Files in this item



application/pdfECE499-Sp2012-law.pdf (3MB)Restricted to U of Illinois
(no description provided)PDF


Title:Reaching Algorithm Implemented Using Gradient Descent
Author(s):Law, Allan
Contributor(s):Levinson, Stephen
artificial intelligence
humanoid robots
robot vision
robotic control
gradient descent algorithms
Abstract:The aim of this thesis is to provide a simple visual closed loop control method to enable a humanoid robot to reach out towards an object in its visual field in 3D space. To achieve this task, the robot utilizes stereo vision to estimate the relative depths of its end-effector and the desired gripping location of the object. Gradient descent will be applied to the output obtained from the stereo vision to enable the robot to correctly position its end-effector at a desired position. By utilizing this gradient decent method, we will be able control the robot’s end-effectors without the need to map the robot’s end-effectors in world coordinates to the spatial locations found using stereo vision.
Issue Date:2012-05
Publication Status:unpublished
Peer Reviewed:not peer reviewed
Date Available in IDEALS:2014-01-09

This item appears in the following Collection(s)

Item Statistics