See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/242677427

Applied Linear Optimal Control: Examples and Algorithms

Article in Applied Mechanics Reviews · July 2003

DOI: 10.1115/1.1579457

CITATIONS READS
62 2,071

2 authors:

Arthur Bryson M. Reyhanoglu

Stanford University Columbus State University
121 PUBLICATIONS 13,881 CITATIONS 174 PUBLICATIONS 5,150 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by M. Reyhanoglu on 06 June 2014.

The user has requested enhancement of the downloaded file.

 Applied Linear Optimal Control: Examples and Algorithms
Arthur E. Bryson, Cambridge University Press, United Kingdom, 2002, 362
pp., $45.00, ISBN 0 521 01231 7 paperback

The author states in the preface, “This book discussion on estimator divergence, with a
is about the optimization of dynamic systems in common solution to this problem. Problem sets
the presence of uncertainty.” Here, uncertainty are scattered throughout the text, which are
includes both random inputs and errors, and generally given at the end of each section. This
parametric uncertainties. The stated intent of allows the reader a chance to test their skills
this book is to be the successor to the second immediately after reading about a certain topic.
half of Applied Optimal Control,1 where Ref. 2 Clearly, the author used a deliberate and
is the successor of the first half. The book meaningful thought process in the preparation of
generally succeeds in its goal. this book.
The book comprises twelve chapters and Intrinsic with the reading material is the
two appendices. The first chapter is dedicated wide use of MATLAB. The book contains a
to static estimation, which includes a review of CD-ROM that includes a toolbox as well as
random variables. Chapter 2 provides an codes for nearly all the examples, figures and
overview of random processes, which leads to problems within the text. The reader is
dynamic estimation using filters and smoothers expected to have at least some experience with
shown in Chapters 3 and 4, respectively. In MATLAB programming. However, the author
Chapter 5 Linear-Quadratic State-Feedback successfully backs up the codes with detailed
Follower-Controllers are shown, which leads to explanations in the book that provide a clear
the standard Linear Quadratic Gaussian (LQG) understanding of the overall algorithmic
control problem discussed in Chapter 6. In programming. Therefore, anyone with even a
Chapter 7 the results of Chapter 4 are extended terse background in MATLAB should be able to
to handle controlled plants. The next three easily comprehend the relationships between the
chapters are devoted to the application of the stated applications in the book and the coded
results from previous chapters to time-invariant scripts.
systems. Chapter 11 presents Worst-Case The intended audience of this book includes
Controllers, while Chapter 12 presents first year graduate students and practicing
Parameter-Robust LQG Controllers. Appendix engineers with some optimal control experience.
A gives various filter and controller designs for The author inherently applied the old adage of
systems with colored measurement noise, and “the best way of learning is doing” in preparing
Appendix B gives the plant models used for the this book. This is clearly demonstrated by the
examples in the book. plethora of examples shown in the book. These
The book is aesthetically very well written. examples cover a wide variety of systems,
The chapters progress in a logical fashion. In including: spacecraft attitude and orbit
particular, the developments shown in the first maneuver control, aircraft and helicopter
ten chapters lead to several advanced topics, control, robotic-type systems and ground
such as Linear-Quadratic Worst-Case Control, vehicle motion systems, as well as standard
Disturbance Rejection and Minimax Control, in problems such as planetary gear problems, the
the last chapters. Algorithms for both overhead crane and the inverted pendulum.
continuous-time and discrete-time systems are These real-world systems and nontrivial
provided. Several practical discussions are problems provide the reader a good foundation
given that enhance the pedagogical experience for the application of linear optimal control
as well. For example, the author gives a theory to more complicated systems.
 Since the style of this book involves a learn-
by-practice approach, it is not as mathematically
rich as the author’s classic book of Ref. 1.
Several of the detailed derivations shown in Ref.
1, for example, the Bayesian approach to
optimal filtering and optimal filtering/control
for nonlinear systems, have been omitted in the
present text. Also, even though the premise of
the book is linear systems, a more detailed
treatment of nonlinear systems would be useful,
since nonlinear estimation and control have now
reached the mainstream engineering practice.
Therefore, some of the concepts may have to be
supplemented with more detailed treatments,
and the book seems to be most effective when
used in conjunction with more mathematically
rich texts or detailed theoretical notes.
In summary, Applied Linear Optimal
Control successfully fills the gap between the
theory of optimal control design and practical
applications in the face of uncertainties. The
student should find this book to be an invaluable
supplemental tool combined with a solid
theoretical framework in the classroom. The
practicing engineer should find this book to be
an invaluable reference tool in understanding
the issues involved with practical control
design. Both the student and the practicing
engineer will especially appreciate the several
examples backed up with MATLAB codes,
which serve to provide more interesting
problems than the book’s predecessor. This
book should be a valuable addition to an
individual’s collection or to an institution’s
library.
References
1. Bryson, A.E., and Ho, Y.-C., Applied
Optimal Control: Optimization, Estimation
and Control (revised printing), Taylor &
Francis, London, 1975.
2. Bryson, A.E., Dynamic Optimization,
Addison-Wesley, Menlo Park, CA, 1999.

John L. Crassidis
University at Buffalo

View publication stats