accelerate those learnings."

The particulars of automated systems in agriculture and automotive are different. Cars use lidar to help detect objects such as pedestrians and use cameras to localize themselves from images of road signs and bridges. That doesn't work among featureless fields of crops.

But fundamentals remain the same. Both agriculture and automotive are grappling with the same basic challenge — how to serve growing populations on a finite swath of land.

For employees, that can be a rewarding pursuit. Farming and mining served as stopovers for many engineers who eventually went to work on self-driving passenger vehicles, but a few have followed the reverse trajectory. Willy Pell worked for Google, then moved to Blue River Technology, a computer vision and machine learning company acquired by John Deere in 2017. He's thrilled.

"Self-driving cars are awesome, but I wake up every morning and say, 'I helped put food on somebody's plate in a sustainable way,' " he said. "It's very mission driven. And the self-driving world is extremely populated. There's thousands of engineers working those problems, and their impact is relatively small. In ag, they could have a huge impact."