John Deere announced on November 19, 2026 that it had acquired Orchard Robotics, which operates as Orchard AI, adding vehicle-mounted cameras and software intended to turn tree-by-tree crop measurements into recommendations that guide fieldwork for specialty-crop growers.
The San Francisco company, founded in 2022, develops crop-intelligence products for growers of high-value specialty crops. The acquisition brings that plant-level data capability into Deere’s precision-agriculture portfolio. Tim Marquis, Deere’s director of farm systems innovation, described the objective as connecting crop intelligence with equipment that can execute fieldwork precisely throughout the growing season.
How Orchard AI Builds Plant-Level Maps
Orchard AI’s cameras travel through orchards and vineyards on farm vehicles, collecting data on individual trees, vines and plants across the season. Deere says the system analyzes hundreds of images per tree on each pass, producing yield estimates, fruit counts, fruit size and color data, and canopy measurements. Its FruitScope software makes that information available for farm-management decisions.
Orchard AI’s product documentation says FruitScope Vision cameras can mount on tractors, all-terrain vehicles and other farm vehicles already in use, operate at 1 to 12 mph, and return scan results the next morning. The hardware is designed for operations ranging from 100 acres to more than 10,000 acres. FruitScope OS provides per-tree maps covering fruit count, size, color, disease and canopy conditions, with harvest-planning and crop-load-management tools on phones and computers.
Orchard AI describes Canary AI as software that reasons over farm data to answer questions and recommend what to do next. Its listed crops include apples, wine and table grapes, blueberries, cherries, almonds, pistachios and citrus. The company offers a per-acre, per-year subscription covering the camera hardware, unlimited scans, FruitScope OS, Canary AI, training and support, with no separate hardware purchase. That model gives Deere a subscription software offering built around repeated crop scans, in addition to the sensing equipment itself.
From Crop Maps to Equipment Instructions
Deere framed the technology around the cost of poor timing: a missed spray application or mistimed harvest can affect fruit quality, yield and a season’s profitability, while weather, labor constraints and changing crop conditions narrow decision windows. The workflows identified for the combined technology include pruning, thinning, applying crop treatments where needed and determining which blocks are ready for harvest.
One proposed workflow uses higher bloom counts to generate a prescription for where to spray. The recommendation could be delivered to a mobile device for in-field application or to an in-cab G5 display. Deere presented these examples as possibilities, not as a commitment to specific product availability. The proposed connection would move crop data beyond a dashboard and into instructions for work in specific blocks.
Deere’s Plans for Orchard AI
Tyler Klick of Redwood Empire Vineyard Management, an early user of Orchard AI’s technology, said the scans provided useful precision-farming insights into vineyard conditions, and that CEO Charlie Wu’s team had listened to customer needs and put them into action.
Justin Rose, president of Deere’s Worldwide Agriculture & Turf Division for Small Ag & Turf, said the company wants to use its scale to help Orchard AI reach more growers. “A big part of what makes Orchard AI successful is its speed of innovation and the close relationships it has built directly with customers,” he said, adding that Deere wants to preserve those strengths.
For Deere, the strategic value is the potential to make plant-level intelligence an input to fieldwork rather than another standalone dashboard. Turning Orchard AI’s maps into routine equipment workflows, while retaining the grower relationships that shaped the product, is the central integration question raised by the acquisition.
Credit: Source link

























