Keywords

Computers -- Optical, Robots -- Industrial

Abstract

One of the solutions to the world's industrial and economic recovery and survivability is "Flexible Automation," wherein vision is the key to its cost-effective flexibility. Vision gives the robot a measure of environmental intelligence. The majority of present vision systems are patterned after the human eye, but this is not a requirement. It has been shown that the multi-aperture insect eye cannot image an object. The physiology of an insect eye is fundamentally different than that of the human eye in that it is not binocular by multioccular in nature. In order to study multi-aperture optics and indirectly the functioning of the insect eye an extremely high resolution robot system, resembling in some respects an insect eye was constructed. This system provided a wide field of view, and an excellent resolution with no need for optical alignment. An individual eyelet of the vision system consists of a Gradient Index (GRIN) lens, and seven optical fibers which transfer the incident light on the lens to individual detectors. Each detector or pixel, integrates the intensity over the unique region that it sees, and gives a voltage as a function of that intensity. A uniform white sphere was chosen as a test object in order to reduce five degrees of freedom (x, y, z, ╬ÿ, ╔╕) into three, namely x, y, z. It was experimentally shown that each set of voltage outputs produced by a set of detectors was unique when the object was placed at exact coordinate points, in a cube of 5X5X10 cm. The voltage readings were cataloged in a data file. Identification of the coordinates of the object was possible by a simple comparison of input voltages to the cataloged values. Also, a mathematical model for this system was developed.

Notes

If this is your thesis or dissertation, and want to learn how to access it or for more information about readership statistics, contact us at STARS@ucf.edu

Graduation Date

Fall 1983

Advisor

Walters, Roy A.

Degree

Master of Science (M.S.)

College

College of Engineering

Degree Program

Engineering

Format

PDF

Pages

79 p.

Language

English

Rights

Public Domain

Length of Campus-only Access

None

Access Status

Masters Thesis (Open Access)

Identifier

DP0014010

Included in

Engineering Commons

Share

COinS