logo资料库

控制理论最经典的25篇论文.pdf

第1页 / 共517页
第2页 / 共517页
第3页 / 共517页
第4页 / 共517页
第5页 / 共517页
第6页 / 共517页
第7页 / 共517页
第8页 / 共517页
资料共517页,剩余部分请下载后查看
Contents
Regeneration Theory
Stabilized Feedback Amplifiers
Relations Between Attenuation and Phase in Feedback Amplifier Design
The Linear Filter for a Single Time Series
Control System Synthesis by Root Locus Method
The Structure of Dynamic Programming Processes
Optimal Regulation Processes
Contributions to the Theory of Optimal Control
A New Approach to Linear Filtering and Prediction Problems
Dual Control Theory
Absolute Stability of Nonlinear Systems of Automatic Control
A Steepest-Ascent Method for Solving Optimum Programming Programs
The Solution of Certain Matrix Inequalities in Automatic Control Theory
Mathematical Description of Linear Dynamical Systems
On the Input-Output Stability of Time-Varying Nonlinear Feedback Systems
An Invariance Principle in the Theory of Stability
Decoupling and Pole Assignment in Linear Multivariable Systems: AGeometric Approach
System Theory on Group Manifolds and Coset Spaces
Controllability of Nonlinear Systems
Dissipative Dynamical Systems-Part I: General Theory
On Self Tuning Regulators
Nonlinear Controllability and Observability
Analysis of Recursive Stochastic Algorithms
Discrete-Time Multivariable Adaptive Control
Feedback and Optimal Sensitivity: Model Reference Transformations, Multiplicative Seminorms, and Approximate Inverses
Index
About the Editor
CONTROL THEORY
IEEE Press 445 Hoes Lane, P.O. Box 1331 Piscataway, NJ 08855-1331 IEEE Press Editorial Board Robert J. Herrick, Editor in Chief M.Akay J. B. Anderson P. M. Anderson J. E. Brewer M.Eden M. E. El-Hawary R. F. Hoyt S. V. Kartalopoulos D. Kirk M.S.Newman M. Padgett W.D.Reeve G. Zobrist Kenneth Moore, DirectorofIEEE Press Catherine Faduska, SeniorAcquisitions Editor Robert H. Bedford, AssistantAcquisitions Editor Anthony VenGraitis, ProjectEditor Marilyn G. Catis, Marketing Manager IEEE Control Systems Society, Sponsor CSS Liaison to IEEE Press, Bruce M. Krogh Cover design: William T. Donnelly, WT Design Books of Related Interest from IEEE Press PERSPECTWES IN CONTROL ENGINEERING: Technologies, Applications, and New Directions Edited by Tariq Samad 2001 Hardcover 536 pp IEEE Order No. PC5798 ISBN 0-7803-5356-0 PHYSIOLOGICAL CONTROL SYSTEMS: Analysis,Simulation, and Estimation A volume in the IEEE Press Book series on Biomedical Engineering Michael C. K. Khoo 2000 Hardcover 344 pp IEEE Order No. PC5680 ISBN 0-7803-3408-6 THE CONTROL HANDBOOK A CRC Handbook published in cooperation with IEEE Press Edited by William S. Levine 1996 Hardcover 1566 pp IEEE Order No. PC5649 ISBN 0-8493-8570-9 ROBUSTVISIONFOR VISION-BASED CONTROL OF MOTION A volume in the SPIE/IEEE Series on Imaging Science & Engineering Edited by Markus Vincze and Gregory D. Hager 2000 Hardcover 264 pp IEEE Order No. PC5403 ISBN 0-7803-5378-1
CONTROL THEORY Twenty-Five Seminal Papers (1932-1981) Edited by Tamer Basar University ofIllinoisat Urbana-Champaign Editorial Board Brian D. O. Anderson Karl J. Astrom John Baillieul TamerBasar (Chair) Bruce A. Francis Alberto Isidori Petar V. Kokotovic Huibert Kwakemaak WilliamJ. Levine Lennart Ljung David Q. Mayne Jan C. Willems IEEE Control Systems Society,Sponsor A Selected Reprint Volume IEEE PRESS The Institute of Electrical and ElectronicsEngineers,Inc., New York
This book and other books may be purchased at a discount from the publisher when ordered in bulk quantities. Contact: IEEE Press Marketing Attn: Special Sales 445 Hoes Lane, P.O. Box 1331 Piscataway, NJ 08855-1331 Fax: +1 732 981 9334 For more information on IEEE Press products, visit the IEEE Online Catalog and Store: http://www.ieee.org/store. © 2001 by the Institute of Electrical and Electronics Engineers, Inc., 3 Park Avenue, 17th Floor, New York, NY 10016-5997. All rights reserved. No part of this book may be reproduced in anyform, nor may it be storedin a retrieval systemor transmitted in anyform, withoutwrittenpermission from the publisher. Printed in the United States of America. 10 9 8 7 6 5 4 3 2 1 ISBN 0-7803-6021-4 IEEE Order No. PC5870 Library of Congress Cataloging-in-Publication Data Control theory: em. p. twenty-five seminal papers (1931-1981) / edited by Tamer Basar, "IEEE Control Systems Society, sponsor." "A selected reprint volume." ISBN 0-7803-6021-4 1. Automatic control. 2. Control theory. Society. TJ213.7.C662000 629.8--dc21 I. Basar,Tamer. II. IEEE Control Systems 00-058171 CIP
Contents Preface vii Regeneration Theory Nyquist, H. (Bell Syst. Tech. J., Vol. 11, 1932, pp. 126-147.) 1 Stabilized Feedback Amplifiers Black, H. S. (Bell Syst. Tech. J., Vol. 13, 1934, pp. 1-18.) 25 Relations Between Attenuation and Phase in Feedback Amplifier Design Bode, H. W. (Bell Syst. Tech. J., Vol. 19,1940, pp. 421-454.) 45 The Linear Filter for a Single Time Series Wiener, N. (Chapter III from Extrapolation, Interpolation, and Smoothing ofStationary Time Series, 81 The M.I.T. Press, Cambridge, MA, 1949, pp. 81-103.) Control System Synthesis by Root Locus Method Evans, W. R. (Trans. Amer. Inst. Electric.Engineers, Vol. 69, 1950, pp. 66-69.) 107 The Structure of Dynamic Programming Processes Bellman, R. (Chapter 3 from DynamicProgramming, Princeton University Press, Princeton, NJ, 113 1957, pp. 81-89.) Optimal Regulation Processes Pontryagin, L. S. (Uspekhi Mat. Nauk, USSR, Vol. 14,1959, pp. 3-20. (English translation: Amer. 125 Math. Society Trans., Series 2, Vol. 18, 1961, pp. 321-339.» Contributions to the Theory of Optimal Control Kalman, R. E. (Bol. Soc. Mat. Mexicana, Vol. 5,1960, pp. 102-119.) 147 A NewApproach to Linear Filtering and Prediction Problems Kalman, R. E. [Trans. ASME (J. Basic Engineering), Vol. 82D, March 1960, pp. 35-45.] 167 Dual Control Theory, Parts I and II Feldbaum, A. A. [Automation and Remote Control, Vol. 21, April 1961, pp. 874-880, and May 1961, 181 pp. 1033-1039. (Russian originals dated September 1960, pp. 1240-1249, and November 1960, pp. 1453-1464.)] Absolute Stability of Nonlinear Systems of Automatic Control Popov, V. M. [Automation and Remote Control, Vol. 22, February 1962, pp. 857-875. (Russian original 197 dated August 1961, pp. 961-979.)] A Steepest-AscentMethod for Solving Optimum Programming Problems Bryson, A. E., and Denham, W. F. [Trans. ASME (J. Appl. Mechanics), June 1962, pp. 247-257.] 219 v
vi CONTENTS The Solution of Certain Matrix Inequalities in Automatic Control Theory Yakubovich, V. A. (DANDokladyAkademiiNauk SSSR),Vol. 143, 1962, pp. 1304-1307. (English 233 translation: SovietMathematics (by AmericanMath. Society), 1962, pp. 620-623.) Mathematical Description of Linear Dynamical Systems Kalman, R. E. (SIAMJ. Control, Vol. 1, 1963, pp. 152-192.) 239 On the Input-Output Stability of Time-Varying Nonlinear Feedback Systems-Part I: Conditions derived using concepts of loop gain, conicity, and positivity; Part II: Conditions involving circles in the frequency plane and sector nonlinearities 283 Zames, G. (IEEETrans. Automat. Contr., Vol. AC-ll, 1966, pp. 228-238 and 465-476.) An Invariance Principle in the Theory of Stability Lasalle, J. P. (in Differential Equations and Dynamical Systems, J. Hale and J. P. LaSalle, Eds., 309 Academic Press, New York, 1967, pp. 277-286.) Decoupling and Pole Assignment in Linear Multivariable Systems: A Geometric Approach Wonham, W. M., and Morse, A. S. (SIAMJ. Control, Vol. 8,1970, pp. 1-18.) 321 System Theory on Group Manifolds and Coset Spaces Brockett, R. W. (SIAMJ. Control, Vol. 10, 1972, pp. 265-284.) 341 Controllability of Nonlinear Systems Sussmann, H. J., and Jurdjevic, V. (J. Diff. Eqns., Vol. 12, 1972, pp. 95-116.) 363 Dissipative Dynamical Systems-Part I: General Theory Willems, J. C. (Arch. Ratl. Mech. and Analysis,Vol. 45, 1972, pp. 321-351.) 389 On Self-Tuning Regulators Astrom, K. J., and Wittenmark, B. (Automatica, Vol. 9,1973, pp. 185-199.) 423 Nonlinear Controllability and Observability Hermann, R., and Krener, A. J. (IEEETrans. Automat. Contr., Vol. AC-22, 1977, pp. 728-740.) 441 Analysis of Recursive Stochastic Algorithms Ljung, L. (IEEETrans. Automat. Contr., Vol. AC-22, 1977, pp. 551-575.) 457 Discrete Time Multivariable Adaptive Control Goodwin, G. C., Ramadge, P. J., and Caines, P. E. (IEEETrans. Automat. Contr., Vol. AC-25, 1980, 485 pp.449-456.) Feedback and Optimal Sensitivity: Model Reference Transformations, Multiplicative Seminorms, and Approximate Inverses 495 Zames, G. (IEEETrans. Automat. Contr., Vol. AC-26, 1981, pp. 301-320.) Index 517 About the Editor 523
Preface CONTROL is in one sense a fairly young discipline. Even though it would be possible to push its historical origins back by about two millennia to the days of the Babylonians, in modem terms the real creation of the field has been in the twentieth century. It is in this century that CONTROL became a scientificdiscipline, with an intellectual core shaped by revolutionary ideas, novel concepts, and a wealth of analytical and computational tools. As a young and intellectually stimulatingdiscipline, it attracted someof thebrightestmindsto itsranksand,withitstheorydriven by real applications, it provided versatile tools for generations of practicing engineers. Prior to the twentieth century, there were of course also sev- eral key contributions. Perhaps the firsttime the study of control systemsattracted serious scientificattention was towardthe end of the eighteenth century, following James Watts's invention of the governor in 1788, which was designed to regulate the speed of the rotary steam engine. A related work on governors by Huygens! actuallypredated that of Wattsby about a century; Huygensinventedthe centrifugalgovernoras a meansof regulat- ing a clock, which was adapted for windmills and water wheels in the Netherlandsas early as the seventeenthcentury.Feedback played an important role in all these inventions,and soon it was realized and widely acknowledged that it is a concept that lies at the foundation of any successful control design. However, to make effective use of feedback, there was a need for a careful mathematicalstudy of its impact on control design.James Clerk Maxwell was the first to realize this need and to respond to it by developingin his now famous paperv'consideredby many to be the starting point of the scientific approach to control research, mathematicalmodelsfor variousgovernormechanismsbasedon linear differentialequations. He worked out in his paper a com- pletetheoryof stabilityfor constantcoefficientlineardifferential equations up to fourth order, and obtained some conditions for stabilityof fifth-order systems.Around the same time, and inde- pendently of Maxwell,a Russian engineer,Vyshnegradskii, had recognized the importance of control in industrial applications, IC. Huygens, "Horologii oscillatorii," Part 5, Paris, 1673. 2J.C. Maxwell, "Ongovemors," Proc.RoyalSoc.London,16, 1868,pp. 270- 283. and the need for developing a sound theory.' Where Maxwell left off was then picked up by Edward John Routh," and inde- pendently by Adolf Hurwitz.l who came up with what is known todayas theRouth-Hurwitz stabilitycriteria,solvingcompletely the problem of stability of constant coefficient linear differen- tial equations of any finite order. At about the same time, and as the nineteenth century was coming to a close, another trend- settingdevelopmenttook place, again in the area of stability, but this time for nonlinear dynamical systems. Motivatedby prob- lems that arise in astronomy in connection with the motion of the planets, a topic studied earlier by Henry Poincare," among others,AleksandrMikhailovichLyapunovdevelopedin his doc- toral thesis in Russia a new approach for testing the stability of the equilibrium of a system described by nonlinear ordinary differential equations, known today as the Second Method of Lyapunov. Hence, there was quite a bit of accumulatedactivityin control at the beginning of the twentieth century. But what this century, and particularly its secondhalf, deliveredwas somethingdiffer- ent in terms of both content and sheer volume of diverse con- tributions.The incessant growth caused by an explosion of new fresh ideas, and drivenby numerous applicationsfrom different domains, brought this activity to unprecedented levels. As we are coming to the close of this century,we thought that it would be useful to reflect back and ask the questions: What have been the major results of this century in control? What have been the greatest hits in control? How has control theory evolved since the times of Maxwell, Routh, Hurwitz, and Lyapunov (among others)?There is of course no unique way of answeringall these questions, but one possible way is to collect under one cover, 3J. Vyshnegradskii, "Sur la theorie generale des regulateurs (On the general theory of control)," Compt. Rend. Acad. Sci. Paris, 83,1876, pp. 318-321. 4E.J. Routh, A Treatise on theStabilityofa GivenStateofMotion,Macmillan (London), 1877. 5A. Hurwitz, "Uber the Bedingungen unter welschen eine Gleichung nur Wurzeln mit negativen reelen Teilen besitzt (On the conditions under which an equation has only roots with negative real parts)," Mathematische Annalen,46, 1895, pp. 273-284. 6H. Poincare, Les MethodesNouvellesde la Mechanique Celeste (The New Methods of the Cellestial Mechanics), Vol. 1, Gauthier-Villars (Paris), 1892. vii
viii PREFACE and as a representative of the major research developments and accomplishments in control in this century, some key papers that have made major impact on the field, along with some introduc- tory material for each. These considerations have led to the present volume, which contains twenty-five carefully selected seminal papers covering the period 1932-1981, and begins with Harry Nyquist's famous "Regeneration Theory" paper, which introduced the so-called Nyquist criterion (still a versatile tool for control engineers) and laid the foundations of a frequency-domain approach to stabil- ity analysis of linear control systems. The volume ends with the 1981 paper by George Zames, which marks the beginning of the "robustness" era in control-an era that we are leaving to the coming generations to evaluate (perhaps by the middle of the twenty-first century), along with other exciting developments the control field has experienced for the past two decades, and will undoubtedly continue to do so in the next century. This volume was prepared under the auspices of the IEEE Control Systems Society, by an Editorial Board consisting of twelve members, namely: Brian D.O. Anderson, Karl J. Astrom, John Baillieul, Tamer Basar (Chair), Bruce A. Francis, Alberto Isidori, Petar ~ Kokotovic, Huibert Kwakernaak, William J. Levine, Lennart Ljung, David Q. Mayne, and Jan C. Willems. Based on nominations received in response to solicitations that appeared in the IEEE Control Systems Magazine, and in the E-LETTER on Systems, Control, and Signal Processing (based in the Netherlands), and nominations generated by the Board members, and after several rounds of voting, the Board unan- imously agreed on the selection of the twenty-five papers in- cluded in this volume. The preambles to each paper were written by the Board members (in some cases jointly), with the primary author(s) in each case identified by his (their) initials. The twenty-five papers included in this volume cover a broad spectrum of major developments in control theory in the twen- tieth century, but still the volume should not be viewed as pro- viding an exhaustive coverage of all topical areas of control, as this has not been a criterion set by the Editorial Board in their selection of the papers. The focus here has been the papers selected rather than the areas of research in control. Still, we be- lieve that the selected papers clearly outline the path which the control discipline has followed during its rapid growth from the 1930s through the 1980s. To help the reader in this journey, in the preambles to individual papers we have discussed develop- ments in areas neighboring the topic of a particular paper, so as to place its contributions and impact in proper perspective, and to maintain continuity in the flow of ideas from one topic to another. What is the path we can trace from the chronologically or- dered papers in the volume? First comes the basic feedback the- ory, as represented by the first three papers, by Nyquist, Black, and Bode, which was the outcome of research conducted at the Bell Laboratories in the 1930s driven by the need to de- velop electronic feedback amplifiers for long telephone lines. To this was added later, as a practical tool, the root locus method of Evans. In the 1940s, Wiener's work on prediction, filtering, and smoothing for time series, impacted control in many ways, fostering further developments not only in filtering theory but also in control design. The "Sputnik effect" and the ensuing space research propelled the development of a mathematically advanced optimal control theory in the late 1950s and early 1960s, with dynamic programming, maximum principle, and the LQ regulator design (with its associated concepts of con- trollability and observability) as its centerpieces, as represented in the works of Bellman, Pontryagin, and Kalman included in this volume. Also developed during this period were compu- tational techniques to make dynamic programming and maxi- mum principle practicable, as represented by the paper of Bryson and Denham. During the same period, efforts were intensified to develop an applicable stability theory of nonlinear feedback systems in the absolute stability framework, represented by the papers of Popov, Yakubovich, and Zames and, as extensions and refinements of Lyapunov concepts, in the papers by LaSalle and Willems. The need to operate in the presence of noise and other disturbances was also recognized in the early 1960s, as shown in papers by Kalman and Feldbaum. Then came the establishment of the precise relationship between input-output descriptions and state-space representations of linear systems, with its multifold ramifications, as presented in the 1963 paper by Kalman, and the establishment of a geometric theory for linear systems, in the paper by Wonham and Morse, with the introduction of the novel concept of controlled invariance, which found applications in much broader domains (than linear systems) later. We see in the 1970s the emergence of a nonlinear system theory, with as- sociated richer concepts of controllability and observability, as shown in the papers by Brockett, Sussmann and Jurdjevic, and Hermann and Krener. Adaptive control is another area where a comprehensive theory started emerging in the 1970s, where the important notion of "self-tuning" was introduced (in the paper by Astrom and Wittenmark), and methodologies were developed for establishing the convergence of adaptive control algorithms, as well as recursive stochastic algorithms, as presented in the papers by Goodwin, Ramadge and Caines, and Ljung. Robust control is yet another topic that gained steam in the late 1970s, with the 1981 paper by Zames (the last paper in this volume) marking the beginning of some intense activity in this domain. Of course, there were many other accomplishments in control during this period, than those represented by the twenty-five pa- pers selected, and it is hoped that the preambles to the papers will convey to the reader this richness in the field. I hope that the reader will enjoy this journey, and will de- velop a real sense of the evolution of the control field during the fifty-year period, 1932-1981, through the milestone accom- plishments embodied in these twenty-five seminal papers and further discussed in the introductory material provided by our Editorial Board. It is our hope that the volume will be a valu- able resource in the twenty-first century (and beyond) especially for young control researchers and engineers, and will be instru- mental in the realization of an even more explosive century for control theory and its applications. Tamer Basar University ofIllinois at Urbana-Champaign
分享到:
收藏