Intelligent control and computer engineering

Lecture Notes in Electrical Engineering

Volume 70

For other titles published in this series, go to

www.springer.com/series/7818

Sio-Iong Ao Oscar Castillo Xu Huang

Editors

Intelligent

Control

and Computer

Engineering

Editors

Sio-Iong Ao

International Association of Engineers

Hung To Road 37-39

Hong Kong, Unit 1, 1/F

People’s Republic of China

[email protected]

Oscar Castillo

Tijuana Institute of Technology

Computer Science

Tijuana

Mexico

[email protected]

Xu Huang

University of Canberra

Fac. Information Science & Engineering

Canberra, Aust. Capital Terr.

Australia

[email protected]

ISSN 1876-1100

ISBN 978-94-007-0285-1

e-ISSN 1876-1119

e-ISBN 978-94-007-0286-8

DOI 10.1007/978-94-007-0286-8

Springer Dordrecht Heidelberg London New York

© Springer Science+Business Media B.V. 2011

No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by

any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written

permission from the Publisher, with the exception of any material supplied specifically for the purpose

of being entered and executed on a computer system, for exclusive use by the purchaser of the work.

Cover design: VTEX, Vilnius

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

Preface

A large international conference on Advances in Intelligent Control and Computer

Engineering was held in Hong Kong, March 17–19, 2010, under the auspices of

the International MultiConference of Engineers and Computer Scientists (IMECS

2010). The IMECS is organized by the International Association of Engineers

(IAENG). IAENG is a non-profit international association for the engineers and

the computer scientists, which was founded in 1968 and has been undergoing rapid

expansions in recent years. The IMECS conferences have served as excellent venues

for the engineering community to meet with each other and to exchange ideas.

Moreover, IMECS continues to strike a balance between theoretical and application

development. The conference committees have been formed with over two hundred

and fifty members who are mainly research center heads, deans, department heads

(chairs), professors, and research scientists from over thirty countries. The confer￾ence participants are also truly international with a high level of representation from

many countries. The responses for the conference have been excellent. In 2010,

we received more than one thousand manuscripts, and after a thorough peer review

process 56.26% of the papers were accepted (http://www.iaeng.org/IMECS2010).

This volume contains 25 revised and extended research articles written by promi￾nent researchers participating in the conference. Topics covered include artificial

intelligence, control engineering, decision supporting systems, automated planning,

automation systems, systems identification, modelling and simulation, communica￾tion systems, signal processing, and industrial applications. The book offers the state

of the art of tremendous advances in intelligent control and computer engineering

and also serves as an excellent reference text for researchers and graduate students,

working on intelligent control and computer engineering.

Sio-Iong Ao

Oscar Castillo

Xu Huang

v

Contents

Intelligent Control of Reduced-Order Closed Quantum Computation

Systems Using Neural Estimation and LMI Transformation ..... 1

Anas N. Al-Rabadi

Optimal Guidance and Control for Space Robot Operation . . . . . . . . 15

Takuro Kobayashi and Shinichi Tsuda

The Application of Genetic Algorithms in Designing Fuzzy Logic

Controllers for Plastic Extruders . . . . . . . . . . . . . . . . . . . . 25

Ismail Yusuf, Nur Iksan, and Nanna Suryana Herman

Automatic Weight Selection and Fixed-Structure Cascade Controller for

a Quadratic Boost Converter . . . . . . . . . . . . . . . . . . . . . . 39

Somyot Kaitwanidvilai and Pitsanu Srithongchai

Availability Studies and Solutions for Wheeled Mobile Robots . . . . . . 47

Adrian Korodi and Toma L. Dragomir

The Use of Higher-Order Spectrum for Fault Quantification of

Industrial Electric Motors . . . . . . . . . . . . . . . . . . . . . . . . 59

Juggrapong Treetrong

A Newly Cooperative PSO – Multiple Particle Swarm Optimizers with

Diversive Curiosity, MPSOα/DC . . . . . . . . . . . . . . . . . . . . 69

Hong Zhang

Predicting the Toxicity of Chemical Compounds Using GPTIPS: A Free

Genetic Programming Toolbox for MATLAB . . . . . . . . . . . . . 83

Dominic P. Searson, David E. Leahy, and Mark J. Willis

Diversity-Driven Self-adaptation in Evolutionary Algorithms . . . . . . . 95

Fanchao Zeng, James Decraene, Malcolm Yoke Hean Low, Suiping

Zhou, and Wentong Cai

vii

viii Contents

A New Rearrangement Plan for Freight Cars in a Train . . . . . . . . . . 107

Yoichi Hirashima

Coevolving Negotiation Strategies for P-S-Optimizing Agents . . . . . . . 119

Jeonghwan Gwak and Kwang Mong Sim

Policy Gradient Approach for Learning of Soccer Player Agents . . . . . 137

Harukazu Igarashi, Hitoshi Fukuoka, and Seiji Ishihara

Genetic Algorithm for Forming Buyer Coalition with Bundles of Items

in E-Marketplaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

Anon Sukstrienwong

Inside Virtual CIM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163

Ning Zhou, Sev Naglingam, Ke Xing, and Grier Lin

Supreme Court Sentences Retrieval Using Thai Law Ontology . . . . . . 177

Tanapon Tantisripreecha and Nuanwan Soonthornphisaj

Genetic Algorithm Based Model for Effective Document Retrieval . . . . 191

Hazra Imran and Aditi Sharan

An Agent-Based Cloud Service Discovery System that Consults a Cloud

Ontology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203

Taekgyeong Han and Kwang Mong Sim

Possible Applications of Navigation Tools in Tilings of Hyperbolic Spaces 217

Maurice Margenstern

Graph Pattern Matching with Expressive Outerplanar Graph Patterns . 231

Hitoshi Yamasaki, Takashi Yamada, and Takayoshi Shoudai

Setvectors – An Efficient Method to Predict Cache Contention . . . . . . 245

Michael Zwick

New Material Model for Describing Large Deformation of Pressure

Sensitive Adhesive . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259

Kazuhisa Maeda, Shigenobu Okazawa, and Koji Nishiguchi

QoS Provisioning in EPON Systems with Interleaved Two Phase

Polling-Based DBA . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271

I-Shyan Hwang, Jhong-Yue Lee, and Zen-Der Shyu

The Game of n-Player Shove and Its Complexity . . . . . . . . . . . . . . 285

Alessandro Cincotti

Contents ix

Modeling the Vestibular Nucleus . . . . . . . . . . . . . . . . . . . . . . . 293

Alexandru Codrean, Adrian Korodi, Toma-Leonida Dragomir, and Vlad

Ceregan

SPECT Lung Delineation . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

Alex Wang and Hong Yan

Intelligent Control of Reduced-Order Closed

Quantum Computation Systems Using Neural

Estimation and LMI Transformation

Anas N. Al-Rabadi

Abstract A new method of intelligent control for closed quantum computation

time-independent systems is introduced. The introduced method uses recurrent su￾pervised neural computing to identify certain parameters of the transformed system

matrix [A˜ ]. Linear matrix inequality (LMI) is then used to determine the permuta￾tion matrix [P] so that a complete system transformation {[B˜ ],[C˜ ],[D˜ ]} is achieved.

The transformed model is then reduced using singular perturbation and state feed￾back control is implemented to enhance system performance. In quantum computa￾tion and mechanics, a closed system is an isolated system that can’t exchange energy

or matter with its environment and doesn’t interact with other quantum systems. In

contrast to an open quantum system, a closed quantum system obeys the unitary

evolution and thus is information lossless that implies state reversibility. The exper￾imental simulations show that the new hierarchical control simplifies the model of

the quantum computing system and thus uses a simpler controller that produces the

desired performance enhancement and system response.

Keywords Linear matrix inequality · Model reduction · Quantum computation ·

Recurrent supervised neural computing · State feedback control system

1 Introduction

Due to the fact that current dense hardware implementations are heading towards

the critical atomic threshold, quantum computing will rapidly occupy an increas￾ingly important position in building nano-size, super-fast, and ultra-low power con￾suming systems [1–3, 6, 8, 12]. Other motivations for implementing circuits and

systems using quantum computing would include items such as: (1) power where

A.N. Al-Rabadi ()

The University of Jordan, Faculty of Engineering & Technology, Computer Engineering

Department, Amman, Jordan 11942

e-mail: [email protected]

S.-I. Ao et al. (eds.), Intelligent Control and Computer Engineering,

Lecture Notes in Electrical Engineering 70,

DOI 10.1007/978-94-007-0286-8_1, © Springer Science+Business Media B.V. 2011

1

2 A.N. Al-Rabadi

State Feedback Control System

Model Reduction

System Transformation: {[B˜ ],[C˜ ],[D˜ ]}

LMI-Based Permutation Matrix: [P]

Neural-Based State Transformation: [A˜ ]

Time-Independent Quantum Computing System: {[A],[B],[C],[D]}

Fig. 1 The introduced control methodology utilized for closed quantum computing systems

the internal computations in quantum computing systems consume no power and

only power is consumed when reading and writing operations are performed [1, 6,

8, 12]; (2) size where, at the atomic dimension, quantum mechanical effects have to

be accounted for; and (3) speed where if the properties of superposition and entan￾glement of quantum mechanics can be usefully employed in the design of circuits

and systems, significant computational speed enhancements can be expected [1, 6,

12]. Figure 1 illustrates the layer layout of the introduced closed-system quantum

computing control methodology.

2 Fundamentals

This section presents important background on quantum computing systems, super￾vised neural networks, linear matrix inequality, and model order reduction that will

be used later in Sects. 3, 4 and 5.

2.1 Quantum Computation

Quantum computing is an efficient method of computation that uses the dynamic

process which is governed by the Schrödinger equation [1, 6, 12]. The one￾dimensional time-dependent Schrödinger equation (TDSE) is as follows [1, 5, 6,

12]:

−(h/2π)2

2m

∂2|ψ

∂x2 + V |ψ = i(h/2π)

∂|ψ

∂t (1)

or H|ψ = i(h/2π)

∂|ψ

∂t (2)

where h is Planck constant (6.626 · 10−34 J·s = 4.136 · 10−15 eV ·s), V (x,t) is the

applied potential, m is the particle mass, i is the imaginary number, |ψ(x,t) is the

quantum state, H is the Hamiltonian operator where H = −[(h/2π)2/2 m]∇2 +V ,

and ∇2 is the Laplacian operator.

Intelligent Control of Reduced-Order Closed Quantum Computation Systems 3

A general solution to the TDSE is the expansion of a stationary (i.e., time￾independent for spatial) basis functions (i.e., eigen states) Ue(r) using time￾dependent (i.e., temporal) expansion coefficients ce(t) as follows:

(r,t) =n

e=0

ce(t)ue(r)

The expansion coefficients ce(t) are a scaled complex exponentials as follows:

ce(t) = kee

−i Ee (h/2π) t

where Ee are the energy levels. In quantum computing, the time-independent

Schrödinger equation (TISE) is normally used [1, 12]:

∇2ψ = 2m

(h/2π)2 (V − E)ψ (3)

where the solution |ψ is an expansion over orthogonal basis states |φi defined in

a linear complex vector space called Hilbert space H as:

|ψ =

i

ci|φi (4)

where the coefficients ci are called probability amplitudes and |ci|

2 is the probability

that the quantum state |ψ will collapse into the (eigen) state |φi. The probability

is equal to the inner product |φi|ψ|2, with the unitary condition |ci|

2 = 1. In

quantum computing, a linear and unitary operator  is used to transform an input

vector of quantum bits (qubits) into an output vector of qubits. In the two-valued

quantum computing, the qubit is a vector of bits which is defined as follows [1, 12]:

qubit0 ≡ |0 = 

1

0



, qubit1 ≡ |1 = 

0

1



(5)

A two-valued quantum state |ψ is a superposition of quantum basis states |φi.

Thus, for the orthonormal computational basis states {|0,|1}, one has the following

quantum state:

|ψ = α|0 + β|1 (6)

where αα∗ = |α|

2 = p0 ≡ the probability of having state |ψ in state |0,ββ∗ =

|β|

2 = p1 ≡ the probability of having state |ψ in state |1, and |α|

2 +|β|

2 = 1. The

calculation in quantum computing for multiple systems follows the tensor product

(⊗). For example, given the quantum states |ψ1 and |ψ2, one has:

|ψ1ψ2=|ψ1⊗|ψ2

= 

α1|0 + β1|1



⊗ 

α2|0 + β2|1



= α1α2|00 + α1β2|01 + β1α2|10 + β1β2|11 (7)

A physical system (e.g., the hydrogen atom) that is described by the following

equation:

|ψ = c1|Spinup + c2|Spindown (8)

4 A.N. Al-Rabadi

can be used to physically implement a two-valued quantum computing. Another

common alternative form of Eq. 8 is as follows:

|ψ = c1









+

1

2

+ c2







 − 1

2

(9)

Many-valued quantum computing can also be performed. For the three-valued

case, the qubit becomes a 3D vector quantum discrete digit (qudit), and in general,

for an m-valued quantum computing the qudit is of dimension “many” [1, 12]. For

example, one has for the 3-state case, the following qudits:

qudit0 ≡ |0 =

1

0

0

, qudit1 ≡ |1 =

0

1

0

, qudit2 ≡ |2 =

0

0

1

(10)

A three-valued quantum state is a superposition of three quantum orthonormal basis

states (vectors). Thus, for the orthonormal computational basis states {|0,|1,|2},

one has the following quantum state:

|ψ = α|0 + β|1 + γ |2

where αα∗ = |α|

2 = p0 ≡ the probability of having state |ψ in state |0,ββ∗ =

|β|

2 = p1 ≡ the probability of having state |ψ in state |1,γγ ∗ = |γ |

2 = p2 ≡ the

probability of having state |ψ in state |2, and |α|

2 + |β|

2 + |γ |

2 = 1.

The calculation in quantum computing for m-valued multiple systems follow

the tensor product in a manner similar to the one demonstrated for the higher￾dimensional qubit in the two-valued quantum computing. Several quantum comput￾ing systems were used to implement quantum gates from which complete quantum

circuits and systems were constructed [1, 6, 12], where several of the two-valued

and m-valued quantum circuit implementations use the two-valued and m-valued

quantum Swap-based and Not-based gates [1, 12].

In general, for an m-valued logic, a quantum state is a superposition of m quan￾tum orthonormal basis states (i.e., vectors). Thus, for the orthonormal computational

basis states {|0,|1,...,|m − 1}, one has the quantum state:

|ψ =

m

−1

k=0

ck|qk (11)

where m−1

k=0 ckc∗

k = m−1

k=0 |ck|

2 = 1. The calculation in quantum computing for

m-valued multiple systems is done similar to the case for the two-valued system.

In quantum mechanical systems, a closed system is an isolated system that

doesn’t exchange energy or matter with its environment (i.e., doesn’t dissipate

power) and doesn’t interact with other quantum systems. While an open quantum

system interacts with its environment and thus dissipates power which results in a

non-unitary evolution producing information loss, a closed quantum system doesn’t

exchange energy or matter with its environment and therefore doesn’t dissipate

power which results in a unitary evolution (i.e., unitary matrix) and thus it is in￾formation lossless.