Files in this item



application/pdf3337809.pdf (2MB)Restricted to U of Illinois
(no description provided)PDF


Title:Plant Specific Direct Chemical Application Field Robot
Author(s):Jeon, Hong Young
Doctoral Committee Chair(s):Lei F. Tian
Department / Program:Agricultural Engineering
Discipline:Agricultural Engineering
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):Engineering, Robotics
Abstract:This study addressed the issues in the conventional chemical application such as chemical drift and application inaccuracy via developing a plant specific direct application field robot. The robot was designed to autonomously detect plants, to identify weeds and to control weeds in the field. The developed robot was equipped with a stereovision with processing algorithms to detect and identify individual weeds in the field image and the custom designed end effector with a robotic arm to carry out plant specific direct application. The developed vision system with the algorithms was dealt with unpredictable outdoor illuminations to sense plants in the field via a series of image processing. The developed algorithm processed field images with the processing error less than 3 % in terms of identifying individual plants, and the algorithm identified 92.5 % and 95.1 % of weeds in the image via trained artificial neural network. The direct application end effector with the arm executed the direct application via cutting weed stem and wiping chemical on the weed surface. The indoor experiment resulted that 90.9 % of the weed treated by the end effector had herbicide symptoms after 6 day from the application. The plant specific direct chemical application field robot was developed by utilizing the vision system and the end effector with the arm to a field robot platform. The potential of developed robot was examined by testing the robot in the field. While testing in the field, the robot autonomously controlled 63.6 % of weeds, and required 12.923 seconds to control a weed from detection to application inspection for reapplication via machine vision.
Issue Date:2008
Description:128 p.
Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 2008.
Other Identifier(s):(MiAaPQ)AAI3337809
Date Available in IDEALS:2015-09-28
Date Deposited:2008

This item appears in the following Collection(s)

Item Statistics