Development of an Automated Weeding Machine for Precision In-Row Weed Control

Automated/robotic weeding machines are commercially available; however, their adoption has been limited because the machines 1) provide only partial weed control and follow up hand weeding is necessary, 2) have slow travel speed which results in low work rates and high operating costs and 3) lack precision and cannot remove doubles in lettuce crops. To overcome these limitations, we propose to develop an innovative high-speed, high-precision automated weeding machine. The machine will utilize an artificial intelligence (AI) based imaging system to identify weeds and a precision sprayer to spot spray targeted weeds at the 1-cm level of resolution. The 1-cm resolution spot sprayer has already been developed and successfully tested in a laboratory setting. The primary focus of the proposed project is to develop the AI-based imaging system. An additional goal is to integrate the imaging system with the precision spot sprayer and evaluate the system in the field. With promising preliminary data, we plan to seek funding to fully develop and prototype a high-precision automated weeding machine.