A spray boom used to treat a field usually has one job: keep moving and cover ground. Verdant Robotics' SharpShooter gives it a more particular assignment. Its cameras look down at individual plants, while software decides whether a nozzle has a reason to open.
That distinction matters most in California's vegetable fields, where a pass can carry herbicide across plenty of bare soil and crop rows in search of a comparatively small number of weeds. At TopFlavor Farms, a SharpShooter system has been used to distinguish weeds from vegetable plants and direct the spray only at the weeds, according to Ag Alert's account of the machine.
A Tractor That Keeps Its Appointment
The new piece is not another camera or another nozzle. Sabanto's autonomous tractor technology is being combined with SharpShooter, so the machine carrying the sprayer can work without a driver seated in the cab. The companies describe the arrangement as a pairing of autonomous tractor operation with Verdant's Aim & Apply system, which identifies plants and weeds before applying inputs where needed.
Sensors along SharpShooter's boom image plants at 30 frames per second, feeding an onboard computer that sorts the scene quickly enough for targeted application. The hardware still has to travel through a real field—with dust, uneven ground, crop residue and the occasional plant that refuses to present itself neatly—but the aim is to make each spray decision at the nozzle rather than at the field boundary.
The companies say the system can identify weeds across 100 crops, a deliberately broad claim that reaches beyond one lettuce block or one vegetable season. For growers, the useful question is narrower: how reliably does it recognize the weeds and crop varieties in a particular field, under the light and weather conditions that California supplies for free?
Specialty Crops Are the Hard Test
California specialty crops put pressure on both halves of the arrangement. Labor shortages make repetitive tractor work harder to staff, while high-value crops leave little room for careless misses. An autonomous carrier can address the first problem; plant-level spraying is meant to address the second.
That does not make the system a substitute for agronomy. Application labels, field conditions, crop stage, weed size and the performance of the vision system still govern whether a targeted pass is useful. A machine that can recognize a weed in one crop does not automatically prove itself in a young orchard, a dense lettuce stand or a field with unusual residue.
The technology is arriving alongside a larger California push toward machine vision in row crops. A UC Cooperative Extension field day in the Salinas Valley scheduled demonstrations from Carbon Robotics, Mantis Ag Technology and Niqo Robotics, including laser weeding, lettuce thinning and targeted applications. The machines are beginning to look less like isolated curiosities and more like competing answers to the same labor equation.
For now, the partnership is best understood as an operating model rather than a finished prescription: an autonomous tractor supplies the movement, SharpShooter supplies the plant-level judgment, and the grower still supplies the field knowledge. The value will show up in the less glamorous accounting—hours covered, acres completed, chemical saved and rework avoided.