Files
Abstract
Phenotyping robots are being investigated to automate the tasks associated with high-throughput phenotyping (HTP) that can aid the development of high yield crops. The aim of this work is to develop an autonomous mobile field robot that can perform both HTP and soil sensing with the onboard LiDAR sensor and a manipulator. In the first part of the thesis, a simulation of the robot in a high-fidelity environment utilizing Robotic Operating System is presented to validate the use of an actuated LiDAR configuration that allows for simultaneous phenotyping and autonomous navigation. The use of this LiDAR configuration was shown to estimate plant height and volume with comparable accuracy to other LiDAR configurations while also navigating through crop rows with 0.2% error. In the second part of the thesis, a differential drive robot was designed and implemented that can autonomously navigate in the field based on Global Navigation Satellite Systems. This autonomous mobile robot system validates the simulation study by implementing an actuated 2D LiDAR for both phenotyping and navigation. Additionally, a three degree of freedom articulated robotic manipulator was designed to be re-configurable and expected to perform a variety of tasks such as soil sampling and sensing. The robotic system developed in this thesis will benefit high throughput phenotyping and precision agriculture.