2021/12/03 by Yayun Du, Du, Yayun, Bhrugu Mallajosyula +13 · 1 citation
Agricultural and Biological Sciences · #FOS: Computer and information sciences #Irrigation Practices and Water Management #Robotics (cs.RO) #Smart Agriculture and AI
paper · pdf · doi:10.48550/arxiv.2112.02162
openalex publication_date 2021/12/03 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Modern herbicide application in agricultural settings typically relies on\neither large scale sprayers that dispense herbicide over crops and weeds alike\nor portable sprayers that require labor intensive manual operation. The former\nmethod results in overuse of herbicide and reduction in crop yield while the\nlatter is often untenable in large scale operations. This paper presents the\nfirst fully autonomous robot for weed management for row crops capable of\ncomputer vision based navigation, weed detection, complete field coverage, and\nautomatic recharge for under \$400. The target application is autonomous\ninter-row weed control in crop fields, e.g. flax and canola, where the spacing\nbetween croplines is as small as one foot. The proposed robot is small enough\nto pass between croplines at all stages of plant growth while detecting weeds\nand spraying herbicide. A recharging system incorporates newly designed robotic\nhardware, a ramp, a robotic charging arm, and a mobile charging station. An\nintegrated vision algorithm is employed to assist with charger alignment\neffectively. Combined, they enable the robot to work continuously in the field\nwithout access to electricity. In addition, a color-based contour algorithm\ncombined with preprocessing techniques is applied for robust navigation relying\non the input from the onboard monocular camera. Incorporating such compact\nrobots into farms could help automate weed control, even during late stages of\ngrowth, and reduce herbicide use by targeting weeds with precision. The robotic\nplatform is field-tested in the flaxseed fields of North Dakota.\n