How Deere plans to build a world of fully autonomous farming by 2030

Could John Deere become one of the world’s leading artificial intelligence and robotics companies in the next decade, along with Tesla and Silicon Valley tech giants?

That notion doesn’t jive with the common image of the 185-year-old company as a heavy-metal manufacturer of tractors, bulldozers and lawnmowers painted green and yellow.

That’s what the company sees in the future, according to Jorge Heraud, vice president of automation and autonomy at Deere’s Moline, Illinois company, a preview of which was shown last January at Deere’s Las Vegas Consumer Electronics Show. A fully autonomous 8R farm tractor driven by artificial intelligence rather than a farmer behind the wheel.

The autonomous 8R is the culmination of nearly two decades of strategic planning and investment by Deere in automation, data analytics, GPS guidance, Internet of Things connectivity and software engineering. While much of this R&D is in-house, the company is also working on multiple acquisitions and partnerships with agtech startups, gathering know-how and talent.

“It comes from our recognition that technology will drive value creation and increase productivity, profitability and sustainability for farmers,” Heraud said.

While Deere made a big splash at CES and has the investment community interested, Jefferies equity research analyst Stephen Volkmann said, “We said it’s very, very, very early in the process.”

“The total global fleet of Deere autonomous tractors is less than 50 today,” he said. While Deere’s goal is to have a fully autonomous farming system for row crops by 2030, Volkmann said, “In Wall Street time, that’s an eternity.”

Currently, Deere creates value and profit with well-built automated systems that can be retrofitted into its existing tractors, such as GPS-based self-driving and precision seeding that measures how deep and far to plant. Those steps have to be in place before you can put full autonomy around them, Volkmann said.

The autonomous 8R represents a giant leap forward in current technology, not to mention the marketing benefit. “Before it was presented at CES, everyone was thinking [full autonomy] It was pie in the sky,” said Scott Shearer, chair of the Department of Food, Agricultural and Biological Engineering at Ohio State University.

According to Shearer, 30 different autonomous tractor projects are in the works worldwide, but none are commercially available. “But when Deere, which has 60% of the tractor market in North America, comes out with a vehicle, reality sets in,” Shearer said.

This reality reflects Deere’s autonomy strategy. “The artificial intelligence we’re using involves computer vision and machine learning,” said Heraud, whose science was well underway at Blue River Technology, a Silicon Valley startup that Deere bought for $305 million in 2017 — also a Blue River co-founder and CEO. Blue River’s “look and spray” robotic platform uses dozens of sophisticated cameras and processors to separate weeds from crops while applying herbicides.

The autonomous tractor is equipped with a 120-foot-wide boom lined with 6 pairs of stereo cameras that “see” an obstacle in the field—whether it’s a rock, tree, or person—and determine its size and relative distance. The images captured by the cameras are passed through a deep neural network that classifies each pixel in about 100 milliseconds and decides whether the tractor should continue moving.

“We selected hundreds of thousands of images from different farm locations, different weather and lighting conditions,” said Heraud, “so that with machine learning, the tractor can understand what it sees and react accordingly. This capability also allows the farmer, instead of being on the tractor, to remote control while doing something.”

Heraud was talking about autonomous driving, another piece of Deere’s agtech puzzle that came together when it bought Bear Flag Robotics for $250 million last year. Also, Silicon Valley startup Bear Flag’s autonomous navigation system, which launched in 2017, can be equipped with existing tractors, in this case Deere’s latest 8R model, which will be launched in 2020.

Since its CES unveiling, Deere has acquired AI assets from two other agtech pioneers. In April, Deere formed a joint venture with GUSS Automation to develop semi-autonomous orchard and vineyard sprayers. Using artificial intelligence and the Internet of Things, multiple GUSS (Global Unmanned Spray System) sprinklers can be controlled remotely by one operator, and up to eight sprinklers can be operated simultaneously from a laptop. GUSS can detect trees and determine how much to spray each, regardless of height or canopy size.

A month later, Deere announced it had acquired numerous patents and other intellectual property from artificial intelligence startup Light, according to The Robot Report. Light’s depth perception platform enhances existing stereo-vision systems by using additional cameras that mimic the structure of the human eye to provide more accurate 3D vision. Deere plans to integrate the Light platform into future versions of autonomous farm equipment.

To closely follow other agtech R&D efforts, Deere has created a Startup Collaborator program to test innovative technologies with customers and dealers who do not have a more formal business relationship. “The hope is that they find the diamonds before they are discovered [competitors] and keep them in the loop,” Volkmann said. Current products include Pittsburgh-based Four Growers, which provides robotic harvesting and analytics for high-value crops ranging from greenhouse tomatoes, and Philadelphia-based Burro, which produces small fruit, autonomous robots that can help farm workers with various handling tasks. .

Not surprisingly, Deere’s biggest rivals are also developing automation and autonomy for its agricultural machinery. AGCO, whose brands are Massey Ferguson and Fendt, “has been automating farming operations since the mid-1990s,” said Seth Crawford, senior vice president and general manager of the Duluth, Georgia-based company’s precision agriculture and digital division. “We’re at the stage of what we call controlled autonomy, where there’s still someone in the cockpit of the car,” he said. “The buzz is around fully autonomous operations, but where farmers are willing to pay for automation is in terms of features.”

While Deere is focused on adding full autonomy to its farm equipment, AGCO is looking at the broader supply market, Crawford said. “In the summer of 2023, we will have a set of performance-enhancing upgrades for many car brands,” he said. “Where other people say we give you autonomy with a half-million dollar tractor,” he said, pointing to the price tag of Deere’s 8R, “we have kits that allow you to do that with your existing fleet. We see a huge opportunity with the installed base, where farmers can increase their results.” They want to adopt the technology, but they don’t want to replace their entire fleet and make this massive investment.”

In 2016, Case IH, a subsidiary of London-headquartered CNH Industrial, debuted what it called an Autonomous Concept Vehicle at the Farm Progress Show. The elegant prototype tractor, which excluded the driver’s cab, hinted at the appearance of autonomy at the time. Fast forward six years to the Farm Progress Show in September, when Case IH showcased its Trident 5550 autonomous applicator.

Source link