» BOOKSHELF
From Kits to Chaos
T
he book reviews in this issue cover diverse topics. We
begin with a review of the book Stiquito Controlled!,
which is also a mechatronics kit. Next, we review the recent
book by Laub on matrix theory for scientists and engineers.
The Colonius and Pritchard tome on mathematical systems
theory is reviewed. Finally, we have a review of the book by
Gonzalez-Miranda on control of chaos.
These books are a useful reminder of the breadth of top-
ics of interest to IEEE Control Systems Magazine readers. I
am continually looking for new books to have reviewed. If
you are the author of a recently published or soon-to-be-
released book relating to any aspect of systems and control,
please contact me. I look forward to hearing from you.
Scott Ploen
Guidance and Control Analysis Group
Jet Propulsion Laboratory
California Institute of Technology
4800 Oak Grove Drive
Pasadena, CA, 91109 USA
Phone: +1 818 354 0581
Fax: +1 818 393 4440
E-mail: Scott.R.Ploen@jpl.nasa.gov
Stiquito Controlled!
Making a Truly
Autonomous Robot
By JAMES M. CONRAD
STIQUITO:
A SMALL HEXAPOD ROBOT
In 1992, Dr. Jonathan W. Mills
envisioned a small, low-cost
hexapod robot named Stiquito
Wiley, New York, 2005,
US$59.95 for studying colony robotics and
ISBN 0-472-48882-8 emergent systems. The six legs of
this tiny robot were actuated by
nitinol shape-memory alloy wires [1]. Since then, variations
on the design with two-degree-of-freedom (DOF) legs and
different power sources have been constructed by various
researchers. Dr. James Conrad has introduced a variation of
Stiquito that can be controlled and programmed by a com-
puter (see Figure 1). The assembled robot is 85-mm long,
70-mm wide, and 25-mm high and weighs 10 g without bat-
teries. Stiquito can carry up to 50 g while walking at a speed
of 3–10 cm/min.
92 IEEE CONTROL SYSTEMS MAGAZINE » APRIL 2006
FIGURE 1 Stiquito robot. Controlled Stiquito robots are well suited for
control systems and mechatronics education.
The most important features of this variation of Stiquito
are its low cost and controllability. For the price of a single-
wheeled mobile robot, it is now possible to build tens of walk-
ing Stiquito robots and program them to study computational
sensors [2], subsumption architectures [3], neural gait control
[4], behavior of social insects [5], and machine vision [6]. Con-
trolled Stiquito, which is well suited for control systems and
mechatronics education, is the subject of the book being
reviewed here.
This book is the successor of two earlier books, namely, [7]
and [8], published in 1997 and 1999, respectively. Both [7] and
[8] come with a robot kit and provide instructions for assem-
bling the kit. The former examines advanced research on the
role of various sensors in the development of a mobile robot’s
autonomy, while the latter focuses on Stiquito’s uses in educa-
tion by presenting lab exercises.
Stiquito Controlled! comes with all of the parts needed to
build the Stiquito and includes new design-oriented material.
In particular, the book presents an introduction to embedded
systems, layout, and manufacturing of the robot’s printed cir-
cuit board (PCB) as well as information on assembling and
programming the Stiquito and optimizing the robot’s motion.
Several changes are introduced to simplify the process of
building the base robot. The inclusion of material for the lay-
out and manufacturing of the PCB is a radical feature not seen
in many books.
CONTENTS
Chapter 1 presents an overview of robotics and the origin of
Stiquito, as well as suggestions on how to proceed with read-
ing the book and building the kit. This chapter starts with a list
of the skills and time required to produce a good working
1066-033X/06/$20.00©2006IEEE
robot. In our case, the time it took for one person to build the
kit for the first time was four hours, which is the longest period
of time suggested by the book. This chapter continues with a
definition of a robot, Asimov’s three laws of robotics, a brief
mention of robotic applications in industry and entertainment,
and a detailed history of how Stiquito has evolved into its cur-
rent form. After presenting this useful information, Chapter 1
ends by stressing important safety practices necessary for
building the kit. However, the corresponding section lacks a
descriptive title to attract the readers’ attention.
An interesting introduction to embedded systems and their
increasing applications is presented in Chapter 2. Embedded
systems, which are small computers with microprocessors
that are integrated into everyday items from cell phones to
automobiles, can be programmed to perform specific tasks.
Stiquito, which has a microprocessor and is programmable,
fits the definition of an embedded system. Understanding
how Stiquito is designed and built requires some background
in embedded systems. Chapter 2 provides this background
and discusses the procedure for product development and
some microprocessor basics. This chapter concludes by pre-
senting the design steps for the Stiquito controller board as an
embedded system and the final design in detail. This chapter
can be used as a tutorial for electrical engineering students
taking courses in embedded systems.
Chapter 3 is dedicated to the layout and manufacturing of
PCBs in general. This chapter discusses how to complete the
schematic for a circuit design using a computer-aided design
(CAD) schematic-capture tool, which aids in drawing and
checking the design as well as listing the required components.
Chapter 3 continues with a detailed presentation of the manu-
facturing processes of PCBs, from a single layer to six layers.
Testing the Stiquito’s PCB is introduced at the end of Chapter 3
with a reference to its CAD files on the Internet. Although
Chapter 3 complements Chapter 2 by describing the imple-
mentation of an embedded-system design, this chapter, due to
its general nature, seems unrelated to the rest of the book.
Chapter 4, the central part of the book, explains how to
build the Stiquito for use with the microcontroller PCB and
how to add other connectors to program the board. We fol-
lowed the instructions provided by the author to build the
Stiquito. The list of required tools summarized in a table was
useful for setting up our workbench. In addition to the listed
tools, however, a flathead screwdriver was needed to tighten
the wires and a marker was needed to designate the lengths of
wire that had to be cut. The instructions for bending the legs
and inserting them in the holes on the robot’s body were easy
to follow. The legs also hold a power bus (a copper wire) onto
the robot’s body. The power bus was a little longer than the
size indicated in the book. Also, we suggest that the author
include a top-view figure showing which holes on the robot’s
body are to be used for inserting the legs. This simple figure
would minimize the guesswork associated with this step.
Bending the legs at the knees is as easy as suggested in the
book, although mentioning which tool to use would have
been helpful. We used a pair of pliers, which worked well.
Trimming the legs such that the robot stood level on the
ground was fairly straightforward. The next step was to
ensure that the screws with washers were inserted into the
correct body holes. Recognizing the correct holes was difficult
since the front and rear of the body are not marked properly.
When following the instructions in the book, we did not
encounter any problems with cutting the aluminum tube to
make the leg crimps or cutting the nitinol wires to the
required length. Next, we took pieces of nitinol, tied knots in
them, and fed them into the leg crimps. The book emphasizes
that tying the knots requires much patience. Our experience
confirmed this statement. Crimping the leg crimps to hold the
knot, wrapping the other end of the wire around the screws,
and tightening the screws were straightforward steps. Build-
ing the robot was completed by testing the resistance of each
leg, applying current to the legs to observe their motion, bend-
ing the end of the legs to make the feet, mounting the micro-
controller PCB on top of the body, and attaching the power
bus to the battery compartment. The instructions for these
tasks were clear and error free.
The last section of Chapter 4 describes how to operate the
Stiquito robot. Our robot worked after readjusting the tension
of the nitinol wires and adjusting the speed of the legs by
turning a potentiometer on the PCB, as instructed by the book.
During the robot’s initial test walk, three of the legs in one tri-
pod fully contracted backwards and then relaxed and
returned to their original positions. This same sequence of
movements was then performed by the other tripod. Chapter
4 finishes with a well-organized and easy-to-use troubleshoot-
ing guide. We were not able to test the instructions in the
Appendix of Chapter 4 due to the lack of optional parts.
The Texas Instruments microcontroller used on the robot’s
PCB as well as programming the Stiquito are discussed in
Chapter 5. The controller is preprogrammed to control the one-
and 2-DOF legs. In 1-DOF mode, only the backward motion of
the legs is controlled, whereas in 2-DOF mode, both the back-
ward and lifting motions of the legs are controlled. The con-
troller switches between the modes by adding or removing a
jumper. In addition to performing these movements, the
microcontroller is preprogrammed for pulse-modulation,
which reduces the power consumption of the robot. The micro-
controller board code is written in C and is compiled using the
free IAR Embedded Workbench Integrated Development Envi-
ronment (IDE) 3.0B, which is a part of tool Kick-start V2. The
software architecture and algorithms of these programs are
given in Chapter 5 as a guide for programming the gait of the
robot. At the end of Chapter 5, the user is invited to download
code examples from the IEEE Computer Society Press Stiquito
support Web site [9] and change the program to achieve a dif-
ferent gait, add and activate sensors, and program behaviors in
the robot based on sensor inputs.
Chapter 6 introduces yet another variation of Stiquito.
This variation, which has 2 DOF, uses the standard Stiquito
design with the addition of six shape-memory alloy actuator
APRIL 2006 « IEEE CONTROL SYSTEMS MAGAZINE 93
wires, which lift the legs off the ground. Mechanical and
electrical designs are presented in Chapter 6. This Stiquito
variation uses a BASIC Stamp 2 microcontroller, which
interprets BASIC language programs for instructions. The
PCB design and construction details are also shown in
Chapter 6. At the end of the chapter, a sample of the Paral-
lax BASIC program, run on the microcontroller to control
the 2-DOF legs, is listed. This program can also be viewed at
the IEEE Computer Society Press Web site [9]. Chapter 6
shows how the basic skills and knowledge developed by
building the first Stiquito can be used to build more
advanced walking robots.
Optimizing the Stiquito robot for speed by altering the
robot’s design is the topic of Chapter 7. The parameters that
affect the speed of the robot include the number of DOFs
per leg, the friction of the legs with the ground, the number
of legs, whether the robot moves forward during the con-
traction or relaxation of the nitinol wire, the lengths of the
legs, and the speed of the leg contractions. Chapter 7 lists
the results of experiments with three different variations of
the robot that were altered based on the above parameters,
including a six-legged robot with 2-DOF standard legs, a
six-legged robot with 1-DOF long legs, and an eight-legged
robot with 1-DOF legs. Chapter 7 concludes that these
experimental results cannot provide a definitive answer
about which parameter is the most important, due to varia-
tions in weight and power supply.
Chapter 8 summarizes innovative variations of Stiquito by
different people. Among these designs are a robot controlled
by an analog controller rather than a digital microcontroller, a
BEAMStiquito controlled by the neural-network method, a
Stiquito with an Amtel AVR-based controller, two four-legged
Stiquitos, and a Caterpillar Stiquito.
CONCLUSIONS
Stiquito Controlled! is an interesting book and combo kit pre-
senting the details of the development, design, implementa-
tion, production, and programming of a mechatronic system.
The novelty of this product lies in the opportunity it gives the
reader to participate in this development by providing the
parts to build a mechatronic system. The instructions included
for building the robot are fairly accurate, which makes the
building process more enjoyable.
Matrix Analysis for Scientists
and Engineers
By ALAN J. LAUB
TEACHING LINEAR ALGEBRA
A basic course in linear algebra or matrix theory is mandatory
in every engineering, mathematics, and science program, and
a great variety of textbooks are available. Is another one needed?
94 IEEE CONTROL SYSTEMS MAGAZINE » APRIL 2006
The Stiquito kit and book can be inspiring learning aids for
students serving to tweak their curiosity about robotics and
introduce them to a hands-on experience. Stiquito Controlled!
can be used as a source of ideas for design projects for senior
students. This book is suitable as a supplemental text for
undergraduate courses in robotics, mechatronics, and embed-
ded systems. However, since the topic of Stiquito Controlled! is
about the variations of a specific robot design, this book can-
not be used as a textbook for a standalone course.
—Farbod Fahimi and Andrew Browne
REVIEWER INFORMATION
Farbod Fahimi is an assistant professor at the University of
Alberta, Edmonton, Alberta, Canada. He received his Ph.D. in
1999 in mechanical engineering from Sharif University, Tehran,
focusing on control and dynamics. In 2002, he held a postdoc-
torate fellowship in the Department of Mechanical Engineering,
Villanova University, Pennsylvania, where he worked on the
control of mobile robots. He was with the same department as
an assistant professor until August 2005. His research interests
include obstacle avoidance and the control of autonomous vehi-
cle group formations, advanced robotics, and haptic devices.
Andrew Browne is a second-year mechanical engineering co-
op student at the University of Alberta, Canada. He is consid-
ering robotics or pressure vessel design as his degree focus.
REFERENCES
[1] R.G. Gilbertson, Working with Shape-Memory Wires. San Leandro, CA:
Mondo-Tronics, Inc., 1992.
[2] W.J. Mills, “Area-efficient implication circuits for very dense Lukasiewicz
logic arrays,” in Proc. 22nd Int. Symp. Multi-Valued Logic, Sendai, Japan,
1992.
[3] R. Brooks, “A robot that walks: Emerging behavior from a carefully
evolved network,” Neural Computat., vol. 1, no. 2, pp. 253–262, 1990.
[4] R. Beers, “An artificial insect,” Amer. Scientist, vol. 79, no. 5,
pp. 444–452, 1990.
[5] E.O. Wilson, Sociobiology: The Synthesis. Cambridge, MA: Harvard Univ.
Press, 1975.
[6] D.H. Ballard and C.M. Brown, Computer Vision. Englewood Cliffs, NJ:
Prentice-Hall, 1982.
[7] J.M. Conrad and J.W. Mills, Stiquito: Advanced Experiments with a Simple
and Inexpensive Robot. Los Alamitos, CA: IEEE Computer Science Press,
1997.
[8] J.M. Conrad and J.W. Mills, Stiquito for Beginners: An Introduction to Robot-
ics. Los Alamitos, CA: IEEE Computer Science Press, 1999.
[9] Stiquito controller board code listings. IEEE Computer Science Press Stiq-
uito support page [Online]. Available: http://computer.org/books/stiquito
I would definitely say “yes,” and if I had written one, it would
be just like the one that Alan J. Laub has written. Why is a
textbook in this form needed?
There are essentially three basic approaches to teaching
linear algebra.
1) The abstract approach, which includes vector spaces,
homomorphisms, eigenvalues of endomorphisms, and
related topics. This coordinate-free approach is elegant,
and I learned the subject this way. However, students
1066-033X/06/$20.00©2006IEEE
 who are not mathematics for such a reduced course while providing the students with
majors cannot relate this the content that they need.
approach to their engineer- The 13 chapters of the book cover vector spaces, linear
ing or physics courses, transformations, pseudoinverses, singular-value decomposi-
which may involve linear tion, linear systems, inner products and norms, least-squares
systems and eigenvalue and problems, eigenvalue and eigenvectors, canonical forms, lin-
singular-value problems. ear differential and difference equations, generalized eigen-
2) The abstract-plus-coordi- value problems, and Kronecker products. For engineers and
nate-based approach, which scientists, these topics are the most important topics from lin-
includes vector spaces, ear algebra.
homomorphisms, and eigen- The author does a great job in concisely presenting the
values of endomorphisms, results. The explanations are accompanied by examples that
SIAM Publications, but with a parallel treatment are easy to understand as well as many exercises that are good
Philadelphia, 2005,
of coordinate vectors and for reinforcing understanding. With these features, the book is
US$25.20
ISBN 0-89871-576-8 matrices as representations of ideal for a reduced course.
linear mappings. Combined A single point of criticism that I have concerning this book
with an introduction to software such as MATLAB, such relates to my own experience in teaching this course. I would
a course (which consists of approximately 64 lectures of have slightly reordered the presentation and moved the singu-
45 minutes each in my institution) presents the field in lar-value decomposition and the pseudoinverse to after the
depth while relating the contents to real-world problems. chapter on canonical forms.
I teach the course in this way for mathematics students
and I believe in this approach. CONCLUSIONS
3) The reduced abstract-plus-coordinate-based approach. For engineering and science students, the modern curriculum
Forced by the unfortunate development that many engi- limits the amount of time devoted to a basic course in linear
neering schools reduce the mathematics contents of their algebra or matrix theory. For such a course, Matrix Analysis for
curricula to the absolute minimum, the number of hours Scientists and Engineers is an ideal textbook due to its topical
available for a basic linear algebra course has become limit- coverage, clarity, examples, and exercises.
ed. Currently, in my institution, this course occupies —Volker Mehrmann
approximately 32 lectures of 45 minutes each. This allot-
ment is not enough, and I strongly feel that this situation is REVIEWER INFORMATION
a disaster for the development of science and technology. In Volker Mehrmann is a professor at TU Berlin specializing in
teaching such a course, the instructor is forced to strip away numerical mathematics. His research interests are in the areas
the abstract theory as much as possible and teach the basics of applied and numerical linear algebra, control theory, and
that are most needed by future engineers and scientists. the theory and numerical solution of differential-algebraic
equations. He is editor-in-chief of Linear Algebra and Its Appli-
WHERE DOES THIS BOOK FIT IN? cations and author of several books, including a basic intro-
The textbook Matrix Analysis for Scientists and Engineers pro- ductory textbook on numerical analysis for scientists and
vides an exact fit for the third category. The book can be used engineers (in German).

Mathematical Systems Theory I:
Modelling, State Space Analysis,
Stability and Robustness
By DIEDERICH HINRICHSEN and ANTHONY J. PRITCHARD
SYSTEMS THEORY
The basic input-state-output concept of a system introduced
by Kalman [1] has become a cornerstone of control theory. In
the first chapter of [1], we are told that
The terms “systems,” “system concepts,” “systems
approach,” and “systems science” are used so widely
and so broadly today that they tend to connote fuzzy
thinking. For us, however, a system, or more correctly a
dynamical system, is a precise mathematical object; the
study of systems theory is then largely, although not
entirely, a branch of mathematics.
Within the control community, systems theory is typically
thought of as the foundation for control analysis. Indeed, it is
common practice to intertwine systems and control in a work
on systems theory [2], [3]. An alternative approach is embod-
ied in the general systems theory of von Bertalanffy and oth-
ers, who define the field in much more general terms [4].
The approach taken by Hinrichsen and Pritchard in Mathe-
matical Systems Theory I is, as decreed above, that systems
theory is a branch of mathematics. Consequently, their text is a
mathematics book and a fairly heady one at that. This statement

1066-033X/06/$20.00©2006IEEE APRIL 2006 « IEEE CONTROL SYSTEMS MAGAZINE 95

 should not be taken as suggesting
that the authors disregard the
importance of the interconnections
between systems and control theo-
ry. On the contrary, the second vol-
ume in this set will be devoted to
issues of control. The presentation
in the current volume, while pre-
sumably laying the groundwork for
the next volume, can be taken on its
own as a treatment of systems theo-
ry. In this effort the authors succeed
Springer, 2005, US$79.95
ISBN 3-540-44125-5
admirably, although discussions of
control topics are occasionally held
at bay somewhat artificially, as when observability is intro-
duced as an unmotivated algebraic condition.
This book provides a thorough description of the theory of
linear systems. It may come as a surprise that a text on the
graduate level would focus almost exclusively on linear sys-
tems. The common perception is that students would already
have covered the subject as presented in undergraduate texts
such as [5]. The treatment in Mathematical Systems Theory I
goes far deeper than the typical understanding.
As a consequence, the required background for the text is not
a mastery of linear systems theory but rather a grounding in the
mathematical tools required to digest mathematically sophisticat-
ed results, such as a solid foundation in linear algebra, analysis,
and differential equations as well as some knowledge of func-
tional analysis. The reader should expect to see the use of tools
such as the Arzela-Ascoli theorem and isometric embeddings of
Banach spaces. The approach is unabashedly mathematical, with
most discussions proceeding from the general treatment to spe-
cific cases. Although examples are presented throughout, it is
clear that the goal of the text is to address the mathematical
results, not to develop their engineering applications.
CONTENTS
The first chapter introduces various systems and models. Pop-
ulation dynamics and economics are included alongside stan-
dard examples from mechanics, electronics, and digital
systems. The discussion of digital switching networks is a wel-
come inclusion that provides a solid context for the discrete-
time results that follow. The examples from mechanics and
circuit theory are accompanied by a general discussion of
Lagrangian and Hamiltonian approaches as well as a deriva-
tion of Maxwell’s equations. The presentation of the material
in this introductory chapter indicates the level of sophistica-
tion expected from the reader. These topics are followed by an
application that is rarely treated in a general text, namely, heat
transfer. This material helps motivate the subsequent discus-
sions on infinite-dimensional systems.
Chapter 2, on state-space theory, begins with a well-crafted
intuitive introduction to dynamical systems. This discussion is
followed by a careful definition of a system in the spirit of [1].
Several examples are chosen to illustrate the concept, with
96 IEEE CONTROL SYSTEMS MAGAZINE » APRIL 2006
automata and delay systems included along with the standard
pendulum and satellite. The discussion of linear systems is far
deeper than the usual treatment. For example, the careful dis-
cussion of eigenmotions provides a rigorous definition of
modes, a term that is sometimes used in a vague sense. Linear
infinite-dimensional systems are introduced as well, and the
authors are careful to point out the differences from the finite-
dimensional case, including the resulting complications.
Transfer matrices are discussed and the irrational transfer
functions derived from infinite-dimensional systems are rep-
resented by an example of a system with time delay. Although
the frequency response is defined, the classical single-input,
single-output tools for their study are mentioned only in pass-
ing. A section is devoted to sampling and approximation,
leading to a proof of the sampling theorem. This section is fol-
lowed by a discussion of approximation of continuous sys-
tems by discrete systems and methods for numerical
simulation, including an interesting treatment of the potential
pitfalls of applying multistep methods to controlled systems.
Chapter 3, which covers stability theory, begins with an
historical overview of the development of the mathematical
notion of stability. Stability is first defined abstractly, specifi-
cally in terms of nonlinear flows on metric spaces. Lyapunov’s
direct method, which is developed in the same general setting
(again with a welcome historical introduction), is illustrated
by application to a delay equation. The results are specialized
to time-varying, finite-dimensional systems and then to time-
invariant systems. This coverage is quite complete, including a
discussion of instability results as well as computation of
regions of attraction. The presentation then turns to linear sys-
tems with a discussion of Lyapunov and Bohl exponents, the
latter being the analog of Lyapunov exponents for addressing
uniform behavior of time-varying systems. Finally, stability
criteria for polynomials are presented in a compendium of
algebraic and analytical results that are rarely covered in a
course in control theory. The discussions provide deep insight
into stability algorithms that are commonly used. The two
applications worked in this section illustrate the most fre-
quently used tools in this area, namely, the Routh criterion
and the Jury test. Overall, this chapter provides an elegant dis-
play of mathematics.
Chapter 4 addresses the behavior of indicators of stability
under perturbations. Beginning with a discussion of the roots
of polynomials, the authors then turn to eigenvalues of matri-
ces. The singular value decomposition is introduced, and the
behavior of singular values under perturbations is addressed.
This discussion leads naturally to a corresponding treatment
of structured singular values in the context of µ-analysis. The
chapter ends with a detailed treatment of the computation of
eigenvalues and singular values. While the discussion of con-
dition number follows naturally from the rest of the material
in the chapter, the lengthy treatment of computational algo-
rithms (complete with pseudocode) seems a bit out of place.
The last chapter continues the discussion from Chapter 4
on the consequences of perturbations. Spectral value sets for
families of matrices are introduced, and their computation is
discussed. Stability radii (generalizations of gain margins) are
then addressed in a treatment that relies heavily on the
authors’ own research. The discussion next turns to a subject
often left untreated, namely, stability bounds on transient
behavior and the corresponding robustness analysis.
CONCLUSIONS
In the preface to [1], the authors state that they have “tried to con-
vey the fundamental notion that systems theory is not simply a
branch of applied analysis, but provides a source of problems
and intuition for a rich interplay between algebra and analysis.” The
book under review is a celebration of that interplay.
I heartily recommend this book as a remarkably complete ref-
erence for researchers doing theoretical work involving linear
systems. The book is a unique compilation of the mathematics
underlying the field, and I look forward to seeing the author’s
treatment of synthesis problems in the second volume.
More care must be taken in recommending the use of this
text for students. There is no doubt that parts of the text could
be successfully used for a graduate course in a mathematics
program; in fact, exercises are provided throughout to facilitate
that use. The authors indicate that they have also had success
using the first few chapters in undergraduate classes. However,
it is my experience that undergraduates typically balk at the
level of rigor required by Sontag’s text [6]. Mathematical Systems
Theory I, which is written in a similar vein, is considerably more
demanding and would undoubtedly invoke a similar reaction.
The book is a welcome addition to the library of work on
systems theory and will no doubt serve as a valuable reference
in many studies. Most readers will come to the book with
control-theoretic applications in mind, and while the connec-
Synchronization
and Control of Chaos:
An Introduction
for Scientists
and Engineers
By J.M. GONZÁLEZ-MIRANDA
CHAOS CONTROL
Chaos control refers to manipu-
lating the dynamical behavior
Imperial College Press,
of a chaotic system, in which
London, 2004, 212 pp + xi,
US$52.00 the goal is to suppress chaos
ISBN 1-86094-488-4 when it is harmful or to
enhance or create chaos when it
is beneficial. Chaos synchronization, on the other hand, refers
to the task of enabling or disabling the dynamical synchrony
of several connected chaotic systems by means of control tech-
1066-033X/06/$20.00©2006IEEE
tions to those applications are often left implicit, the text skill-
fully highlights the rigorous basis on which control engineer-
ing rests. Even von Bertalanffy, a biologist who applied his
general system theory to psychology and the social sciences, is
quick to remind us that “systems theory . . . is preeminently a
mathematical field” [4]. It is those mathematical foundations
that have been carefully documented by the authors in Mathe-
matical Systems Theory I.
—Brian Ingalls
REVIEWER INFORMATION
Brian Ingalls is an assistant professor in the Department of
Applied Mathematics at the University of Waterloo. He
received his Ph.D. in mathematics from Rutgers University in
2001 and held a postdoctoral fellowship at the California Insti-
tute of Technology in 2001–2002. His main research interests lie
in the application of systems and control theory to biochemical
regulation. He is an associate editor of IEE Proceedings Systems
Biology and serves on the editorial board of the International
Journal of Applied Mathematical Analysis and Applications.
REFERENCES
[1] R.E. Kalman, P.L. Falb, and M.A. Arbib, Topics in Mathematical
Systems Theory. New York: McGraw-Hill, 1969.
[2] H.K. Khalil, Nonlinear Systems, 3rd ed. New Jersey: Prentice-Hall,
2002.
[3] S. Sastry, Nonlinear Systems: Analysis, Stability and Control. New York:
Springer-Verlag, 1999.
[4] L. von Bertalanffy, General Systems Theory: Foundations, Development,
Applications. New York: George Braziller, 1968.
[5] C.-T. Chen, Linear System Theory and Design, 3rd ed. Oxford: Oxford Univ.
Press, 1999.
[6] E.D. Sontag, Mathematical Control Theory: Deterministic Finite Dimensional
Systems, 2nd ed. New York: Springer-Verlag, 1998.
niques or through specially designed coupling configurations.
These topics are addressed in the book Synchronization and
Control of Chaos: An Introduction for Scientists and Engineers by
J.M. González-Miranda.
Numerous research monographs, textbooks, and edited
volumes are devoted to the subjects of synchronization and
control of chaos. Examples include [1]–[10], although none of
these works are included in the 217 citations of the book.
Since the preface states that “The purpose of this book is to
provide a systematic and broad account of that research for a
wide audience,” I was interested in seeing how these topics
are presented in a book intended as an introduction for scien-
tists and engineers.
CONTENTS
The book has eight chapters, none of which include any exercises.
Chapter 1 begins with an introduction to standard exam-
ples of free and damped second-order linear mechanical har-
monic oscillators, evolving to one external-force driven and
APRIL 2006 « IEEE CONTROL SYSTEMS MAGAZINE 97
two coupled oscillators, then to a cubic-nonlinear (Duffing)
oscillator. This short introductory chapter is quite well writ-
ten, preparing basic material required in the sequel.
Chapter 2 reviews basic concepts in chaos theory, includ-
ing nonlinear continuous flows and discrete maps, using the
Lorenz system and Hénon map as examples. The chapter also
introduces Poincaré maps and bifurcation diagrams using
Chua’s circuit as the platform. Chaotic time series analysis,
typically on Lyapunov exponents and power spectra, as well
as a special feature of chaotic orbits (unstable periodic orbits)
are discussed in detail. In this regard, however, the standard
references [11], [12] on nonlinear and chaotic time-series
analysis are regrettably not cited.
Chapter 3 introduces the concept of periodically driven
chaotic oscillation along with the related topics of phase syn-
chronization and chaos suppression through external-force
input. Experimental and numerical observations of these phe-
nomena are demonstrated using physical examples such as
lasers, Josephson junctions, and single-neuron models. The
author provides a good description of the basic phenomenon
of chaotic phase synchronization. In referencing this topic, the
overview article [13], which is not cited, can be beneficial for
the reader’s further edification on state-of-the-art develop-
ments in chaotic phase synchronization. As discussed below,
several references cited in the book are either not the best
selections from the available literature or are out of date
despite the statement that “the book combines tutorial and
review techniques . . . All the above is provided together with
a bibliography aiming to allow the reader to probe deeper into
the particular issues that especially interest him or her.”
Chaos suppression by means of external weak resonant
forces is discussed in the second half of Chapter 3, based on an
open-loop control technique suggested in the early 1990s; see,
for example, Chapter 4 of [5]. Brief comments on variants of
this methodology are provided at the end of the chapter. I
would refer the reader to the more updated and comprehen-
sive reference [14].
Chapter 4 introduces the concept as well as elementary
techniques for chaos synchronization, mainly based on the
drive-response (also called masterslave) configuration. Com-
plete and generalized synchronizations, for both identical and
nonidentical systems, are discussed. The reader can learn the
essentials of chaos synchronization by reading through this
well-written chapter. In particular, “marginal synchronization
of chaos” in the phase space, attributed to the author of the
book, is studied in detail. The chapter closes with observations
on generalized chaos synchronization. It is important to note
that there exist alternative synchronization settings, such as
chaotic switching and chaotic modulation, that are not intro-
duced in the chapter. For additional chaos synchronization
configurations, see Chapter 11 of [5]. As an introductory trea-
tise on synchronization and control of chaos, this omission
leaves the book incomplete.
Chapter 5 is devoted to the study of chaos control by
means of perturbations, mainly the Ott-Grebogi-Yorke
98 IEEE CONTROL SYSTEMS MAGAZINE » APRIL 2006
(OGY) method and delayed feedback control. This 21-page
chapter (pp. 111–131), as well as the eight-page “chaos sup-
pression” portion in Chapter 3 (pp. 70–77) reviewed above,
constitute the main subject of chaos control as covered by the
book. Compared to the materials on chaos synchronization,
this emphasis is unbalanced. As a result, all other approach-
es to chaos control available in the literature, which includes
more than 20 well-known methods and techniques (see, for
example, [1]–[9] and particularly Chapters 5–10, pp. 173–534,
of [5]), are not presented here. Control engineers are familiar
with linear and nonlinear feedback control, adaptive control,
optimal control, sliding-mode control, self-tuning control,
robust control, digital control, stochastic control, intelligent
control, distributed control, and many other system regula-
tion and stabilization methodologies. Seeing the eye-catching
title of the book, an engineer who is interested in delving
into this new research field may wonder whether these con-
ventional control techniques work equally well for chaotic
systems and, if so, whether there are any special features to
be anticipated by taking advantage of chaos theory.
Although intended as an introduction for scientists and engi-
neers, presumably targeting control engineers, this omission
fails to provide the whole picture of chaos control but,
rather, shows only a very small piece of it. Setting aside this
important issue, the key topics of OGY control and delayed
feedback control are not introduced with updated informa-
tion. Although these approaches are described quite well in
the text, the OGY method is further developed in subsections
3.2.6–3.2.7 of [5] and Chapter 12 of [8]. Also, the delayed
feedback control method has an inherent defect of the odd
number limitation [15], [16], which should be noted in an
introductory text.
Chapters 6 and 7 extend the chaos synchronization frame-
work from a pair of coupled oscillators to a network of
multi-coupled oscillators. The case of identical chaotic oscil-
lators is studied in Chapter 6, while the nonidentical setting
is discussed in Chapter 7. In these chapters, network syn-
chronization is presented by numerical simulations, with
physical experiments for lower dimensional cases. Phase
synchronization is presented with detailed discussions. The
recent trend of chaotic network synchronization [17], [18],
however, is not reflected in these chapters or elsewhere in
the book.
Chapter 8, in the author’s own words, “is devoted to a
summary of the main results presented in the book, to discuss
their possible technical and scientific applications, and to spec-
ulate on the perspectives of the field both from the fundamen-
tal and the applied points of view.” This well-written
summary includes three short subsections on applications,
namely, synchronization in neurobiology, synchronization in
the earth sciences, and chaotic communications. Chaos control
and synchronization have many more potential applications;
see, for example, chapters 11 and 12 of [5].
The subject of chaos synchronization in the human brain
is a good selection for study. As noted in [19], “The brain
transforms sensory messages into conscious perceptions
almost instantly. Chaotic, collective activity involving
millions of neurons seems essential for such rapid recogni-
tion.” Within this context, the chaotic and collective activity
often refers to chaos synchronization [19]. As the author
points out on page 193, “Modelling of climatic and meteoro-
logical scenarios is another application of interest.” The topic
of synchronized motions in patterns of atmospheric circula-
tion is a natural choice for discussion in this section, based
on the familiar Lorenz model. Generally speaking, these top-
ics are well presented.
Readers of IEEE Control Systems Magazine may be interest-
ed in the third subject, chaotic communications. This topic,
unfortunately, is not well presented in the book, and only
some initial work from the early 1990s is introduced. If one is
interested in the novelty and merits of chaotic communica-
tions versus conventional communications, then the best
research includes [20] and [21]. If the reader is interested in
the security aspect of chaotic communications, then the setting
based on chaotic signal masking and discovery introduced in
this chapter is known to be insecure, as cryptanalyzed by sev-
eral papers (see, for example, [22] and the references in [23]).
In general, chaotic communication and encryption technolo-
gies are not well reflected in this part of the book. The goal of
the author “to exemplify what might be a major line of devel-
opment of this field in the next few years: the application of
the new ideas and techniques in all the sciences, medicine and
engineering” (p. 196) is far from being reached.
CONCLUSIONS
As noted in the preface, the book’s author has “been working
on coupled and driven chaotic oscillators for the last ten
years.” Indeed, the corresponding part of the contents is clearly
written and well presented. If the reader seeks to learn or
teach chaos synchronization, this book is useful. In particular,
the topic of chaotic phase synchronization is well presented,
and the discussion is thorough; thus, this part of the text is
highly recommended.
However, I do not recommend this text as an introduction
for scientists and engineers. I believe that the goal of this book,
as stated in the preface, “to provide a systematic and broad
account of that research,” has not been achieved. For the read-
er’s information and further reading, given the vast literature
on the rapidly evolving research subjects of chaos control and
chaos synchronization, the bibliographies [24] and [25] in con-
junction with [1]–[10] are more informative. Nevertheless, in
the information-explosive era today, writing a good research
monograph or textbook is never an easy task. From this point
of view, I commend the author for the courage, time, and
effort devoted to this book.
—Guanrong Chen
REVIEWER INFORMATION
Guanrong Chen is a chair professor and the founding director
of the Centre for Chaos Control and Synchronization at the
City University of Hong Kong. He is a Fellow of the IEEE for
fundamental contributions to the theory and applications of
chaos control and bifurcation analysis.
REFERENCES
[1] M. Lakshmanan and K. Murali, Eds., Chaos in Nonlinear Oscillators: Con-
trolling and Synchronization. Singapore: World Scientific, 1995.
[2] T. Kapitaniak, Controlling Chaos. New York: Academic, 1996.
[3] L. Fradkov and A. Yu. Pogromsky, Introduction to Control of Oscillations
and Chaos. Singapore: World Scientific, 1997.
[4] K. Judd, A. Mees, K.L. Teo and T.L. Vincent, Eds., Control and Chaos.
Boston: Birkhauser, 1997.
[5] G. Chen and X. Dong, From Chaos to Order: Methodologies, Perspectives and
Applications. Singapore: World Scientific, 1998.
[6] H.G. Schuster, Ed., Handbook of Chaos Control. New York: Wiley, 1999.
[7] G. Chen, Ed., Controlling Chaos and Bifurcations in Engineering Systems.
Boca Raton: CRC Press, 2000.
[8] C.W. Wu, Synchronization in Coupled Chaotic Circuits and Systems. Singa-
pore: World Scientific, 2002.
[9] G. Chen and X. Yu, Eds., Chaos Control: Theory and Applications. Berlin:
Springer-Verlag, 2003.
[10] H. Nijmeijer and A. Rodriguez-Angeles, Synchronization of Mechanical
Systems. Singapore: World Scientific, 2003.
[11] H. Kantz and T. Schreiber, Nonlinear Time Series Analysis, 2nd ed. Lon-
don: Cambridge Univ. Press, 2004.
[12] J.C. Sprott, Chaos and Time-Series Analysis. London: Oxford Univ. Press,
2003.
[13] S. Boccaletti, J. Kurths, G. Osipov, D.L. Valladares, and C.S. Zhou, “The
synchronization of chaotic systems,” Phys. Rep., vol. 366, no. 1–2,
pp. 1–101, 2002.
[14] R. Chacón, “Maintenance and suppression of chaos by weak harmonic
perturbations: A unified view,” Phys. Rev. Lett., vol. 86, no. 9, pp. 1734–1740,
2001.
[15] T. Ushio, “Limitation of delayed feedback control in nonlinear discrete-
time systems,” IEEE Trans. Circuits Syst. I., vol. 43, pp. 815–816, 1996.
[16] H. Nakajima and Y. Ueda, “Limitation of generalized delayed feedback
control,” Physica D, vol. 111, pp. 143–150, 1998.
[17] X.F. Wang and G. Chen, “Synchronization in small-world dynamical
networks,” Int. J. Bifurcation Chaos, vol. 12, pp. 187–192, 2002.
[18] X.F. Wang and G. Chen, “Synchronization in scale-free dynamical net-
works: Robustness and fragility,” IEEE Trans. Circuits Syst. I, vol. 49, pp.
54–62, 2002.
[19] W.J. Freeman, “The physiology of perception,” Sci. Amer., pp. 78–85,
Feb. 1991.
[20] G. Kolumbán and M.P. Kennedy, “Overview of digital communica-
tions,” in Application of Chaotic Electronics to Telecommunications, M. P.
Kennedy, R. Rovatti, and G. Setti, Eds. Boca Raton: CRC Press, 2000, pp.
131–149.
[21] F.C.M. Lau and C.K. Tse, Chaos-Based Digital Communication Systems.
New York: Springer-Verlag, 2003.
[22] K.M. Short, “Signal extracting from chaotic communications,” Int. J.
Bifurcation Chaos, vol. 7, no. 7, pp. 1579–1597, 1997.
[23] T. Yang, “A survey of chaotic secure communication systems,” Int. J.
Computat. Cognit., vol. 2, no. 2, pp. 81–130, 2004.
[24] Bibliography of chaos control and chaos synchronization (1987–1997)
[Online]. Available: http://www.ee.cityu.edu.hk/~gchen/chaos-bio.html
[25] Bibliography of chaos control and chaos synchronization (1997–2000)
[Online]. Available: http://www.rusycon.ru/chaos-control.html#g
APRIL 2006 « IEEE CONTROL SYSTEMS MAGAZINE 99