Pictured above: John Deere’s fully autonomous tractor; and below: Deere’s “See & Spray Ultimate” targeted spray technology in action.

John Deere (NYSE: DE) has achieved leadership in automated machinery powered by Artificial Intelligence (AI). The self-driving tractor pictured above was revealed in CES 2022 and has six pairs of stereo cameras. Images captured by the cameras are passed through a deep neural network that monitors the path ahead and stops the tractor if it senses an obstacle. The autonomous tractor also functions within an inch of its preset working boundary. (VentureBeat)

The image below shows, in action, Deere’s See & Spray™ Ultimate, a factory-installed system available for model year 2023 John Deere 410R, 412R, and 612R Sprayers. It enables targeted spraying of non-residual herbicide on weeds among corn, soybean, and cotton plants.

John Deere’s new carbon-fiber truss-style boom features cameras and processors to distinguish weeds from crop plants as the machine passes over the field.

The system can reduce non-residual herbicide use by more than two-thirds by only spraying weeds in corn, soybeans, and cotton.

“Data Seeds”

However, the expertise that went into the development of the above machines was decades in the making, and involved the company’s painstaking and systematic collection of valuable data through its platform, machine connectivity, and GPS.

Deere is also reaping a rich harvest from canny and forward-looking decisions in the past. Among these, a mid-nineties foray into GPS technology and its application to farm machinery such as a GPS-based steering system for tractors – “a big unlock,” in the words of Julian Sanchez, director of emerging technology at John Deere. “We’ve been building on that ever since.”

Another path-breaking breakthrough was tagging a geospatial location to every sensor on its vehicles, allowing for geospatial mapping of the field in the early 2000s. However, USB-based transfer of data from these sensors to offsite PCs was a clumsy procedure.

In 2010, Deere found a way to scrap transfer vide thumb drives, and instead commenced the use of vehicle-mounted telematics boxes that transferred data using a cellular network. Mobile and cloud innovations in the 2010s gave a boost to the possibilities of using these large volumes of data in AI applications.

In 2017, Deere added more firepower to its AI capabilities by acquiring Blue River Technologies, a machine learning outfit. “That immediately doubled or tripled the number of people working on AI,” Sanchez told VB. “That was a pivot point.”

All this is now supported by a hundreds-strong data science team, and dedicated units for the development of automation solutions, as well as digital tools.

“Our goal is by 2030, we want to have a fully-autonomous production system, meaning we want an autonomous combine and sprayer and tractor planter,” says Sanchez.

That would be a boon for farmers given today’s shortage of skilled and experienced agricultural hands, exacerbated by a tendency to move out of rural areas.

Another ‘holy grail’ AI goal for Deere is real-time measurement of soil health using cloud and 5G connectivity.

“AI already has had an impact in agriculture,” observes Sanchez. “That’s why we see that runway of opportunity there. Agriculture has all kinds of these perfect examples that are prime for AI, as opposed to broader, more generalized applications.”

Related Story: Blue River’s Massive Robots For Precision Spraying Weeds