Masters Theses

Date of Award

5-1994

Degree Type

Thesis

Degree Name

Master of Science

Major

Electrical Engineering

Major Professor

Paul Benjamin

Committee Members

David Koch, Herbert Neff

Abstract

This thesis examines a specialized solution to the problem of automatically locating a stationary, three-dimensional, circular docking port that is to be accessed by a robotic arm or boom without intervention by a human operator under conditions varying from high ambient light to total darkness.

Image formation, object identification, and pose determination for the port are the subject of this thesis. A special docking aid marker designed in conjunction with the recognition algorithm ensures that the proper object is identified and can be recognized when imaged over a specified range of distances and orientations under adverse lighting conditions. This special technique uses colored markers about the port that are easily isolated via a color camera.

The feasibility of the proposed design and technique for automated recognition and location is tested and demonstrated by using simulated images and standard and custom modules within an image processing software package. Reliable and repeatable results are obtained and documented where the true orientation relative to the camera is extracted from the algorithm to locate the port in the image.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS