A learning automaton approach to trajectory learning and control system design using dynamic recurrent neural networks

Persistent Link:
http://hdl.handle.net/10150/291987
Title:
A learning automaton approach to trajectory learning and control system design using dynamic recurrent neural networks
Author:
Condarcure, Thomas A., 1952-
Issue Date:
1993
Publisher:
The University of Arizona.
Rights:
Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
Abstract:
This thesis presents a method for the training of dynamic, recurrent neural networks to generate continuous-time trajectories. In the past, most methods for this type of training were based on gradient descent methods and were deterministic. The method presented here is stochastic in nature. The problem of local minima is addressed by adding the enhancement of incremental learning to the learning automaton; i.e., small learning goals are used to train the neural network from its initialized state to its final parameters for the desired response. The method is applied to the learning of a benchmark continuous-time trajectory--the circle. Then the learning automaton approach is applied to stabilization and tracking problems for linear and nonlinear plant models, using either state or output feedback as needed.
Type:
text; Thesis-Reproduction (electronic)
Keywords:
Engineering, Electronics and Electrical.; Artificial Intelligence.; Computer Science.
Degree Name:
M.S.
Degree Level:
masters
Degree Program:
Graduate College; Electrical and Computer Engineering
Degree Grantor:
University of Arizona
Advisor:
Sundareshan, Malur K.

Full metadata record

DC FieldValue Language
dc.language.isoen_USen_US
dc.titleA learning automaton approach to trajectory learning and control system design using dynamic recurrent neural networksen_US
dc.creatorCondarcure, Thomas A., 1952-en_US
dc.contributor.authorCondarcure, Thomas A., 1952-en_US
dc.date.issued1993en_US
dc.publisherThe University of Arizona.en_US
dc.rightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.en_US
dc.description.abstractThis thesis presents a method for the training of dynamic, recurrent neural networks to generate continuous-time trajectories. In the past, most methods for this type of training were based on gradient descent methods and were deterministic. The method presented here is stochastic in nature. The problem of local minima is addressed by adding the enhancement of incremental learning to the learning automaton; i.e., small learning goals are used to train the neural network from its initialized state to its final parameters for the desired response. The method is applied to the learning of a benchmark continuous-time trajectory--the circle. Then the learning automaton approach is applied to stabilization and tracking problems for linear and nonlinear plant models, using either state or output feedback as needed.en_US
dc.typetexten_US
dc.typeThesis-Reproduction (electronic)en_US
dc.subjectEngineering, Electronics and Electrical.en_US
dc.subjectArtificial Intelligence.en_US
dc.subjectComputer Science.en_US
thesis.degree.nameM.S.en_US
thesis.degree.levelmastersen_US
thesis.degree.disciplineGraduate Collegeen_US
thesis.degree.disciplineElectrical and Computer Engineeringen_US
thesis.degree.grantorUniversity of Arizonaen_US
dc.contributor.advisorSundareshan, Malur K.en_US
dc.identifier.proquest1356820en_US
dc.identifier.bibrecord.b314800449en_US
All Items in UA Campus Repository are protected by copyright, with all rights reserved, unless otherwise indicated.