Graduation Year

2006

Document Type

Thesis

Degree

M.S.Cp.E.

Degree Granting Department

Computer Science and Engineering

Major Professor

Kimon Valavanis, Ph.D.

Committee Member

Steven Wilkerson, Ph.D.

Committee Member

Miguel Labrador, Ph.D.

Keywords

robotics, modular, UAV, UGV, VTOL

Abstract

This work describes the design and implementation of a generic lightweight onboard processing system for miniature Unmanned Vehicles (UVs) that is computationally powerful and highly adaptable. First, several classical approaches to giant scale and full size UV onboard processing systems are described along with their corresponding limitations. Second, a detailed study is presented that describes the key characteristics of an onboard system along with associated limitations. Next, an implementation of a generic onboard system capable of vision processing and servo based control is presented along with detailed hardware specifications and implementation software. Last, experimental data, both laboratory and field, are presented to show validation of the onboard processing system design, functionality, and key characteristics presented.

Two primary contributions are made in this work. i) Identification of key characteristics of an onboard system allows for a high level validation of the hardware of an onboard system along with a design template for a reconfigurable, platform independent, processing system for UVs. ii) Detailed design and implementation of an adaptable onboard processing system that is both computationally powerful and easily adapted.

This system is validated by showing satisfiability of the design characteristics necessary for an adaptable onboard system along with fully operational field test and their corresponding results.

Share

COinS