Evolving circuits on a field programmable analog array using genetic programming
Name
62521782-MIT.pdf
Description
Full printable version
Size
3.9 MB
Format
Adobe PDF
Checksum (MD5)
65cc48b948151d5bfa98e79934d8ddc7
Author(s)
Terry, Michael A. (Michael Allen)
Advisor(s)
Una-May O'Reilly.
Date Issued
2005
Publisher
Massachusetts Institute of Technology
Abstract
This thesis describes the design and implementation of the Genetic Programming Intrinsic Circuit (GPIC) design system. Inspired by a number of recent advances in the field of Evolvable Hardware, the intended purpose of GPIC is to automate the design of analog circuits with minimal domain knowledge, computational resources, and cost using Genetic Programming with candidate solutions implemented in real hardware. This system has been constructed out of commercially available hardware and software, and the components were integrated through the development of a modular device-independent software system. The fitness evaluations of the candidate solutions of the Genetic Programming module are realized through a C interface to a National Instruments Data Acquisition Card. This Genetic Programming approach to analog circuit design decreases the fitness evaluation time of previous approaches by substituting expensive circuit simulation for real-time hardware testing. Since performing fitness evaluations in simulation is limited by the known model for a given environment, intrinsic testing provides additional benefit through the inherent incorporation of any unknown environmental conditions during tests. This feature is especially important for autonomous systems in unknown environments, and systems that must perform well in extreme environments.
Description
Thesis (M. Eng.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2005.
Includes bibliographical references (p. 57-60).
Subjects
Electrical Engineering and Computer Science.
MIT Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
M.I.T. theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. See provided URL for inquiries about permission.
Persistent DSpace Link
