Cornell robot electrifies weed control race

Cornell University undergraduates have won the grand prize at The Farm Robotics Challenge with an autonomous robot that kills weeds using targeted electric shocks, strengthening investor interest in farm automation as growers look for alternatives to herbicides, fuel-heavy machinery and costly manual labour.

The winning project, developed by Rootline Robotics, beat 95 other teams in the 2026 competition, which drew entries from 13 countries and spanned schools, two-year colleges and four-year universities. The award was announced at Plug and Play Tech Center in Sunnyvale, California, after judges reviewed student-built systems aimed at problems ranging from weed pressure and pest detection to harvesting, irrigation and labour shortages.

The Cornell team’s machine is designed for orchards and vineyards, where weeds compete with trees and vines for water and nutrients during critical growth periods. Unlike conventional electric weeders that require an operator and can consume large amounts of energy, the Cornell prototype moves autonomously through crop rows and applies pulsed voltage microshocks directly to unwanted plants.

The system uses computer vision, machine learning and depth sensing to identify targets, then deploys a two-degree-of-freedom mechanical arm fitted with a comb-based electrode array. The array conditions the plant before treatment, allowing the robot to deliver electricity more precisely. Competition data placed the treatment at roughly 30 joules per plant, with total energy use below 0.1 kilowatt-hours even at 10,000 plants.

Rootline Robotics was led by Andrew James, an agricultural sciences student at Cornell’s College of Agriculture and Life Sciences. The team included students from agricultural science, biological engineering, software engineering, mechanical engineering and electrical and computer engineering, reflecting the interdisciplinary direction of modern farm technology. Michael Neiss, Neil Morrison, Harrison Sachs, Rohan Alapati, Nidhish Kumar and Natalia Kurz were among the students named with the project.

The prize gives the group a path towards commercial development. The $50,000 grand prize was backed by Reservoir, an agricultural technology investor and incubator, and the team is moving forward with company formation under the Rootline Robotics name. Field validation is continuing with grower partners, including Crist Bros Orchards, a 500-acre commercial apple operation in Orange County, New York.

Weed control has become one of the sharpest pain points in specialty crop farming. Orchards and vineyards often have narrow rows, uneven terrain and sensitive root zones, making broad mechanical intervention difficult. Organic growers are especially exposed because their options are limited to mowing, mulching, string trimming and other labour-intensive methods that may not keep pace with spring weed growth.

The Cornell robot addresses a market gap between manual weeding and expensive industrial-scale systems. Existing electric weeders can cost about $150,000 on average, putting them beyond the reach of many smaller growers. By focusing on low-energy pulsed treatment, autonomy and modular deployment on an established robotic platform, Rootline is attempting to cut both operating cost and labour dependence.

The project also underscores the growing role of robotics in agriculture at a time when farms face rising labour costs, ageing workforces and pressure to reduce chemical inputs. Automated weeding systems are gaining attention because they can limit herbicide use, lower fuel demand and allow growers to treat only problem plants rather than whole rows or fields.

The 2026 Farm Robotics Challenge was organised by UC ANR Innovate and the AI Institute for Next Generation Food Systems, with support from the Fresno-Merced Future of Food Initiative. The programme has become a pipeline for student-led agricultural technology, encouraging teams to work directly with farmers before designing prototypes.

Judges also recognised projects from Carnegie Mellon University, UC Davis, Reedley College, Hartnell College and student teams from other regions. The range of entries reflected the breadth of automation now moving into farm operations, from drone-assisted bloom estimation and disease mapping to autonomous sprayers and soil-moisture sensing systems.