For Agriculture, the Artificial Intelligence Test Has Arrived
May 6, 2024 By its nature, technology is a rapidly moving category. Each few months it seems, some new kind of improvement or technological innovation is introduced into the world. And each gets talked about quite a bit, initially. Then, something new comes along and it inevitably becomes the new “it” item. And so on.
However, one recent technological innovation promises to eclipse all overs in terms of market buzz – artificial intelligence (AI). As described by Mark Kuehn, OEM Sales Manager for North America at Trimble: “To put it as simply as possible, AI allows computer systems to complete tasks that are normally performed by humans.”
Given this definition, AI could mean myriad operations — from performing cognitive tasks such as data analytics and forecasting to physical tasks such as spraying weeds or picking produce in crop fields.
Since these items — programs that could “learn and adapt” without human intervention as they manage their particular field of endeavor — burst onto the scene a few years ago, plenty parts of society have raced to adapt them to their everyday ways of doing business.
And the world of agriculture has not been immune to this desire. Indeed, according to some industry insiders, agriculture has already been utilizing these kinds of systems in one form or another for a few decades now. The difference in the 2024 growing season is how they are now being used.
“I’ve been in this market for 26 years, and AI has been around in some form for all of those 26 years,” says Seth Crawford, Senior Vice President and General Manager PTx for AGCO Corp. PTx, is a recently launched AGCO brand representing its precision ag portfolio. “What’s changed is now we have a connected fleet out there. We have smart technology in every field collecting all these data points. We are applying dramatically improved hardware like cameras into the mix. The processing power and the video capabilities are all much better today.”
Of course, adds Crawford, improvements in the speed and accuracy of today’s AI systems is the primary advantage being offered to ag retailers, grower-customers, and technology suppliers/manufacturers as this technology expands its market presence.
“The major advantage is it helps us do better work faster” he says. “It’s a productivity tool. We can leverage AI for the forecasting of our business, of our individual products, and at our dealer level. It helps us plan for the demand that is expected.
“Our software engineers are leveraging AI as they write their software programs,” continues Crawford. “For some, this increases their productivity by 40%. On our products, we are leveraging it on the quality side. We can now identify part failures earlier. We are cataloging weeds – thousands of images are being identified and processed very quickly. This allows the users to make real-time decisions as they go along, on-the-fly so to speak. It brings all the data together quickly so that we can take it all to the growers without the gap that was normally the case. We can provide the answers needed in their fields in a fraction of the time it used to take.”
AI Systems Already in the Market
Trimble has now combined Bilberry’s AI system with its WeedSeeker2 to provide for “green-on-green” identification.
At present, there are several AI-oriented systems already making their market debuts. For example, at the 2024 National Farm Machinery Show (NFMS), held in mid-February in Louisville, KY, Trimble featured the Bilberry Smart Spraying system. Part of the company’s late 2022 acquisition of French AI manufacturer Bilberry, this system is now being combined with Trimble’s existing WeedSeeker2 product.
“As the green-on-green complement to Trimble’s green-on-brown solution, WeedSeeker2, a farmer can now reduce herbicides and spray by up to 90%,” said Cory Buchs, Senior Product Director, Agricultural Software at Trimble, speaking at NFMS 2024. He added that the company will continue to test the Bilberry system in new markets during the year, including in North and Latin America.
One month later in mid-March, Bayer announced the pilot of an expert GenAI system to benefit farmers and up-level agronomists in their daily work. According to Bayer, GenAI was developed in collaboration with Microsoft Corp. as a technology partner and Ernst & Young as an industry partner. Bayer said it has been using proprietary agronomic data to train a large language model with years of internal data, insights from thousands of trials within its vast testing network, and centuries of aggregated experience from Bayer agronomists around the world. The result is an expert system that quickly and accurately answers questions related to agronomy, farm management, and Bayer agricultural products. Instead of a time-consuming process, the intuitive system responds to natural language and can generate expert information within seconds.
“Our unique GenAI system has the potential to serve agronomists and benefit farmers all over the world, further advancing AI as an indispensable technology for agriculture,” said Amanda McClerren, CIO and Head of Digital Transformation & Information Technology for Bayer’s Crop Science Division, in a press release announcing the system’s introduction. “We’ll continue to use traditional AI to develop better products, and we’re also committed to harnessing new GenAI technology in a thoughtful way that augments and supports knowledgeable experts across the industry, bringing value to farmers and those who serve them.”
Introduced in 2022, the See & Spray Ultimate system from John Deere uses AI to help users identify weeds vs. crops when applying herbicides.
Of course, one of the equipment companies most involved in AI has been John Deere. In fact, in March 2022, the company formally introduced its See & Spray Ultimate to the marketplace. See & Spray Ultimate’s targeted spray technology was designed by Blue River Technology, a wholly-owned subsidiary of John Deere. In operation, the system uses cameras and processors mounted on John Deere’s carbon-fiber truss-style boom, combining computer vision and machine learning to detect weeds from crop plants. There is one camera mounted every one meter across the width of the boom, which equates to 36 cameras on a 120-foot boom scanning more than 2,100 square feet at once. The system was available for model year 2023 John Deere 410R, 412R, and 612R Sprayers.
“We’ve been on a technology journey with our customers since broadcast spraying,” said Franklin Peitz, Marketing Manager for John Deere, announcing the system’s debut. “While operating at up to 12 mph, See & Spray Ultimate’s targeted spray technology can help farmers reduce their non-residual herbicide use by more than two thirds and maintain a hit rate comparable to traditional spraying. For farmers, this decreased herbicide use can significantly lower herbicide costs, reduce tendering stops, and helps them cover more acres per day.”
In 2025, John Deere will offer its AI-powered See & Spray Premium on Hagie STS sprayers.
And John Deere continues to expand See & Spray’s market presence. At the 2024 Commodity Classic show in late February, the company announced that See & Spray Premium would be available on model-year 2025 Hagie STS sprayers.
“The availability of See & Spray Premium on Hagie STS sprayers means that more farmers and more custom applicators will have access to one of the most powerful tools available for maximizing cost-efficiency and in-field productivity,” says Tim Deinert, Market Manager for Application Equipment. “Used in conjunction with a good residual herbicide, the ability to selectively spray only targeted weeds – without spraying an entire field – can potentially reduce postemergence herbicide use by more than 50%.”
He adds that using See & Spray Premium could also help users become smarter. “For growers, AI might not be about being more tech-oriented,” says Deinert. “They just want to be smarter about how they manage every acre. See & Spray Premium does this by allowing them to be smarter about every acre they apply. The systems using 120 cameras scanning in front of the boom and six vision processing units that take those pictures and tells the system: ‘There’s a weed out here, spray now.’ The AI allows this system to get things down to a plant-by-plant basis.”
A Question of Autonomy
In some corners of agriculture, there is a debate regarding the difference between AI and an already existing technology for ag equipment – autonomous vehicles. However, Deinert definitely sees a difference.
“I see one all right,” he says. “AI is a system that can get smarter and adapt to the things it is processing. It learns, it gets smarter about what it’s seeing. Autonomy is really focused on simple jobs in the field and automating them. It’s two different camps.”
Now that there are several AI-based systems becoming available to the agricultural marketplace, many market observers are wondering just how quickly such systems will catch on with users. According to some statistics looking at this marketplace in depth, the value of AI in agriculture reached $1.5 billion globally during calendar year 2023. It is projected to grow modestly over the next few years – hitting $1.8 billion in 2024, $2.4 billion in 2025, and $3 billion in 2026. However, when combined with all other similar high-tech innovations – such as drones and in-field robots – the entire category for agriculture could easily top $208 billion globally by the end of the decade.
According to Mark Burns, Marketing Manager for Application Equipment for Case IH, one of the market motivators for AI growth over the next few years could tie back to one of the industry’s biggest challenges over the past few growing seasons – finding labor to operate machines/work in crop fields. Indeed, according to statistics from the 2023 CropLife 100 survey of the nation’s top ag retailers, 34% cited finding/keeping employees as their No. 1 market concern going into the 2024 growing season. Furthermore, historically across the 40-year history of the CropLife 100, labor has consistently ranked as the industry’s top challenge year in and year out.
Case IH has already attempted to address this concern as far back as 2022, when the company partnered with technology company Raven Industries to introduce one of the industry’s first autonomous spreaders, the Trident 5550, at that year’s Farm Progress show. However, adopting some form of AI system into its units would be the next step for Case IH and its industry partners, says Burns.
“Labor gets to be more of a challenge all the time,” he says. “As we get into this next year or so, you are going to see introductions through our partnership with Raven. We will integrate those systems into our systems. If we, using AI, can make the operations of a sprayer or spreader easier, so be it.”
Still, the question of how fast AI-oriented will become more commonplace remains. As Michael Gomes, Vice President of Business Development for Agriculture at Topcon, it could take some time for these systems to “catch on.”
“Completely replacing a human is pretty hard to do, today,” says Gomes. “Nobody likes change, but most people do like progress. The difference is when change has a purpose.”
Gary Esselink, Precision Agriculture Specialist at Raven Industries, agrees. “AI won’t be commonplace in 2025,” predicts Esselink. “It will take longer than that. People will need to get more comfortable with these technologies – and that they work properly – before they adopt them in big numbers. I think it will take five years or so. But by the time 2030 comes around, I believe AI system will really be coming into their own in terms of market acceptance.”
Subscribe Today For
