Development of an Autonomous Weeding Robot for Roadside Weeds

Abstract

Roadside weeds can cause various problems, such as obstructing drivers’ views, reducing the aesthetic appeal of the surrounding areas, and serving as habitats for pests. Currently, roadside weeding is primarily performed by human labor, which is risky and time-consuming. We propose an autonomous mobile weeding robot that can travel along the shoulder of a paved road and cut weeds. Our robot is equipped with two cameras, including a forward camera for autonomous movement and a side camera for weed detection. We conducted a demonstration test of the weeding function of the developed robot outdoors. Our approach is novel in that it focuses on roadside weeding, which has received relatively little attention in previous studies on weeding robots. Our results demonstrate the feasibility of using autonomous weeding robots to address the challenges associated with roadside weeding.