Date of Award

6-1-2010

Document Type

Masters Thesis

Degree Name

M.S.

Organizational Unit

Daniel Felix Ritchie School of Engineering and Computer Science

First Advisor

Richard Voyles, Ph.D.

Second Advisor

Matthew Rutherford

Third Advisor

Mohammad Matin

Keywords

Field programmable gate arrays, Hardware, RecoNode, Robot

Abstract

Urban search and rescue (USAR) robots can benefit from small size as it facilitates movement in cramped quarters. Yet, small size limits actuator power, sensor payloads, computational capacity and battery life. We are alleviating these issues by developing the hardware and software infrastructure for high performance, heterogeneous, dynamically-reconfigurable miniature USAR robots, as well as a host of other relevant applications. In this thesis, a generic modular embedded system architecture based on the RecoNode multiprocessor is proposed, which consists of a set of hardware and software modules that can be configured to construct various types of robot systems for dynamic and unforeseen changes in the USAR environment. The benefit of this Reconfigurable Node is that, at run-time, the system can react to unexpected changes in configuration, such as nodes exhausting their batteries or the failure of sensors. These modules include a high performance microprocessor supporting complete on board processing for autonomous control, a reconfigurable hardware component, and diverse sensor and actuator interfaces. The design of all the modules in the electrical subsystem allows for the replacement of the motion control and serial communication capabilities within a dedicated FPGA logic module, which helps gain system performance by releasing the CPU from these tasks. The selection of module components and real-time scheduler and operating system (OS) are described. The portable power supply solution is also designed and tested.

Publication Statement

Copyright is held by the author. User is responsible for all copyright compliance.

Rights Holder

Kang Li

Provenance

Received from ProQuest

File Format

application/pdf

Language

en

File Size

84 p.

Discipline

Computer engineering

Download



Included in

Robotics Commons

Share

COinS