Full season navigation and control of ultra-compact autonomous ag-bot in GPS denied environment

This thesis describes perception and an autonomous navigation system for an ultra-lightweight ground robot in agricultural fields. The system is designed for reliable navigation under cluttered canopies using only a 2-D Hokuyo UST-10LX LiDAR and an RTK GPS as the primary sensors for navigation. Its purpose is to ensure that the robot can navigate through rows of crops without damaging the plants in narrow row-based and high-leaf-cover semi-structured crop plantations, such as corn (Zea mays), sorghum (Sorghum bicolor) and soybean (Glycine max). The fundamental contributions of this work are a GPS-INS based reliable localization system for the robot and a LiDAR-based navigation algorithm capable of rejecting outlying measurements in the point-cloud due to plants in adjacent rows, low-hanging leaf cover, or weeds. Finally, this work describes a behavior-based navigation architecture that enables the system to autonomously traverse a breeding field throughout the planting season.LimitedAuthor requested closed access (OA after 2yrs) in Vireo ETD syste