21 September 2026, British Columbia, Canada: Engineers at Simon Fraser University’s Global Institute for Agritech in Burnaby, British Columbia, Canada, have built and field-tested a laser weeding robot designed to cost a fraction of commercial machines, aiming the technology squarely at small and resource-constrained farms rather than the large mechanized operations that most precision weeding equipment targets today. The peer-reviewed study, led by researcher Woo Soo Kim with eight co-authors, was published in the journal Smart Agricultural Technology, appearing online on June 19, 2026, and in the publication’s August 2026 print volume.
The robot, named LiteWeed, pairs a low-power laser with artificial intelligence that runs directly on the machine itself rather than in the cloud, an approach known as edge computing. It uses a 4-watt blue diode laser, a much smaller and cheaper light source than the industrial carbon dioxide lasers found in commercial laser weeders, paired with a compact computer vision system called YOLOX-nano running on a Raspberry Pi 5, an inexpensive single-board computer popular in hobbyist and research electronics. When the system detects a weed, the robot stops, aligns a small two-jointed robotic arm over the target, and fires an adaptive laser dose calibrated to the plant’s species and size, a method the researchers describe as a stop-and-align control strategy.
Built for Places Big Machinery Cannot Reach
The team tested LiteWeed on three common and difficult weed species, pigweed, purslane and nutsedge, at three growth-stage sizes, in two very different settings: bare-ground field plots in Vancouver, Canada, and in Arusha, Tanzania. That deliberate choice of a second test site in East Africa was meant to demonstrate the system could function outside a controlled, resource-rich research environment. The results were consistent across both locations. The robot successfully eliminated 97 percent of targeted weeds in the Vancouver trials and 96 percent in Arusha, with most weeds killed in under 30 seconds of laser exposure, rising to a maximum of 60 seconds for the largest pigweed plants. The vision system’s detection accuracy, measured using a standard computer vision metric called mean average precision, came in at 0.32, a figure the authors treat as a solid baseline for a low-cost system rather than a state-of-the-art benchmark, alongside a processing delay of about 118 milliseconds per detection on the robot’s onboard computer.
The other headline figure in the study is cost. The researchers priced a complete LiteWeed unit at approximately 500 Canadian dollars, roughly 370 US dollars, built from off-the-shelf components: about 100 dollars for the onboard computing and camera system, 200 dollars for the wheeled locomotion platform, 50 dollars for the targeting arm, 50 dollars for the laser module itself, and 100 dollars for power and safety systems. That places it well below commercial autonomous laser weeders, which are typically sold as large, tractor-mounted implements costing tens of thousands of dollars and designed for row-crop operations at scale.
The authors are candid about the platform’s limits. LiteWeed moves at about 5 centimeters per second and its stop-and-align approach means it can realistically cover only around 0.01 to 0.03 hectares a day, depending on how dense the weeds are, making it unsuitable for broadacre field clearing. Instead, the researchers position it as a tool for early-stage, low-density spot treatment on small plots, market gardens or the edges of larger fields, where herbicide use is often hardest to justify economically or environmentally and where labor for hand-weeding is scarce or expensive.
Why It Matters Beyond the Lab
Laser weeding has drawn significant commercial investment in North America and Europe over the past several years, but nearly all of that investment has gone into large, capital-intensive machines built for row-crop agriculture in wealthy farming regions. LiteWeed’s authors argue that a genuinely low-cost, open approach is needed if laser weeding is to reach smallholder and mid-sized farms in Africa, South Asia and Latin America, where labor-intensive hand-weeding remains the norm and herbicide costs and resistance are growing concerns. The system also avoids the soil microbiome disruption associated with repeated herbicide use, since it kills weeds through localized heat damage rather than chemical action.
Global Agriculture is an independent international media platform covering agri-business, policy, technology, and sustainability. For editorial collaborations, thought leadership, and strategic communications, write to pr@global-agriculture.com
