Download Full Version ebook - Visit ebookmeta.

com

Optimal and Robust Control: Advanced Topics with

MATLAB®, 2nd Edition Fortuna

https://ebookmeta.com/product/optimal-and-robust-control-
advanced-topics-with-matlab-2nd-edition-fortuna/

OR CLICK HERE

DOWLOAD NOW

Discover More Ebook - Explore Now at ebookmeta.com

 Instant digital products (PDF, ePub, MOBI) ready for you
Download now and discover formats that fit your needs...

Start reading on any device today!

Automatic Control Systems: With MATLAB, 2nd Edition S.

Palani

https://ebookmeta.com/product/automatic-control-systems-with-
matlab-2nd-edition-s-palani/

ebookmeta.com

Control Engineering MATLAB Exercises Advanced Textbooks in

Control and Signal Processing Keviczky László

https://ebookmeta.com/product/control-engineering-matlab-exercises-
advanced-textbooks-in-control-and-signal-processing-keviczky-laszlo/

ebookmeta.com

Optimal Control of ODEs and DAEs 2nd Edition Matthias

Gerdts

https://ebookmeta.com/product/optimal-control-of-odes-and-daes-2nd-
edition-matthias-gerdts/

ebookmeta.com

Computer Organization and Architecture: Designing for

performance, Global Edition, 11th Edition William
Stallings
https://ebookmeta.com/product/computer-organization-and-architecture-
designing-for-performance-global-edition-11th-edition-william-
stallings/
ebookmeta.com
The Routledge Companion to Gender, Media and Violence 1st
Edition Karen Boyle

https://ebookmeta.com/product/the-routledge-companion-to-gender-media-
and-violence-1st-edition-karen-boyle/

ebookmeta.com

How to Design a Life Worth Smiling About 1st Edition

Darryl Davis

https://ebookmeta.com/product/how-to-design-a-life-worth-smiling-
about-1st-edition-darryl-davis/

ebookmeta.com

These Thin Lines 1st Edition Milena Mckay

https://ebookmeta.com/product/these-thin-lines-1st-edition-milena-
mckay/

ebookmeta.com

Decolonising Indian Education: National Education Policy

2020, Ars Artium: An International Refereed Research
Journal of English Studies and Culture 2023rd Edition
Susheel Kumar Sharma (Author)
https://ebookmeta.com/product/decolonising-indian-education-national-
education-policy-2020-ars-artium-an-international-refereed-research-
journal-of-english-studies-and-culture-2023rd-edition-susheel-kumar-
sharma-author/
ebookmeta.com

Essentials of Pediatric Urology 3rd Edition Duncan T

Wilcox David F M Thomas

https://ebookmeta.com/product/essentials-of-pediatric-urology-3rd-
edition-duncan-t-wilcox-david-f-m-thomas/

ebookmeta.com
Cardiovascular Pathology 5th Edition L. Maximilian Buja

https://ebookmeta.com/product/cardiovascular-pathology-5th-edition-l-
maximilian-buja/

ebookmeta.com
Optimal and Robust
Control
 Optimal and Robust
Control
Advanced Topics with MATLAB®
Second Edition

Luigi Fortuna
Mattia Frasca
Arturo Buscarino
MATLAB® is a trademark of The MathWorks, Inc. and is used with permission. The MathWorks
does not warrant the accuracy of the text or exercises in this book. This book’s use or discussion of
MATLAB® software or related products does not constitute endorsement or sponsorship by The
MathWorks of a particular pedagogical approach or particular use of the MATLAB® software.

Second edition published 2022

by CRC Press
6000 Broken Sound Parkway NW, Suite 300, Boca Raton, FL 33487-2742

and by CRC Press

2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN

© 2022 Taylor & Francis Group, LLC

First edition published by CRC Press 2012

CRC Press is an imprint of Taylor & Francis Group, LLC

Reasonable efforts have been made to publish reliable data and information, but the author and pub-
lisher cannot assume responsibility for the validity of all materials or the consequences of their use.
The authors and publishers have attempted to trace the copyright holders of all material reproduced
in this publication and apologize to copyright holders if permission to publish in this form has not
been obtained. If any copyright material has not been acknowledged please write and let us know so
we may rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced,
transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or
hereafter invented, including photocopying, microfilming, and recording, or in any information stor-
age or retrieval system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, access www.copyright.
com or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA
01923, 978-750-8400. For works that are not available on CCC please contact mpkbookspermis-
sions@tandf.co.uk

Trademark notice: Product or corporate names may be trademarks or registered trademarks and are
used only for identification and explanation without intent to infringe.

Library of Congress Cataloging‑in‑Publication Data

ISBN: 978-1-032-05300-4 (hbk)

ISBN: 978-1-032-05301-1 (pbk)
ISBN: 978-1-003-19692-1 (ebk)
ISBN: 978-1-032-15155-7 (ebk+)

DOI: 10.1201/9781003196921

Publisher’s note: This book has been prepared from camera-ready copy provided by the authors.

Access the Support Material: https://www.routledge.com/Optimal-and-Robust-Control-Advanced-

Topics-with-MATLAB/Fortuna-Frasca/p/book/9781032053004

eResources are available for this title at: https://www.crcpress.com/9781032053004

Dedicated to our wives
Contents

Preface xiii

Symbol List xv

1 Modelling of Uncertain Systems and the Robust Control

Problem 1

1.1 Uncertainty and Robust Control . . . . . . . . . . . . . . . 1

1.2 The Essential Chronology of Major Findings in Robust Con-
trol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

2 Fundamentals of Stability 11

2.1 Lyapunov Criteria . . . . . . . . . . . . . . . . . . . . . . . 11

2.2 Positive Definite Matrices . . . . . . . . . . . . . . . . . . . 13
2.3 Lyapunov Theory for Linear Time-Invariant Systems . . . . 16
2.4 Lyapunov Equations . . . . . . . . . . . . . . . . . . . . . . 20
2.5 Stability with Uncertainty . . . . . . . . . . . . . . . . . . . 23
2.6 Further Results on the Lyapunov Theory . . . . . . . . . . 26
2.6.1 Hystorical Notes . . . . . . . . . . . . . . . . . . . . . 26
2.6.2 Lyapunov Stability . . . . . . . . . . . . . . . . . . . 27
2.7 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

3 Kalman Canonical Decomposition 35

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2 Controllability Canonical Partition . . . . . . . . . . . . . . 37
3.3 Observability Canonical Partition . . . . . . . . . . . . . . . 39
3.4 General Partition . . . . . . . . . . . . . . . . . . . . . . . . 40
3.5 Remarks on Kalman Decomposition . . . . . . . . . . . . . 47
3.6 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

4 Singular Value Decomposition 51

4.1 Singular Values of a Matrix . . . . . . . . . . . . . . . . . . 51

4.2 Spectral Norm and Condition Number of a Matrix . . . . . 53
4.3 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

vii
viii Contents

5 Open-loop Balanced Realization 59

5.1 Controllability and Observability Gramians . . . . . . . . . 59

5.2 Principal Component Analysis . . . . . . . . . . . . . . . . 63
5.3 Principal Component Analysis Applied to Linear Systems . 64
5.4 State Transformations of Gramians . . . . . . . . . . . . . . 66
5.5 Singular Values of Linear Time-invariant Systems . . . . . . 68
5.6 Computing the Open-loop Balanced Realization . . . . . . 69
5.7 Balanced Realization for Discrete-time Linear Systems . . . 73
5.8 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

6 Reduced Order Models and Symmetric Systems 77

6.1 Reduced Order Models Based on the Open-loop Balanced Re-

alization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
6.1.1 Direct Truncation Method . . . . . . . . . . . . . . . 79
6.1.2 Singular Perturbation Method . . . . . . . . . . . . . 81
6.2 Reduced Order Model Exercises . . . . . . . . . . . . . . . . 82
6.3 Symmetric Systems . . . . . . . . . . . . . . . . . . . . . . . 86
6.3.1 Reduced Order Models for SISO Systems . . . . . . . 86
6.3.2 Properties of Symmetric Systems . . . . . . . . . . . 88
6.3.3 The Cross-gramian Matrix . . . . . . . . . . . . . . . 90
6.3.4 Relations Between Wc2 , Wo2 and Wco . . . . . . . . . 90
6.3.5 Open-loop Parameterization . . . . . . . . . . . . . . 96
6.3.6 Relation Between the Cauchy Index and the Hankel
Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . 98
6.3.7 Singular Values for a FIR Filter . . . . . . . . . . . . 99
6.3.8 Singular Values of All-pass Systems . . . . . . . . . . 102
6.4 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

7 Variational Calculus and Linear Quadratic Optimal Control 107

7.1 Variational Calculus: An Introduction . . . . . . . . . . . . 108

7.2 The Lagrange Method . . . . . . . . . . . . . . . . . . . . . 110
7.3 Towards Optimal Control . . . . . . . . . . . . . . . . . . . 112
7.4 LQR Optimal Control . . . . . . . . . . . . . . . . . . . . . 117
7.5 Hamiltonian Matrices . . . . . . . . . . . . . . . . . . . . . 124
7.6 Solving the Riccati Equation via the Hamiltonian Matrix . 127
7.7 The Control Algebraic Riccati Equation . . . . . . . . . . . 127
7.8 Optimal Control for SISO Systems . . . . . . . . . . . . . . 129
7.9 Linear Quadratic Regulator with Cross-weighted Cost . . . 134
7.10 Finite-horizon Linear Quadratic Regulator . . . . . . . . . . 135
7.11 Optimal Control for Discrete-time Linear Systems . . . . . 136
7.12 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
Contents ix

8 Closed-loop Balanced Realization 139

8.1 Synthesis of a Compensator for High-Order Systems . . . . 139

8.2 Filtering Algebraic Riccati Equation . . . . . . . . . . . . . 140
8.3 Computing the Closed-loop Balanced Realization . . . . . . 142
8.4 Procedure for Closed-loop Balanced Realization . . . . . . . 144
8.5 Reduced Order Models Based on Closed-loop Balanced Real-
ization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
8.6 Closed-loop Balanced Realization for Symmetric Systems . 149
8.7 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150

9 Positive-real, Bounded-real and Negative-imaginary Systems 153

9.1 Passive Systems . . . . . . . . . . . . . . . . . . . . . . . . 154

9.1.1 Passivity in the Frequency Domain . . . . . . . . . . 154
9.1.2 Passivity in the Time Domain . . . . . . . . . . . . . 159
9.1.3 Factorizing Positive-real Functions . . . . . . . . . . 160
9.1.4 Passive Reduced Order Models . . . . . . . . . . . . 161
9.1.5 Energy Considerations Connected to the Positive-real
Lemma . . . . . . . . . . . . . . . . . . . . . . . . . . 161
9.1.6 Closed-loop Stability and Positive-real Systems . . . 162
9.1.7 Optimal Gain for Loss-less Systems . . . . . . . . . . 163
9.2 Circuit Implementation of Positive-real Systems . . . . . . . 165
9.3 Bounded-real Systems . . . . . . . . . . . . . . . . . . . . . 166
9.3.1 Properties of Bounded-real Systems . . . . . . . . . . 169
9.3.2 Bounded-real Reduced Order Models . . . . . . . . . 170
9.4 Relationship Between Passive and Bounded-real Systems . . 170
9.5 Negative-imaginary Systems . . . . . . . . . . . . . . . . . . 171
9.5.1 Characterization of Negative-imaginary Systems in the
Frequency Domain . . . . . . . . . . . . . . . . . . . 171
9.5.2 Characterization of Negative-imaginary Systems in the
Time Domain . . . . . . . . . . . . . . . . . . . . . . 175
9.5.3 Closed-loop Stability and Negative-imaginary Systems 176
9.6 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178

10 Enforcing the Positive-real or the Negative-imaginary

Property in a Linear Model 181

10.1 Why to Enforce the Positive-real and Negative-Imaginary

Property in a Linear Model . . . . . . . . . . . . . . . . . . 181
10.2 Passification . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
10.3 Forward Action to make a System Negative-Imaginary . . . 187
10.3.1 The SISO Case . . . . . . . . . . . . . . . . . . . . . 187
10.3.2 The MIMO Case . . . . . . . . . . . . . . . . . . . . 191
10.4 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
x Contents

11 H∞ Linear Control 195

11.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 195

11.2 Solution of the H∞ Linear Control Problem . . . . . . . . . 197
11.3 The H∞ Linear Control and the Uncertainty Problem . . . 205
11.4 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208

12 Linear Matrix Inequalities for Optimal and Robust Control 209

12.1 Definition and Properties of LMI . . . . . . . . . . . . . . . 210

12.2 LMI Problems . . . . . . . . . . . . . . . . . . . . . . . . . 211
12.2.1 Feasibility Problem . . . . . . . . . . . . . . . . . . . 211
12.2.2 Linear Objective Minimization Problem . . . . . . . 212
12.2.3 Generalized Eigenvalue Minimization Problem . . . . 212
12.3 Formulation of Control Problems in LMI Terms . . . . . . . 213
12.3.1 Stability . . . . . . . . . . . . . . . . . . . . . . . . . 213
12.3.2 Closed-loop Stability . . . . . . . . . . . . . . . . . . 213
12.3.3 Simultaneous Stabilizability . . . . . . . . . . . . . . 214
12.3.4 Positive-real Lemma . . . . . . . . . . . . . . . . . . 214
12.3.5 Bounded-real Lemma . . . . . . . . . . . . . . . . . . 214
12.3.6 Calculating the H∞ Norm Through LMI . . . . . . . 215
12.4 Solving a LMI Problem . . . . . . . . . . . . . . . . . . . . 215
12.5 LMI Problem for Simultaneous Stabilizability . . . . . . . . 218
12.6 Solving Algebraic Riccati Equations Through LMI . . . . . 221
12.7 Computation of Gramians Through LMI . . . . . . . . . . . 223
12.8 Computation of the Hankel Norm Through LMI . . . . . . 224
12.9 H∞ Control . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
12.10 Multiobjective Control . . . . . . . . . . . . . . . . . . . . . 228
12.11 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235

13 The Class of Stabilizing Controllers 237

13.1 Parameterization of Stabilizing Controllers for Stable

Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
13.2 Parameterization of Stabilizing Controllers for Unstable
Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
13.3 Parameterization of Stable Controllers . . . . . . . . . . . . 242
13.4 Simultaneous Stabilizability of Two Systems . . . . . . . . . 245
13.5 Coprime Factorizations for MIMO Systems and Unitary
Factorization . . . . . . . . . . . . . . . . . . . . . . . . . . 245
13.6 Parameterization in Presence of Uncertainty . . . . . . . . . 247
13.7 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
Contents xi

14 Formulation and Solution of Matrix Algebraic Problems

through Optimization Problems 253

14.1 Solutions of Matrix Algebra Problems Using Dynamical Sys-

tems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
14.1.1 Problem 1: Inverse of a Matrix . . . . . . . . . . . . 254
14.1.2 Problem 2: Eigenvalues of a Matrix . . . . . . . . . . 256
14.1.3 Problem 3: Eigenvectors of a Symmetric Positive Def-
inite Matrix . . . . . . . . . . . . . . . . . . . . . . . 258
14.1.4 Problem 4: Observability and Controllability Gramian 259
14.2 Computation of the Open-loop Balanced Representation via
the Dynamical System Approach . . . . . . . . . . . . . . . 260
14.3 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . 263
14.4 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264

15 Time-delay Systems 265

15.1 Modeling Systems with Time-delays . . . . . . . . . . . . . 265

15.2 Basic Principles of Time-delay Systems . . . . . . . . . . . 266
15.3 Stability of Time-delay Systems . . . . . . . . . . . . . . . . 269
15.4 Stability of Time-delay Systems with q = 1 . . . . . . . . . 270
15.5 Direct Method . . . . . . . . . . . . . . . . . . . . . . . . . 275
15.6 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281

Recommended Essential References 283

Appendix A. Norms 291

Appendix B. Algebraic Riccati Equations 295

Appendix C. Invariance Under Frequency Transformations 299

Index 303
Preface

The main aim of this book is to provide for undergraduate and graduate
students, as well as researchers, who already possess the main concepts of
automatic control and system analysis, a self-contained resource collecting
advanced techniques for linear system theory and control design. Selected
theoretical backgrounds are also presented in the book, together with many
numerical exercises and MATLABr examples.
We intend to offer a complete and easy-to-read handbook of advanced
topics in automatic control, including techniques such as the Linear Quadratic
Regulator (LQR) and H∞ control. Large emphasis is also given to Linear
Matrix Inequalities (LMIs) with the purpose of demonstrating their use as a
unifying tool for system analysis and control design.
In presenting the different approaches to control design, the books explic-
itly takes into account the problem of the robustness of the obtained closed-
loop control. Robustness, in fact, represents the capability of a control system
to guarantee the stability in the presence of uncertainty, due to the model itself
or to the use of approximated models, and as such is deemed as particularly
important in view of the practical implementation of the control techniques.
Many books on LQR control and H∞ control have been proposed since
1980. The LMI technique has become well-known in the control community,
and MATLABr toolboxes to solve advanced control problems have been de-
veloped. However, these subjects are often presented for a specialist audience
in materials that are excellent resources for researchers and PhD students.
This book, on the contrary, is oriented to illustrate these topics in an easy
and concise way, using a language suitable for students, yet maintaining the
necessary mathematical rigor.
This book is, therefore, a compendium of many ordered subjects. For spe-
cific proofs, the reader is often referred to the proposed literature. Many ex-
amples and MATLABr based exercises are included here to assist the reader
in understanding the proposed methods. The book can be considered as a
palimpsest of advanced modern topics in automatic control, including an ad-
vanced set of analytical examples and MATLABr exercises. The topics in-
cluded in the book are mainly illustrated with reference to continuous-time
linear systems, even if some results for discrete-time systems are briefly re-
called.
The book is organized into chapters structured as follows. The first chapter
is an introduction to advanced control, the second discusses some fundamental
concepts on stability and provides the tools for studying uncertain systems.

xiii
xiv Preface

The third presents the Kalman decomposition. The fourth chapter is on sin-
gular value decomposition of a matrix, given the importance of numerical
techniques for systems analysis. The fifth and sixth chapters are on open-loop
balanced realization and reduced order models. The seventh chapter presents
the essential aspects of variational calculus and optimal control and the eighth
illustrates closed-loop balancing. The properties of positive-real, bounded-real
and negative-imaginary systems are the subject of the ninth and tenth chap-
ter. In the eleventh and twelfth chapter, the essential aspects of H∞ control
and LMI techniques commonly used in control systems design are dealt with.
The thirteenth chapter is devoted to discuss the class of stabilizing controllers.
The fourteenth chapter reviews some of the problems already discussed in the
book by introducing an approach based on the steady-state solution of a non-
linear dynamical system. Finally, the fifteenth chapter briefly discusses some
fundamental aspects of time-delay systems. The book also includes numerous
examples and exercises, considered indispensable for learning the methodology
of the topics dealt with, and a list of essential references.
This book is targeted at electrical, electronic, computer science, space and
automation engineers interested in advanced topics on automatic control. Me-
chanical engineers as well as engineers from other fields may also be inter-
ested in the topics of the book. The contents of the book can be learned
autonomously by the reader in less than a semester.

For MATLABr and Simulinkr product information, please contact:

The MathWorks, Inc.

3 Apple Hill Drive
Natick, MA 01760-2098 USA
Tel: 508-647-7000
Fax: 508-647-7001
Email: info@mathworks.com
Web: www.mathworks.com
Symbol List

Symbol Description

R the set of real numbers square matrices, i.e.,

C the set of complex num- A ∗ B = {aij bij })
bers det (A) determinant of matrix
Rn real vectors of n compo- A
nents trace(A) trace of matrix A
Cn complex vectors of n P>0 (semi-defined) defined
components positive matrix
Rm×n real matrices of dimen- P≥0 semi-defined positive
sions m × n matrix
Cm×n complex matrices of di- P<0 (semi-defined) defined
mensions m × n negative matrix
<e (x) real part a of a complex P≤0 semi-defined negative
number x = a + jb matrix
=m (x) imaginary part b of a λi (A) i-th eigenvalue of ma-
complex number x = trix A
a + jb σi (A) i-th singular value of
|x| absolute value of a real matrix A
number or the modulus ρ(A) spectral radius (maxi-
of a complex number mum eigenvalue) of ma-
kxk norm of vector x trix A
aij coefficient of line i and S(A, B, C, D) linear dynamical sys-
column j of matrix A tem ẋ = Ax + Bu, y =
sup superior extreme of a Cx + Du
set Mc controllability matrix of
inf inferior extreme of a set a system
I identity matrix of op- Mo observability matrix of
portune dimensions a system
AT transpose of a matrix σi i-th singular value of a
A∗ conjugate transpose of system
A ∈ Cn×n µi i-th characteristic value
A⊗B Kronecker product of a system
A∗B Hadamard product RIC(H) Riccati equation solu-
(component by com- tion associated with
ponent product of two Hamiltonian matrix H

xv
xvi Symbol List

SISO single input single MIMO multi input multi

output system output system
1
Modelling of Uncertain Systems and the
Robust Control Problem

CONTENTS
1.1 Uncertainty and Robust Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 The Essential Chronology of Major Findings in Robust Control 9

In this chapter the main concepts related to robust control are introduced.
In particular, the key notions of robustness and uncertainty are presented. In
the formalization of uncertainty, both the structured and unstructured cases
are dealt with. In this preliminary part, the representation of a linear time
invariant system in term of the realization matrix is also given. Some examples
of uncertainty generated with MATLAB examples are reported. The essential
chronology of the recent history of robust control is also outlined.

1.1 Uncertainty and Robust Control

In control system design, the primary requirement is the asymptotic stability
of controlled systems. In automatic control, this is guaranteed through the de-
sign of an appropriate controller based on the nominal model of the process.
However, in reality there may be several sources of uncertainty which make the
nominal model inaccurate. Robust control deals explicitly with system uncer-
tainties which account for the differences between the real model and nominal
model. Robust control guarantees controlled system performance (primarily,
asymptotic stability) when there are uncertainties.
The robust control issue can be summarized as: given a nominal process
with acceptable interval values for perturbances, a controller should provide
satisfactory performance in a closed-loop system for all the processes with the
“acceptable” perturbances.
As regards stability, the requisites of a robust control system should ensure:
1. Closed-loop stability under nominal conditions;
2. Closed-loop stability although there are uncertainties in the model.

DOI: 10.1201/9781003196921-1 1
2 Optimal and Robust Control: Advanced Topics with MATLAB r

FIGURE 1.1
Unitary feedback control scheme.

As clarified below, uncertainties come in various forms, but, notwithstand-

ing, an acceptable value range for possible uncertainties must be hypothesized
within which interval control system performance is guaranteed. This can also
provide a way to establish if one controller is more robust than another: the
greater the range of acceptable values for uncertainties, the more the controller
can be considered robust.
The automatic control theory refers to the feedback control model repre-
sented in Figure 1.1 (unitary feedback) or in Figure 1.2 (non-unitary feedback).
P represents the process and generally no assumptions are made about its lin-
earity, assuming only it is a time-invariant system. C indicates the controller
to be designed. The main issue is to design C, so the closed-loop system is
asymptotically stable. Furthermore, in addition to this basic specification, the
control system can be required to respond to other criteria. In the case of
robust control, another important factor must be accounted for: the process
model may have uncertainties.
For example, consider a linear process, described by transfer function
1
P (s) = s2 (s+1) . Who can say for sure that the pole of this system is ex-
actly −1? And, in the same way, on what basis could the system gain be
considered perfectly unitary? For this reason, in the theory of robust control,
g1
the system is indicated by P (s) = s2 (s+p 1)
, inputting parameters which ac-
count for system uncertainties. g1 = 1 and p1 = 1 should be considered the
nominal values of the parameters that, however, in practice may eventually
have different values, with a variation range that is usually known. This type
of uncertainty concerns parameter values which characterize the system and
so is defined parametric uncertainty or structured uncertainty. Examples of
parametric uncertainty can be seen in daily life. Think of any size measure-
ment (e.g., the length of a table): the measurement depends on the accuracy
of the measuring instrument.
There are also other causes of parametric uncertainty. Parameter values
may also vary depending on the operating conditions of the system. Think of
a resistor heated by the Joule effect: electrical resistivity increases, modifying
the value of that resistance. Or think of an airplane, as the fuel is consumed,
the plane total mass decreases.
Obviously, if parameter values change (even if in a predictable range),
control system performance drops off compared to a nominal one (in fact,
Modelling of Uncertain Systems and the Robust Control Problem 3

FIGURE 1.2
Feedback control scheme.

the design was related to nominal parameter values). The first goal of robust
control is to ensure that, even with changing parameter values, the system
asymptotic stability is guaranteed.
Above, there was an example of parametric uncertainty about the coeffi-
cients of the transfer function of a linear time-invariant system. Now consider
a nonlinear time-invariant system (with one input and output, for example),
described by the equations:

ẋ = f(x) + g(x)u
(1.1)
y = h(x)u
with x ∈ Rn (state variables), u ∈ R (system input) and y ∈ R (system
output), g : Rn → Rn , f : Rn → Rn and h : Rn → Rn . With robust control,
parametric uncertainty in the model is highlighted by using the parameters
α, β and γ:

ẋ = f(x, α) + g(x, β)u

(1.2)
y = h(x, γ)u
In this case, ᾱ, β̄ and γ̄ indicate the nominal values of the parameters
α, β and γ. Robust control for system (1.2) means designing C to guarantee
closed-loop asymptotic stability, given the variation spans of parameters α, β
and γ (generally, vectors of arbitrary size).
Similarly, one can consider linear systems in state-space form or other
nonlinear models. Especially, for a linear time-invariant system in state-space
form:

ẋ = Ax + Bu
(1.3)
y = Cx + Du
with x ∈ Rn , u ∈ Rm , y ∈ Rp , A ∈ Rn×n , B ∈ Rn×m , C ∈ Rp×n , D ∈ Rp×m ,
coefficients of A, B, C and D vary in certain intervals. What is required for
robust control, once the controller is designed for nominal parameter values,
is that asymptotic stability is guaranteed even when the parameters are not
nominal.
4 Optimal and Robust Control: Advanced Topics with MATLAB r

The linear time-invariant systems can also be described in a more compact

form with the realization matrix :
 
A B
R= (1.4)
C D
If the number of inputs equals the number of system outputs (p = m)
then matrix R is square (R ∈ R(n+m)×(n+m) ). The eigenvalues of this matrix
can easily be shown not to depend on the reference system. In fact, a state
transformation leads to:

R̃ = T−1 RT (1.5)
which is clearly a similarity relation.
To verify if a system is minimal, it is possible to calculate the eigenvalues of
matrix R which are represented by λ̄1 , λ̄2 , . . . , λ̄n+m , whereas the eigenvalues
of A by λ1 , λ2 , . . . , λn . These quantities are system invariants, that is, they do
not depend on the state-space representation. The system is minimal if none
of eigenvalues of A coincide with those of R.
There is another type of uncertainty which is more difficult to deal with
than parametric uncertainty. Called structural uncertainty or unstructured un-
certainty, it concerns the model structure. Consider again the example of the
1
process described by P = s2 (s+1) , structural uncertainty takes into account
the possibility that the modelling did not account perhaps for an additional
pole and an additional zero in the process transfer function:
( αs0 + 1)
P =
s2 (s + 1)( αs + 1)

So, structural uncertainty derives from incorrect modelling, which possibly

overlooked some dynamics of the real process. Neglecting some dynamics in
modelling is actually very common and can have consequences for closed-loop
stability.
Example 1.1
1
Consider system P (s) = (s+1)2
. The transfer function of the closed-loop system
Y (s) C(s)P (s)
(Fig. 1.1) is given by M (s) = R(s) = 1+C(s)P (s) . Using a simple proportional controller
k
C(s) = k, you obtain M (s) = (s+1) 2 +k and the closed-loop system is asymptotically
stable ∀k > 0.
For example, if C(s) = 100, notice that this controller is robust to parametric variations
of position of the pole of P (s) with double multiplicity. In fact, if instead of p = −1,
it was p = −α, the controller C(s) = 100 would continue to guarantee closed-loop
asymptotic stability to a large value of the parameter (α > 0).
In the case of structural uncertainty, the scenario is different. Suppose that there is an
1
uncertainty, say on system order, so that P (s) = (s+1) 3 . The characteristic closed-loop

equation is given by (s + 1)3 + k = 0, i.e., s3 + 3s2 + 3s + 1 + k = 0. Applying the Routh

criterion, we note that in this case the system is asymptotically stable if 0 < k < 8.
1
Controller C(s) = 100, designed for the nominal system P (s) = (s+1) 2 , no longer guar-
antees asymptotic stability because of the structural uncertainty of an additional pole
Modelling of Uncertain Systems and the Robust Control Problem 5

at −1. The conclusion is that the consequences of uncertainty on system order, having
overlooked some dynamics (which could also be a parasite dynamic which triggers in
certain conditions), can be very important.

Defining parametric uncertainty is done by expressing a parameter varia-

tion range. For example, given a linear time-invariant system described by its
transfer function
N (s) b0 sm + b1 sm−1 + . . . + bm−1 s + bm
P (s) = =
D(s) sn + a1 sn−1 + . . . + an−1 s + an

once the minimum and maximum values of the various coefficients are as-
signed, parametric uncertainty is completely characterized:

am
1 ≤ a1 ≤ a1
M

...
am
n ≤ an ≤ an
M
(1.6)
bm
0 ≤ b0 ≤ bM
0
...
bm
m ≤ bm ≤ bm
M

If there is structural uncertainty we assume P (s) = P0 (s) + ∆P (s), where

P0 (s) represents the nominal model and ∆P (s) represents uncertainty. This
structural uncertainty is additive. Structural uncertainty can also be multi-
plicative: P (s) = P0 (s)∆P (s). Structural uncertainty is measured by a norm
of the transfer function or of the transfer matrix. We will see how to define
such a norm in Chapter 9.
The greater the required robustness, the greater the precision required to
model the process. A particularly important role in robust control is test-
ing. The popularity of personal computers now makes it possible to perform
tests through numerical simulations even when the robust control problem
has no analytical solutions. Numerical simulations help verify operation and
controller performance in the variation interval indicated by the uncertainty.
The most critical issue is structural uncertainty. Suppose we consider a
horizontal beam fixed at one end and subject to a stimulus at the other. This
system has distributed parameters and could be modelled by an infinite series
of mass-spring systems. The free-end deflection has infinite
P∞ modes which can
be described with a transfer function of type G(s) = i=0 Gi (s), where Gi (s)
is the transfer function describing the mode associated with the n-th section.
However, this approach is not so easy to carry out and so an approximation
G(s) = G0 (s) + ∆G(s) is used which takes into account the most important
modes (modelled by G0 (s)). Approximation is necessary for the design of a
non-distributed controller (note too that additive uncertainty helps model
the high frequency dynamic neglected by the approximation). In these cases
only the dominant part is considered, but it is essential that the controller is
robust enough for any uncertainty in the dynamics not described in the model.
This introduces another very important issue treated below: how to figure out
6 Optimal and Robust Control: Advanced Topics with MATLAB r

reduced order models which can allow to deal with the control problem in a
simpler way. It seems clear that this operation introduces an error which can
be evaluated and taken into account in the robust controller design.
Now, let us turn to the case of the linear time-invariant systems described
by equations (1.3). In the next example, we will discuss parametric and struc-
tural uncertainty in the state matrix A and illustrate how the position of the
eigenvalues of this matrix is affected by this uncertainty.

MATLABr Exercise 1.1

Let us consider a linear time-invariant system with state matrix
 
0 1 0
A=  0 0 1  (1.7)
−1 + k1 −3 + k2 −3
It contains two parameters k1 and k2 which are uncertain. This system is therefore an
example of a system subject to parametric uncertainty. Note that for k1 = k2 = 0 the
system is asymptotically stable, being all eigenvalues of A equal and negative, namely,
λ1 = λ2 = λ3 = −1.
Now consider the case that k1 and k2 are random numbers drawn from a normal distri-
bution and calculate the eigenvalues repeating the computation for different instances
of the random process. To this aim, let us the following MATLAB commands:
for i=1:1000
k1=randn;
k2=randn;
A=[0 1 0;0 0 1;-1+k1 -3+k2 -3];
figure(1); plot(k1,k2,’k.’); hold on;
figure(2); plot(real(eig(A)),imag(eig(A)),’k.’); hold on;
end
The result is illustrated in Figure 1.3, which shows that in most of the cases the stability
is preserved, but in some cases eigenvalues with positive real part appear, yielding an
unstable system.
The exercise can be repeated considering a smaller uncertainty, for instance, using
k1=0.1*randn; and k2=0.1*randn; to obtain a system that with high probability re-
mains stable.
The effect of other distributions for k1 and k2 can also be studied. For instance, con-
sidering an uniform distribution in [−0.5, 0.5] with the commands k1=rand-0.5 and
k2=rand-0.5, we obtain the result shown in Figure 1.4. In this case, the system is
always stable.
The characteristic polynomial of A is given by p(λ) = λ3 + 3λ2 + (3 − k2 )λ + 1 − k1 as
one can also verify using symbolic calculus in MATLAB:
>> syms k1 k2 real positive
>> A=[0 1 0;0 0 1;-1+k1 -3+k2 -3];
>> p=charpoly(A)
Applying the Routh criterion for stability, one derives that the following conditions
need to be satisfied: k1 < 1, k2 < 3 and 8 − 3k2 + k1 > 0. This explains why, when
k1 and k2 are drawn from an uniform distribution in [−0.5, 0.5], one always obtains a
stable polynomial.
Let us now discuss an example of unstructured uncertainty, where the uncertainty is
no more located on some specific parameters of the state matrix A. In particular, let
us consider the following matrix:
 
0 1 0
A=  0 0 1  + ∆A (1.8)
−1 −3 −3
Modelling of Uncertain Systems and the Robust Control Problem 7

(a) (b)

FIGURE 1.3
Stability of linear systems in presence of parametric uncertainty: (a) distribu-
tion of the parameters k1 and k2 ; (b) eigenvalues of the state matrix A (1.7)
when k1 and k2 are drawn from a normal distribution.

where ∆A is the uncertainty. Note that, in absence of perturbation, i.e., ∆A = 0, the

same state matrix previously considered is recovered.
Suppose now that the coefficients of ∆A assume random values drawn from a normal
distribution with zero mean and unitary variance. The following MATLAB command
can be used to numerically study the problem:
A=[0 1 0; 0 0 1;-1 -3 -3]
for i=1:1000
DeltaA=1*randn(3);
A=A+DeltaA;
plot(real(eig(A)),imag(eig(A)),’.k’)
hold on
end
Figure 1.5 shows an example of the location of the eigenvalues of A.

For linear time-invariant systems described by equations (1.3), under

opportune hypotheses, a controller can be designed using a control law which
operates on state variables (the linear state regulator u = −kx). Remem-
ber, should not all the state variables be accessible, they would have to be
reconstructed through an asymptotic state observer. This common technique
usually includes the design of a linear regulator and an asymptotic observer
which together constitute the compensator.
When designing a linear state
 regulator, the system must be completely
controllable (i.e., matrix Mc = B AB A2 B . . . An−1 B has maximum
rank), whereas designing an asymptotic observer needs the system to be com-
T
pletely observable (i.e., matrix Mo = C CA CA2 . . . CAn−1

has
maximum rank).
Furthermore, note that, since the observer is a dynamical system (linear
and time-invariant) of order n, then the compensator obtained is a dynam-
ical system of order n. High-order systems require the design of high-order
8 Optimal and Robust Control: Advanced Topics with MATLAB r

(a) (b)

FIGURE 1.4
Stability of linear systems in presence of parametric uncertainty: (a) distribu-
tion of the parameters k1 and k2 ; (b) eigenvalues of the state matrix A (1.7)
when k1 and k2 are drawn from an uniform distribution in [−0.5, 0.5].

compensators. So, now, the problem of formulating a lower-order model also

includes the design of a compensator. To do this, there are two different tech-
niques. The first is to design a compensator and then build the lower-order
model. The second is to design the compensator directly on the lower-order
model (e.g., considering P0 (s) and neglecting ∆P (s)).
It is important to remark that the uncertainties regard the process P and
not the controller C. This is so, as it is supposed that the controller is the re-
sult of a design, so any uncertainties about its parameters are minimal. More
recently, it has been discovered that compensators designed with the theory of
robust control are fragile, that is, that compared to the compensator param-
eters, robustness is poor, so even minimal uncertainties about compensator
coefficients can lead to system de-stabilization. At the heart of the compen-
sator issue is that when it is designed to be robust against the uncertainties
in the model of process P , it remains fragile with respect to the uncertainties
in the model of controller C. Fragility means that a small perturbation in the
controller parameters can even destabilize the closed-loop system.
This chapter has briefly described the main robust control issues to be dealt
with in subsequent chapters. Before that, certain subjects will be briefly pre-
sented which are the pillars of systems theory (e.g., stability). This approach
was decided on for two reasons: first, is the importance of these subjects to
robust control, and, second, is that undergraduate courses are organized on
two levels. The course which this book is part of is a Master course for stu-
dents who have likely dealt with automation previously. So, this makes a brief
discourse on systems theory necessary.
Modelling of Uncertain Systems and the Robust Control Problem 9

FIGURE 1.5
Stability of linear systems in presence of unstructured uncertainty as in equa-
tion (1.8): eigenvalues of the state matrix A.

1.2 The Essential Chronology of Major Findings in Ro-

bust Control
Parameterization of stabilizing controllers (Youla, 1976).
Poor robustness of LQG controllers (Doyle, 1978).
Formulating H∞ problems for SISO systems (Zames, 1981).
Balanced realizations (Moore, 1981).
Definition of µ (Doyle, 1982).
Definition of multivariable stability margins (Safonov, 1982).
H∞ synthesis algorithms for systems in state-space form: high order control
(Doyle, 1984).
First book on robust control (Francis, 1987).
H∞ synthesis algorithms for systems in state-space form: low order control
(Doyle, Glover, Khargonekar, Francis, 1988).
2
Fundamentals of Stability

CONTENTS
2.1 Lyapunov Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2 Positive Definite Matrices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.3 Lyapunov Theory for Linear Time-Invariant Systems . . . . . . . . . . . 16
2.4 Lyapunov Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.5 Stability with Uncertainty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.6 Further Results on the Lyapunov Theory . . . . . . . . . . . . . . . . . . . . . . . 26
2.6.1 Hystorical Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.6.2 Lyapunov Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.7 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

This chapter deals with the problem of stability, the main requirement of a
control system. In going over some basics in system analysis, the emphasis
is on some advanced mathematical tools (e.g., Lyapunov equations), which
will be useful below. Particular attention is given to Lyapunov theory for the
case of linear time-invariant systems. The Lyapunov linear matrix equation
is introduced in detail and the concept of positive definite matrix is also dis-
cussed. Some criteria to verify this property are reported and the solution
of Lyapunov equations via vectorization is presented with numerical exam-
ples. The Kharitonov criterion to test the stability of uncertain system is also
introduced. At the end of the chapter, several worked examples are included.

2.1 Lyapunov Criteria

The equilibrium points of an autonomous dynamical system ẋ = f (x) with
 T  T
x = x1 x2 . . . xn and f (x) = f1 (x) f2 (x) . . . fn (x) can
be calculated by ẋ = 0 (i.e., solving the systems of equations f (x) = 0). Gener-
ally, the equations are nonlinear and may have one or more solutions, whereas
for autonomous linear systems, the equilibrium points can be calculated by
Ax = 0, which has only one solution (x = 0) if det(A) 6= 0.

DOI: 10.1201/9781003196921-2 11
12 Optimal and Robust Control: Advanced Topics with MATLAB r

So, nonlinear systems may have more than one equilibrium point. In
addition, each of these equilibrium points has its own stability characteristics.
For example, think of a pendulum. It has two equilibrium points as shown in
Figure 2.1. Only the second of the two equilibrium points is stable. If the pen-
dulum were to start from a position near equilibrium point (a), it would not
return to its equilibrium position, differently from what happens at point (b).
For this reason, in nonlinear systems, stability is a property of equilibrium
points, and not of the whole system. The stability of an equilibrium point
can be studied through the criteria introduced by the Russian mathematician
Aleksandr Mikhailovich Lyapunov (1857–1918).

FIGURE 2.1
Equilibrium points of a pendulum.

The first of Lyapunov’s criteria defines stability of an equilibrium point.

It says that if a system starts from initial conditions very close to a stable
equilibrium point, the state evolution is confined to the neighborhood of that
equilibrium point. If the initial conditions x(0) = x0 are close to equilibrium,
the response obtained by x0 , i.e., the perturbed response, is sufficiently close to
equilibrium point x̄ (the nominal response, i.e., the motion obtained starting
exactly from x̄, is in fact constant over time and equal to the equilibrium
point). To measure the distance between perturbed and nominal motion in
finite space, any vector norm can be used. The criterion is formally expressed
by:
Definition 1 (Lyapunov I criterion) The equilibrium point x̄ of the dy-
namical system ẋ = f (x) (f (x̄) = 0) is defined as stable when:
∀ε > 0, ∃δ > 0 such that ∀x0 with k x0 − x̄ k< δ one has
k x(t) − x̄ k< ε for ∀t > 0
Stability is said to be asymptotic, if this additional condition applies:

lim x(t) = x̄ (2.1)

t→+∞
In this case, the perturbed motion not only stays in the neighborhood of
the equilibrium point, but asymptotically tends to it.
Fundamentals of Stability 13

Definition 2 (Asymptotically stable equilibrium point) The equilibrium

point x̄ of the dynamical system ẋ = f (x) (f (x̄) = 0) is defined as asymptot-
ically stable if it is stable and furthermore condition (2.1) holds.

Before discussing Lyapunov’s second criterion, some preliminary notions

need to be discussed.
Note first that, without loss of generality, it is possible to consider x̄ = 0,
since it is always possible to shift the generic equilibrium point x̄ back to 0
by variable translation.
Now, the concept of positive definite functions needs to be introduced. Let
us consider a neighborhood Ω ⊆ Rn of point x̄ = 0.

Definition 3 (Positive definite function) Function V (x) : Rn → R is

positive definite in x̄ = 0 if the following hold:

1. V (0) = 0
2. V (x) > 0 ∀x ∈ Ω, x 6= 0

A function is positive semi-definite if V (0) = 0 and if V (x) ≥ 0 ∀x ∈ Ω,

x 6= 0. A function is negative definite if V (0) = 0 and if V (x) < 0 ∀x ∈ Ω,
x 6= 0. Finally, a function is negative semi-definite if V (0) = 0 and if V (x) ≤ 0
∀x ∈ Ω, x 6= 0.

Theorem 1 (Lyapunov II criterion) The equilibrium x̄ = 0 of the dy-

namical system ẋ = f (x) is stable if there is a scalar function of the state, the
so-called Lyapunov function, which has the following properties:

1. V (x) is positive definite for x̄ = 0

dV (x)
2. V̇ (x) = dt is negative semi-definite for x̄ = 0
The stability is asymptotic if V̇ (x) is negative definite.

From a physical point of view, the Lyapunov function represents system

energy. If the energy grows over time, it means that the state variables grow
(thus diverge). Conversely, if the energy tends to zero, the system is dissipating
energy.
Note that many Lyapunov functions satisfy the first condition of Lyapunov
II criterion; the difficulty lies in verifying the second condition.

2.2 Positive Definite Matrices

For linear systems, a Lyapunov function allowing to assess system stability is
easily found. In fact, for nonlinear systems, there is no general procedure for
Discovering Diverse Content Through
Random Scribd Documents
 178.—Recom. in favours of Sir William Dick. Ib.
179.—Ref. Master James Row. Ib.
180.—Ref. to the Commission for publike affaires, concerning the
Scots in Poleland, &c. Ib.
181.—Ref. for planting the Kirk and Colledge of Aberdene to the
Commission for publike affaires. Ib.
182.—Ref. concerning Masters William Douglas, John Logie,
George Hanna, Richard Maitland, and Coline Mackenzie. Ib.
183.—Ref. and Commission concerning the tryall of Masters Murdo
Mackenzie, John Duncane, and William Cowper. Ib.
184.—Recom. Master William Chalmber to the Committee of
Estates. Ib.
185.—Ref. James Grahame of Claypots to the Presbyterie of
Dundie. Ib.
186.—Recom. of some persons for charitie. Ib.
187.—Ref. for planting the Kirk of Bruntiland to the Commission for
publike affaires. Ib.
188.—Ref. concerning the planting of the Kirk of Prestoun to the
Commission for publike affaires. Ib.
189.—Ref. for planting the vaking Kirk in Glasgow to the
Commission for publike affaires. Ib.
190.—Ref. the Lord Ray his Son and some of his friends, to the
Commission for publike affaires. Ib.
191.—Ref. Master Gilbert Gordon, to the Commission for publike
affaires. Ib.
192.—Recom. for Master George Hannayes wife and children. Ib.
193.—Act for presenting the Confession of Faith to the Parliament.
Ib.
194.—Act concerning the Translaters of the Dutch Notes. Ib.
 195.—Recom. and Ref. concerning the collecting of the Passages
and Occurrances of these late times, to the Commission for publike
affaires. Ib.
196.—Ref. for planting the Kirk of Ancrum, to the Commission for
publike affaires. Ib.
197.—Renovation of the Commission for visitation of the University
of Glasgow. Ib.
198.—Ref. Master John Rosse at Birse, to the Synode of Aberdene.
Ib.
199.—Ref. of Master Thomas Ramsay younger, his petition to the
Commission for publike affaires. Ib.
200.—Indiction of a Thanksgiving and of a Fast. Ib.
FINIS.
 Miscellaneous Historical Documents,

RELATIVE TO THE ECCLESIASTICAL AND POLITICAL

EVENTS IN SCOTLAND—1647.
 1. Excerpts from Principal Baillie’s Account of the
Westminster Assembly, continued from p. 460.
To Mr William Spang. Edinburgh, Jan. 1647.
Dear Cousin,—I wrote to you at length before I came from London.
I have had a long and tedious, but, thanks to God, prosperous
journey. I am now here well. I have made my report in the
commission of the church to all their contentment; our errand in
England being brought near a happy period, so far as concerned us
the commissioners of the church; for, by God’s blessing, the four
points of uniformity, which was all our church gave us in commission
to agent in the assembly at Westminster, were as good as obtained.
The Directory I brought down before. The model of government we
have gotten it through the assembly according to our mind: it yet
sticks in the hands of the Houses. They have passed four ordinances
at least about it, all pretty right, so far as concerns the constitution
and erection of general assemblies, provincial synods, presbyteries,
and sessions, and the power of ordination. In the province of
London and Lancashire the bodies are set up. That the like diligence
is not used long ago in all other places, it is the sottish negligence of
the ministers and gentry in the shires more than the parliament’s.
That the power of jurisdiction in all things we require, excepting
appeals from the general assembly to the parliament, is not put in
ordinances long ago, it is by the coming of the Independents and
Erastians in the House of Commons; which obstacle we trust will
now be removed by the zeal of the city of London; so much the
more, as our nation are taken away sooner and more easily than any
did expect. All grounds of jealousy of our joining with the King, the
greatest prop of the sectaries power in the House. However, in the
jus divinum of Presbytery, printed by the ministry of London, you
may see that burden taken off our shoulders; the body of the
ministry of England, not the assembly and Londoners only, being
fully leavened with our sense in all the point of government, and
become willing, and able abundantly, to manage that cause, without
us, against all opposites. The third point, the Confession of Faith, I
brought it with me, now in print, as it was offered to the Houses by
the assembly, without considerable dissent of any. It is much cried
up by all, even many of our greatest opposites, as the best
Confession yet extant. It is expected the Houses shall pass it, as
they did the Directory, without much debate. Howbeit the retarding
party has put the assembly to add scriptures to it, which they
omitted only to eschew the offence of the House, whose practice
hitherto has been, to enact nothing of religion on divine right or
scriptural grounds, but upon their own authority alone. This
innovation of our opposites may well cost the assembly some time,
who cannot do the most easy things with any expedition; but it will
be for the advantage and strength of the work. The fourth part of
our desired and covenanted uniformity is the Catechism. A
committee has drawn and reported the whole.
The assembly ere I came away had voted more than the half. A
short time will end the rest; for they study brevity, and have voted to
have no other head of divinity into it than is set down in the
Confession. This ended, we have no more ado in the assembly,
neither know we any more work the assembly has in hand, but an
answer to the nine queries of the House of Commons about the jus
divinum of divers parts of the government. The ministers of London’s
late jus divinum of Presbytery does this abundantly. Also a
committee of the assembly has a full answer to all these queries
ready. The authors repent much of that motion. Their aim was, to
have confounded and divided the assembly by their insnaring
questions; but finding the assembly’s unanimity in them, the
Independents principles forcing them to join with the rest, in
asserting the divine right of these points of government, whereupon
the parliament does most stick, the movers of these questions
wishes they had been silent. There is no more work before the
assembly. The translation of the psalms is passed long ago in the
assembly; yet it sticks in the Houses. The Commons passed their
order long ago; but the Lords joined not, being solicited by divers of
the assembly, and of the ministers of London, who love better the
more poetical paraphrase of their colleague Mr Burton. The too great
accuracy of some in the assembly, sticking too hard to the original
text, made the last edition more concise and obscure than the
former. With this the commission of our church was not so well
pleased; but we have got all those obscurities helped; so I think it
shall pass. Our good friend Mr Zachary Boyd has put himself to a
great deal of pains and charges to make a psalter, but I ever warned
him his hopes were groundless to get it received in our churches;
yet the flatteries of his unadvised neighbours makes him insist in his
fruitless design.
When I took my leave of the assembly, I spoke a little to them.
The prolocutor, in name of the assembly, gave me an honourable
testimony, and many thanks for my labours. I had been ever silent in
all their debates; and however this silence sometimes weighted my
mind, yet I found it the best and wisest course. No man there is
desired to speak. Four parts of five do not speak at all; and among
these are many most able men, and known by their writs and
sermons to be much abler than sundry of the speakers; and of these
few that use to speak, sundry are so tedious, and thrusts themselves
in with such misregard of others, that it were better for them to be
silent. Also there are some eight or nine so able, and ready at all
times, that hardly a man can say any thing, but what others, without
his labour, are sure to say as well or better. Finding, therefore, that
silence was a matter of no reproach, and of great ease, and brought
no hurt to the work, I was content to use it, as Mr Henderson also
did.
* * * *
It is very like, if he had done any duty, though he had never taken
the covenant, but permitted it to have been put in an act of
parliament in both kingdoms, and given so satisfactory an answer to
the rest of the propositions, as easily he might, and sometimes I
know he was willing, certainly Scotland had been for him as one
man; and the body of England, upon many grounds, was upon a
disposition to have so cordially embraced him, that no man, for his
life, durst have muttered against his present restitution. But
remaining what he was in all his maxims, a full Canterburian, both in
matters of religion and state, he still inclined to a new war; and for
that end resolved to go to Scotland. Some great men there pressed
the equity of Scotland’s protecting of him on any terms. This
untimeous excess of friendship has ruined that unhappy prince; for
the better party, finding the conclusion of the King’s coming to
Scotland, and thereby their own present ruin, and ruin of the whole
cause, the making the malignants masters of church and state, the
drawing the whole force of England upon Scotland for their
perjurious violation of their covenant, they resolved by all means to
cross that design.
So when others proposed to the parliament the assistance of the
King to recover his government in England, notwithstanding any
answer he might give to the propositions, the better sort, before
they should give answer to so high a question, desired a publick fast
in the parliament, and the advice also of the commission of the
church. Both with some difficulty were obtained. But after that fast,
and the distinct answer of the church, that it was unlawful for
Scotland to assist the King for his recovery of the government in
England, if he approved not the covenant, the parliament was
peremptor to refuse the King free access to Scotland, unless he
satisfied the propositions. This much they signified to him by their
commissioners, which we met at Newcastle. It was easy to be
grieved, and to find what to reprehend in this resolution; for indeed
it was clothed with many dangers and grievances; but to fall at that
nick of time on any conclusion, free of more dangers and grievances,
seemed impossible.
* * * *
July 13, 1647.—These matters of England are so extremely
desperate, that now twice they have made me sick. Except God
arise, all is gone there. The imprudence and cowardice of the better
part of the city and parliament, which was triple or sextuple the
greater, has permitted a company of silly rascals, who call
themselves yet no more than 14,000, horse and foot, to make
themselves masters of the King, parliament, and city, and by them of
all England: so that now that disgraced parliament is but a
committee to act all at their pleasure, and the city is ready to fright
the parliament at every first or second boast from the army. No
human hope remains but in the King’s unparallelled wilfulness, and
the army’s unmeasurable pride. As yet they are not agreed, and
some write they are not like to agree: for in our particular I expect
certainly they will agree well enough, at what distance soever their
affections and principles stand. Always if the finger of God in their
spirits should so far dement them as to disagree, I would think there
were yet some life in the play; for I know the body of England are
overweary long ago of the parliament, and ever hated the sectaries,
but much more now for this their unexpected treachery and
oppression. On the other part, the King is much pitied and desired;
so if they give him not contentment, he will overthrow them. If he
and they agree, our hands are bound: we will be able, in our present
posture, and humour of our highly distracted people, to do nothing.
And whom shall we go to help, when none calls? but the King,
parliament, and city, as their masters command, are ready to declare
against us if we should offer to arm. But if the King would call, I
doubt not of rising of the best army ever we had, for the crushing of
these serpents, enemies to God and man.
* * * *
To a friend in Kilwinning. Edinburgh, August 20, Friday night.
The city’s declaration and diurnal declare in what a brave posture
both the city and parliament once were in. The other papers shew
how soon all was overturned. The army marched through the whole
city by way of triumph; but staid not in it, did no violence to any;
only three or four regiments keep the forts about Westminster, and
guard the parliament still. For all that, the House of Commons vote
sundry things contrary to the mind of the army. How long that
courage will remain, I cannot say. It is thought that people, when
they have felt a little the burden of the army, will break that yoke by
one mean or other. The army’s mind, much of it, may be seen in
their propositions, a paper which I purposed to send, but now it is
fallen by. By it they are clear enough for a full liberty of conscience,
a destroying of our covenant, a setting up of bishops, of inthralling
the King, so far, as in my judgement, he and they will not agree,
albeit many think they are agreed already. If this were, our case
were very hard. Never more appearance of a great discord, both in
our church and state, some few days ago; but, blessed be God, the
appearances are now much changed. Never assembly more
harmonious than this yet has been. Our declaration to England, a
very good piece, is passed without a contrary voice. An act against
vaigers [strollers] from their own ministers, and a large direction for
private worship, drawn by Mr Robert Blair for the correcting all the
faults in worship, which offended many here, is past the committee,
without a contrary voice; and, I think, shall pass the assembly also,
no less unanimously; which demonstrates the truth of what I said in
my assembly-speech, “That for all the noise some made, yet truly
there was no division as yet in our church.”
Yesterday, and this night, our state, after much irreconcileable
difference, as appeared, are at last unanimously agreed to send the
Chancellor and Lanerk to the King and parliament of England, to
comfort and encourage both to keep our covenant, and not to agree
to the propositions of the army. No appearance, as yet, of any
stirring in haste in this kingdom.
To Mr Spang. Edinburgh, September 1, 1647.
—— London has lien like a millstone on my breast now of a long
time. The first week we came to this town, my heart was a little
relieved. I thought God had answered our prayers much sooner than
I expected, and had put London in so good a posture for averting all
our fears as I could have wished; but that joy lasted not full eight
days. Stapleton and Hollis, and some others of the eleven members,
had been the main persuaders of us to remove out of England, and
leave the King to them, upon assurance, which was most likely, that
this was the only means to get that evil army disbanded, the King
and peace settled according to our minds: but their bent execution
of this real intention has undone them, and all, till God provide a
remedy. We were glad when Leslie was recalled from his
Lieutenancy of Ireland, a creature of Cromwell’s, who got that great
trust for no virtue at all but his serviceableness to that faction. This
was the first sensible grievance to that army. The second was the
employing of Skippon and Massie, in the Irish command, and giving
to Fairfax such a command in England as made him not very
formidable. But when the third stroke came, of disbanding the most
of the sectaries, and cashiering of their officers, this put them on
that high and bold design, which as yet they follow, as, I think, not
so much on great preconception, as drawn on by the course of
affairs, and light heads of their leaders. Vane and Cromwell, as I
take it, are of nimble hot fancies for to put all in confusion, but not
of any deep reach. St John and Pierpont are more stayed, but not
great heads. Sey and his son, not _____, albeit wiser, yet of so dull,
sour, and fearful a temperament, that no great atchievement, in
reason, could be expected from them. The rest, either in the army or
parliament, of their party, are not on their mysteries, and of no great
parts either for counsel or action, so far as I could ever observe. The
folly of our friends was apparent, when at the army’s first back-
march, and refusal to disband, they recalled their declaration against
their mutinous petitions. Easily might all their designs have been
crushed at that nick of time, with one stout look more; but it was a
dementation to sit still amazed at the taking of the King, the
accusation of the eleven members, the army’s approaching to the
city. Here, had the city agreed, and our friends in parliament shewed
any resolution, their opposites counsel might even then have been
easily overturned; for all this while, the army was not much above
10,000 ill-armed soldiers. But the irrecoverable loss of all, was the ill-
managing of the city’s brave engagement. Had they then made fast
the chief of the sectarian party in both Houses, and stopt their flight
to the army; had Massey and Waller, with any kind of masculine
activity, made use of that new trust committed to them; Mr Marshal,
and his seventeen servants of the synod, for all Foulks and Gib’s
subornation, should never have been bold to offer that destructive
petition to the Houses and common council, which, without any
capitulation, put presently in the army’s power, the parliament, city,
and all England, without the least contradiction. An example rarely
parallelled, if not of treachery, yet at least of childish improvidence
and base cowardice. Since that time they have been absolute
masters of all. Which way they will use this unexpected sovereignty,
it will quickly appear. As yet they are settling themselves in their new
saddle. Before they got up, they gave the King and his party fair
words; but now, when all is their own, they may put him in a harder
condition than yet he has tasted of. Their proposals, a part of their
mind, give to the King much of his desire in bringing back bishops
and books, in putting down our covenant and presbytery, in giving
ease to malignants and Papists; but spoil him of his temporal power
so much, as many think, he will never acquiesce to; albeit it is
spoken loud, that he and they are fully agreed.
Our state here, after long expectation to have heard something of
the King’s own mind and desires, as yet have heard nothing from
him to account of. Although he should employ their help against his
oppressors, yet he being still altogether unwilling to give us any
satisfaction in the matter of our covenant, we are uncertain what
course to take; only we do resent to our commissioners to oppose
the proposals, and to require a safe-conduct to the Chancellor and
Lanerk to come up to the King and parliament. It cost many debates
before it came to this conclusion. Our great men are not like to pack
up their differences. Duke Hamilton and his friends would have been
thought men composed of peace on any terms, and to have cast on
other designs of embroiling Scotland in a new war. But when all
were weary of jangling debates, the conclusion whereto the
committee was brought, was so far to espouse the King’s quarrel on
any terms, that Argyle and Wariston behoved to protest against our
engagement on any such terms. To avoid invidious protestations,
both parties agreed to pass an act of not engagement. The
proceedings of some are not only double and triple, but so manifold,
that as no other, so, in my mind, themselves know not what they
finally intend. They who made themselves gracious and strong, by
making the world believe that it was their opposites who had
brought the country in all the former trouble, and would yet again
bring it into a new dangerous war, when it came to the point, were
found to precipitate us into dangers, and that in such terms as few
with comfort could have undertaken. We have it from divers good
hands at London, that some here kept correspondence with Sir
Thomas Fairfax, which to me is an intolerable abomination. The
present sense of many is this: if the King and the army agree, we
must be quiet and look to God: if they agree not, and the King be
willing to ratify our covenant, we are all as one man to restore him
to all his rights, or die by the way: if he continue resolute to reject
our covenant, and only to give us some parts of the matter of it,
many here will be for him, even in these terms; but divers of the
best and wisest are irresolute, and wait till God give more light.
David Leslie, with a great deal of fidelity, activity, and success, has
quieted all our highlands and isles, and brought back our little army;
which, we think, shall be quartered here and there, without
disbanding, till we see more of the English affairs. The pestilence,
for the time, vexes us. In great mercy Edinburgh and Leith, and all
about, which lately were afflicted with more of this evil than ever
was heard of in Scotland, are free. Some few infections now and
then, but they spread not. Aberdeen, Brechin, and other parts of the
north, are miserably wasted. St Andrew’s and Glasgow, without
great mortality, are so threatened, that the schools and colleges now
in all Scotland, except Edinburgh, are scattered.
While I had written thus far, by the packet this day from London I
learn, that the army daily goes higher and higher, which to me is a
hopeful presage of their quicker ruin. The chief six of the eleven
members were coming to you, Stapleton, Esler, Hollis; the second
gentleman, for all gallantry in all England, died at Calais. I think it
will be hard to the parliament and city to bear these men long; and I
hope, if all men were dead, God will arise against them. Munster is
not like to be a school to them long. Cromwell and Vane are like to
run on to the end of Becold and Knipperdolling’s race.
Northumberland has feasted the King at Swahouse; hence he went
to Hampton-court. They speak of his coming to Whitehall. If he
agree no better with the sectaries than yet he does, that journey
may prove fatal. He is not likely to come out of London willingly; and
if the army should draw him, that violence may waken sleeping
hounds. If they let him come to London, without assurance of his
accord with them, they are more bold and venturous than wise; and
if the King agree to their state-designs, I think he is not so
consonant to all his former principles and practice as I took him.
I know you expect some account of our assembly. Take it, if you
have patience to read what I have scribbled in haste, on a very ill
sheet of paper. I have no leisure to double; for our commissioners
enter every day at seven, and we are about publick business daily till
late at night. At our first meeting, there was clear appearance of
formed parties for division; but God has turned it so about, that
never assembly was more harmonious and peaceable to the very
end. The last year, a minister in the Merse, one Mr James Simson,
whose grandsire was, as I take it, an uncle or brother to famous Mr
Patrick of Stirling, a forward, pious, young man, being in suit of a
religious damsel, sister to Mr James Guthrie’s wife, had kept with Mr
James Guthrie, and others, some private meetings and exercises,
which gave great offence to many. When they came before the last
general assembly and commission of the kirk, Mr David Calderwood
and sundry other very honest men, opposite to malignants, were
much grieved, and by that grief moved to join with Mr William Colvil,
Mr Andrew Fairfoul, and such whom some took to be more
favourable to malignants than need were. These two joined together,
made a great party, especially when our statesmen made use of
them to bear down those who had swayed our former assemblies.
The contest was at the chusing of the moderator. The forementioned
party were earnest for Mr William Colvil. Many were for me; but I
was utterly unwilling for any such unfit charge, and resolved to
absent myself from the first meeting, if by no other means I could
be shifted the leet. At last, with very much ado, I got myself off, and
Mr Robert Douglas on the leets; who carried it from Mr William Colvil
only by four voices. God’s blessing on this man’s great wisdom and
moderation has carried all our affairs right to the end; but Mr David
Calderwood having missed his purpose, has pressed so a new way of
leeting the moderator for time to come, that puts in the hand of
base men to get one whom they please, to our great danger. We
spent a number of days on silly particulars. Mr Gillespie came home
at our first downsitting. He and I made our report to the great
satisfaction of all. You have here what I spoke. Mr Calderwood was
much offended with what I had spoken in the end; but my apology
in private satisfied him. He, and others of his acquaintance, came
with resolution to make great din about privy meetings and
novations, being persuaded, and willing to persuade others, that our
church was already much pestered with schism. My mind was clean
contrary: and now, when we have tried all to the bottom, they are
found to be much more mistaken than I; for they have obtained,
with the hearty consent of these men whom they counted greatest
patrons of schism, all the acts they pleased against that evil,
wherein the wisdom and authority of Mr Blair has been exceeding
serviceable. This yielding on our side, to their desires, drew from
them a quiet consent to these things we intended, from which at
first they seemed much averse. We agreed, nemine contradicente, to
that declaration, which was committed to Mr Gillespie and me, but
was drawn by him alone; also, after much debate in the committee,
to the Confession of Faith; and to the printing of the Directory for
government, for the examination of the next general assembly; of
the Catechism also, when the little that remains shall come down;
likewise for printing to that same end two or three sheets of Thesis
against Erastianism, committed to Mr Gillespie and me, but done by
him at London at Voetius’s motion; which we mind, when approven
here, to send to him; who is hopeful to get the consent of your
universities and of the general assembly of France to them, which
may serve for good purpose. We have put the new Psalter also in a
good way.—— We have this day very happily ended our assembly
with good concord; albeit Mr David Calderwood, serving his own
very unruly humour, did oft very much provoke. He has been so
intolerable through our forbearance, that it is like he shall never
have so much respect among us. His importunity forced us, not only
to a new ridiculous way of chusing the moderator, but on a conceit
he has, that a minister deposed should not again be reposed almost
in no case, he has troubled us exceedingly about the power of the
commission of the kirk to depose a minister in any case; yet we
carried it over him. We have obtained leave to print all our English
papers, Catechism, Confession, Propositions, and Directory for
government and ordination, our debates for accommodation against
toleration, our papers to the grand committee, the propositions for
government, albeit passed both in our assembly and parliament
1643. Mr David opposed vehemently the printing, and his grand
followers, Mr John Smith and Mr William Colvil with him, because
they held forth a session of a particular congregation to have a
ground in scripture, which he, contrary to his Altar of Damascus,
believes to have no divine right, but to be only a commission, with a
delegated power from the presbytery, tolerated in our church for a
time. With great difficulty could we get the printing of that paper
passed for his importunity; but at last we got all.
An express from London this day tells us, that the army’s
parliament press the concurrence of our commissioners to send to
Hampton-court the propositions to the King. This seems to import
the King’s refusal of the proposals, and disagreeing yet with the
army. And what they will do with the King, if he refuse the
propositions also, we know not; only their last remonstrance shews
their resolution to cast out of the parliament many more members,
and to take the lives of some for example. The spirit that leads
them, and the mercy of God to that oppressed people, will not
permit these tyrannous hypocrites to rest, till, by their own hands,
they have pulled down their Babel.
* * * *
October 13, 1647.—— We gave in this day to the states a
remonstrance of the hazard of religion and covenant, if our army
should disband. We hope that plot, long hatched, and with too great
eagerness driven on, shall this day or to-morrow be broken. Our
dangers of farther confusion are great, if God be not merciful. The
persecution at London is very intolerable. I am very confident that
party, so much opposite to God and man, cannot long stand. Ere
long, at my leisure, I may give you a particular account of all our
affairs.
 1646.—August 1.

2. His Majesty’s Answer to the Propositions.359

Charles Rex,
The Propositions tender’d to his Majesty by the Commissioners
from the Lords and Commons assembled in the Parliament of
England at Westminster, and the Commissioners of the Parliament of
Scotland, (to which the Houses of Parliament have taken twice so
many Months for deliberation, as they have assigned Days for his
Majesty’s Answer) do import so great Alterations in Government both
in the Church and Kingdom, as it is very difficult to return a
particular and positive Answer, before a full debate, wherein these
Propositions, and the necessary Explanation, true Sense and
Reasons thereof, be rightly weighed and understood; and that his
Majesty upon a full view of the whole Propositions, may know what
is best, as well as what is taken away and changed. In all which he
finds (upon discourse with the said Commissioners) that they are so
bound up from any capacity either to give Reasons for the Demands
they bring, or to give ear to such Desires as his Majesty is to
propound, as it is impossible for him to give such a present
Judgment of, and Answer to these Propositions, whereby he can
answer to God that a safe and well-grounded Peace will ensue
(which is evident to all the World can never be, unless the just
Power of the Crown, as well as the Freedom and Propriety of the
Subject, with the just Liberty and Privileges of the Parliament, be
likewise setled:) To which end his Majesty desires and proposeth to
come to London, or any of his Houses thereabouts, upon the Publick
Faith and Security of the Two Houses of Parliament, and the Scotch
Commissioners, That he shall be there with Freedom, Honour, and
Safety; where by his Personal Presence he may not only raise a
mutual Confidence betwixt him and his People, but also have these
Doubts cleared, and these Difficulties explained unto him, which he
now conceives to be destructive to his just Regal Power, if he shall
give a full Consent to these Propositions, as they now stand.
As likewise, that he may make known to them such his reasonable
Demands, as he is most assured will be very much conducible to
that Peace which all good men desire and pray for, by the setling of
Religion, the just Privileges of Parliament, with the Freedom and
Propriety of the Subject: And his Majesty assures them, That as he
can never condescend unto what is absolutely destructive to that
just Power which by the Laws of God and the Land he is born unto;
so he will chearfully grant and give his Assent unto all such Bills, at
the desire of his Two Houses, or reasonable demands for Scotland,
which shall be really for the Good and Peace of his People, not
having regard to his own particular (much less of any body’s else) in
respect of the Happiness of these Kingdoms. Wherefore his Majesty
conjures them as Christians, as Subjects, and as Men who desire to
leave a good Name behind them, that they will so receive and make
use of this Answer, that all Issues of Blood may be stopped, and
these unhappy Distractions peaceably setled.
Newcastle, Aug. 1, 1646.
To the Speaker of the House of Peers
pro Tempore, to be communicated.
Postscript.—Upon Assurance of a happy Agreement, his Majesty
will immediately send for the Prince his Son, absolutely expecting his
perfect Obedience to return into this Kingdom.
 1646.—September.

3. His Majesty’s Answer to the Scots Commissioners

at Newcastle.360
My Lords,
I shall begin, by answering what you have now said: For I assure
you I had not thus long delay’d my Answer, but to weigh fully those
Reasons and Arguments which you have laid before me, whereby to
use the uttermost of my Endeavours to give you all possible
Satisfaction; for you having told me nothing but what I have heard
before, the Change of Answer could hardly be expected. And now I
do earnestly desire you to consider what it is that I desire, which is,
To be heard; which if a King should refuse to any of his Subjects, he
would for that be thought a Tyrant. For this, if I had but slight
Reasons, it were the less to be regarded; but they are such, upon
which such a Peace as we all desire, doth depend: For albeit it is
possible, that if I should grant all you desire, a Peace might be
slubber’d up, yet it is impossible that it should be durable, unless
there be right understanding betwixt Me and my People; which
cannot be without granting what I desire.
Yet I desire to be rightly understood; for tho’ many like to Æsop’s
Fable will call Ears Horns, yet let men say what they will, I am far
from giving you a Negative, nay, I protest against it, my only desire
being to be heard; For I am confident that upon Debate I shall so
satisfy them in some things, as likewise I believe they may satisfy
me in many things, that we shall come to a most happy Agreement.
This I believe is not much needful to satisfy your Judgments; for I
am not ignorant how really your Commissioners at London have
endeavoured a Satisfactory Answer to my Message, as likewise what
good Instructions have been sent them out of Scotland; so that the
Force of Power more than the Force of Reason, hath made you so
instant with me as you have been; with which I am so far from
finding fault, that what you have done, I take well, knowing that it
proceeds out of the abundance of your Zeal to my Service:
Therefore as you see I do not mistake you; so I am careful not to be
mistaken by you; wherefore again I desire you to take notice, that I
do not give a Denial, my desire being only to be heard; as likewise
that you will take things as they are, since neither you nor I can
have them as we would; wherefore let us make the best of every
thing, and now as you have fully performed your Duty to me, so I
cannot doubt but you will continue to press those at London to hear
Reason: And certainly you can expect little fair dealing from those
who shall reject so much Reason, and of that sort, which you have,
and I hope will offer them. Not to stay too long upon so unpleasing
a Subject, I assure you, that nothing but the Preservation of That
which is dearer to me than my Life, could have hinder’d me from
giving you full Satisfaction: For upon my word, all the Dangers and
Inconveniences which you have laid before me, do not so much
trouble me, as that I should not give full Satisfaction to the Desires
of my Native Country, especially being so earnestly press’d upon me:
And yet here again I must tell you (for in this case Repetitions are
not impertinent) that I do not give you a Denial, nay I protest
against it; and remember, it is your King that desires to be heard.
 1646.—September.

4. Another paper sent by the King to the Scots

Commissioners at Newcastle.361
My Lords,
’Tis a very great Grief to me, that what I spoke to you yesterday,
and offer’d to you in writing, concerning Religion, hath given so little
Satisfaction: Yet lest the Reasons I then told you, should not be so
fully understood, I think it necessary at this time to set them down
to you in this Paper: I then told you, that whatsoever was my
particular Opinion, I did no ways intend to persuade you to do any
thing against your Covenant; wherefore I desire you to consider,
whether it be not a great step to your Reformation (which I take to
be the chief End of your Covenant) that Presbyterial Government be
legally setled. It is true, that I desire that my own Conscience, and
those that are of the same Opinion with me, might be preserved,
which I confess doth not as yet totally take away Episcopal
Government; but then consider withal, that this will take away all the
Superstitious Sects and Heresies of the Papists and Independents; to
which ye are no less obliged by your Covenant, than the taking away
of Episcopacy: And this that I demand is most likely to be but
Temporary; for if it be so clear as you believe, that Episcopacy is
unlawful, I doubt not but God will so enlighten mine Eyes, that I
shall soon perceive it; and then I promise you to concur with you
fully in matters of Religion: But I am sure you cannot imagine that
there is any hope of converting or silencing the Independent Party,
which undoubtedly will get a Toleration in Religion from the
Parliament of England, unless you join with me, and in that way I
have set down for the Re-establishing my Crown, or at least that you
do not press me to do this (which is yet against my Conscience) until
I may do it without sinning: Which as I am confident none of you
will persuade me to do, so I hope you have so much Charity, not to
put things to such a desperate Issue, as to hazard the loss of us all,
because for the present you cannot have full Satisfaction from me in
point of Religion: Not considering, that besides the rest of the
Mischiefs which may happen, it will infallibly set up the innumerable
Sects of the Independents; nothing being more against your
Covenant, than permitting of those Schisms to increase. As for the
Message which I think fit at this time to send, I have chosen rather
to mention the Point of Religion in a general than particular way, lest
(not knowing all these Reasons which I set down to you, which are
most unfit for a Message) it may give less Satisfaction than I desire:
Nevertheless I do conjure you, by that Love and Loyalty you have
always professed unto me, That you make use of what I offered
yesterday in writing, with these Reasons which I have now set down
to you, and those further Hopes I have now given you, for the best
advantages of my Service; with this particular Explanation, That
whereas I mentioned that the Church-Government should be left to
my Conscience, and those of my Opinion, I shall be content to
restrict it to some few Diocesses, as Oxford, Winchester, Bristol, Bath
and Wells, and Exeter; leaving all the rest of England fully to the
Presbyterian Government, with the strictest Clauses you shall think
upon, against Papists and Independents.
Postscript.—I require you to give a particular and full Account
hereof to the General Assembly in Scotland, shewing them that I
shall punctually make good my last Letter to them, and that this is a
very great step to the Reformation desired, not only by the present
putting down all Sects and Independents, but likewise presently
establishing Presbyterian Government; hoping that they, as Ministers
of God’s Word, will not press upon me untimously the matter of
Church-Government and Discipline, until I may have leisure to be so
persuaded, that I may comply with what they desire, without Breach
of Conscience, which I am confident they as Churchmen cannot
press me to do.
 1646.—December 17.

5. A Solemn and Seasonable Warning, to all Estates

and Degrees of Persons throughout the Land; By
the Commissioners of the General Assembly.362
The Conscience of our duty, and of the great trust reposed in us,
suffereth us not to be silent, nor to connive at the present dangers
which may justly be apprehended and expected from the Enemies of
this Cause and Covenant; Who although they cannot in this
conjuncture of time appear in the same manner as formerly they
have done; yet having retained the same principles (while they seem
to lay aside their former practices) do in a more covert and
dangerous way still drive at their own ends; And as Sathan is neither
sleeping nor idle, though he appear not always as a roaring Lion; So
these who are inspired and acted by him, have their wheels still
moving, though sometimes they make no great noise. Wherefore
that we may truly and faithfully contribute what is incumbent to us,
for preventing or removing any occasions of new troubles or
differences between the King and his People, in both, or either of
those United Kingdoms, or between the Kingdoms themselves; And
least the Church of Christ, and the true Reformed Religion be again
tossed with another, and perhaps a greater Tempest in the depth,
after we seemed to be near the Harbour; We have found it, not only
competent to our Place and Calling, but necessary for us (according
to former laudable Presidents both old and late) To emit this new
Seasonable Warning to the People of God in this Land, and to all
Estates and Degrees of men therein; Whom we exhort, That first,
and above all things, they apply their thoughts to make peace with
God, to take notice of the remaining and renewed tokens of divine
displeasure against the Land, To tremble at the remembrance of
former, and appearances of future judgments, To lament after the
Lord, To lye low before the Throne of Grace, To cry mightily to
Heaven for dispelling that cloud of sin which separateth between our
God and us, and for turning away that cloud of wrath which hangeth
over our heads. There is cause to be humbled and to repent, as for
all our iniquities, So for the too little assistance which hath been
given to such as have born the heaviest burthen, and suffered most
in this cause; And for the too much compliance with, and indulgence
to many who have been active in the late execrable Rebellion. We
know that none can reach the perfection of their duty, neither will
the Lord reckon with his People according to his Justice, but spare
them who walk in the integrity of their Spirits, as a man spareth his
own Son, so that they may rejoyce in his mercy, notwithstanding of
their short-comings, wherein they do not allow themselves; But
wilful neglects are just grounds of a great controversie on the Lord’s
part, and of deep humiliation on ours: And we conceive that the
failings of many are such, because the word of the Lord is a burthen
unto them; And though they walk in the ways of their own heart, yet
they say they shall have Peace; We would have none that are thus
guilty to account light of it, and say, Is it not a little one? Every duty
whereto we are obliged in the Covenant, is of great consequence,
and breaches even in smaller things prove inlets unto more grievous
revoltings.
When we consider how many who were once open opposers or
secret underminers, being received to the Covenant, yet remain
disaffected to the ends of the same; We cannot but think that we
walk in the midst of snares, and that mysteries of iniquity work
amongst us, which may produce most sad and lamentable effects,
unto the prejudice of our Religion and Liberties. Therefore, Because
God hath no greater quarrel against a Nation than that of a broken
Covenant; Let all who fear an Oath, remember the vows of God
which are upon them, Watch and Pray, and take good heed that they
be not cheated nor charmed into a violation of all, or any of the
Articles of that Sacred and Solemn League and Covenant; And let
those especially be observed and avoided, who do, or shall
endeavour a division and breach between the Kingdoms, or the
making of any factions or parties contrary to the Covenant, under
pretence of preserving the King and his Authority, whilst they do not
constantly and sincerely prosecute and press our frequent desires of
his subscribing the League and Covenant, and giving satisfaction in
all things to the just desires of both Kingdoms; which underhand
dealing can prove nothing else, but an abusing of His Majesty for
mens own designs: We wish that none suffer themselves to be
deceived by any false glosses of the Covenant, under which some
may possibly urge the keeping of it, so as to draw us into a certain
breach thereof, and press the defence of the King’s Authority and of
Religion, to engage us in those ways that would tend to the ruin of
both: We are not now to press the want of full satisfaction in the
much desired work of Uniformity, as the ground of a breach between
the Nations; Though we still conceive, this Nation will never be
wanting to prosecute that work to the uttermost of their power in all
lawful ways, according to the League and Covenant.
These Kingdoms, after many fervent Supplications and faithful
endeavours of all the Lovers of Truth and Peace, have been happily
united into a League and Covenant, which to this day hath been
kept inviolably, notwithstanding of all the opposition of open
Enemies, and plotting of secret Underminers; And we are confident
that none but such as have hearts full of Atheism and Treachery, will
attempt the violation thereof, in whole, or in part; And that if any
shall do the same, they shall expose themselves to the Curse of
Almighty God, who will be avenged upon all that Swear falsly by his
Name. We know that men of perverse minds, wanting the fear of
God, and measuring all things by their own ends, may conceive of it
as alterable, or at least, that all the Clauses or Heads thereof are not
so to be stuck upon, but that some one or more may be dispensed
with upon civil advantages: But we have not so learned Christ or his
Word: Both Nations have Covenanted with God, and each of them
with another, in things most lawful and necessary for the
preservation and good of both, without any limitation of time: And
therefore we and our Posterity are obliged before God unto the
Observation thereof, as long as the Sun and Moon shall endure. The
Sense of these things ought to be so deeply engraven upon the
hearts of all that are in trust, That as they should from their Souls
abhor every thought of a breach with England; So should they
carefully and wisely study to avoid everything that may prove a
snare and tentation unto the same. Amongst other things, if his
Majesty shall have thoughts of coming to this Kingdom at this time,
he not having as yet subscribed the League and Covenant, nor
satisfied the lawful desires of his Loyal Subjects in both Nations, We
have just cause to fear that the consequences of it may be very
dangerous, both to his Majesty and these Kingdoms; Which
therefore we desire may be timely prevented.
For so long as his Majesty doth not approve in his heart, and seal
with his hand the League and Covenant, we cannot but apprehend,
that according to his former Principles, he will walk in opposition to
the same, and study to draw us unto the violation thereof, and the
dissolution of the Union so happily begun between Us and our
Brethren, To weaken the Confidence and Trust, and to entertain
jealousies, and make divisions amongst our selves; Neither is it
possible, but that our receiving him in this present posture of Affairs,
will confirm the suspicions of the English Nation, of our underhand
dealing with him before his coming to our Army; And make them,
not without cause, to think that we purpose to dispose of him
without their consent, and to their prejudice; which is contrary to the
Profession of those that were in trust at his Majesty’s first coming to
the Scots Quarters, and overthroweth all the Arguments that have
been used by the Commissioners of our Parliament in their Papers
concerning The disposing of his Majesty’s Person by the joynt advice
and common consent of both Kingdoms given in to both Houses of
Parliament in England; Nor do we see how we can vindicate such a
practice from a direct breach of our engagements to them by
Covenant and Treaty; which were not only to expose us into the
hazard of a Bloody War, but to involve us in the guilt of Perjury. And
what greater disservice could be done to his Majesty and his
Posterity, than to give way to a course that might prove prejudicial
to their interest in the Crown and Kingdom of England.
 Our carriage now for many years past, in the midst of many
tentations, hath put us beyond all suspicion in the point of our
Loyalty; nor have we the least thoughts of deserting the King’s
Majesty in a just and good cause, being bound by our Covenant in
our several Vocations to endeavour with our Estates and Lives, to
preserve and defend his Person and Authority, in the defence and
preservation of the true Religion and Liberties of the Kingdoms: And
so far as his Majesty shall be for these, we really are, and we trust
the rest of his Kingdoms will be for him: Yet we cannot deny, but
openly avouch it, That if his Majesty (which the Lord forbid) shall not
satisfie the just desires of his People; Both Nations stand mutually
obliged by that inviolable Covenant to pursue the ends therein
expressed (which cannot be divided) against all lets and
impediments whatsoever. It is therefore our most earnest and
longing desire, That as those who are in trust with the Publick Affairs
of this Kingdom have heretofore with all earnestness and care in all
their addresses dealt with his Majesty, with much strength of Reason
and vehemency of Affection, so they would still deal with him, to
grant his Royal consent to the desires of both Kingdoms, for setling
Religion according to the Covenant, and for securing a perfect and
durable Peace (which we look upon as the only hopeful means of
preserving himself, his Crown, and Posterity) That his Majesty may
return to his Houses of Parliament in England as a reconciled Prince
to satisfied Subjects; And that acclammations of joy may be heard in
all his Majesty’s Dominions, and no sound of War heard therein any
more, except against the bloody Irish Rebels, under whose
barbarous and cruel persecution, our distressed Brethren, both in
this Kingdom and in Ireland, are still groaning and crying out to us
and to our Brethren in England, Be at peace among yourselves, and
come to help us.
Note of Proceedings in the Convention of Estates
relative to the Church, betwixt the Assemblies of
1647 and 1648.
1648.
March 27. Answers of Parliament to the desires of the
Commissioners of the General Assembly, represented by them to
Parliament, vol. vi., p. 290.
March 29. Answers of Parliament to the representation of the
Commission of the General Assembly, of their sense on the
Parliament’s Answer to their right desires, p. 291.
April 11. Act anent the Resolutions of Parliament concerning the
Breaches of the Covenant and Treaties betwixt the Kingdoms of
Scotland and England, and Demands for Reparation thereof, p. 292.
April 16. Act concerning the Desires of the Commissioners of the
General Assembly, p. 295.
April 19. A Declaration of Parliament to all his Majesty’s good
Subjects, concerning their Resolutions for Religion, King, and
Kingdoms, in pursuance of the ends of the Covenant, p. 305.
April 26. A Letter from the Parliament of Scotland to the
Parliament of England, p. 309.
April 27. Desires of the Parliament of Scotland, to the Honourable
Houses of the Parliament of England, p. 309.
May 2. Answer of the Parliament of Scotland to the
Commissioners’ Papers of the 10th and 29th of April, 1648, p. 310. A
Letter sent from the Parliament of Scotland, to the several
Presbyteries within the Kingdom, p. 321.
June 10. Act Ordaining all Ministers to exhort their People to
obedience to the Laws of the Kingdom, and assuring these Ministers
of their Stipends during their lifetime, p. 331. Act and Declaration of
Parliament, in Answer to the Supplications from Synods and
Presbyteries, p. 332.
 THE

GENERAL ASSEMBLY,
AT EDINBURGH, 1648.

The period in our history to which our attention is now attracted—

extending from the beginning of August, 1647, to the 12th of July,
1648—embraces a variety of striking occurrences, and presents, in
the progress of the great national drama which was rapidly
hastening to its tragical termination, a complication of circumstances
well calculated to affect the most sober-minded as well as the more
sanguine student. The pillars of society were then indeed shaken to
their foundations and utterly overthrown.
When we last paused in our narrative, we left the Sovereign of the
British kingdom a prisoner in the hands of a mutinous army in
England—the usurping Parliament truckling to an armed force of its
own creation—the capital of that kingdom in the possession and
under the dominion of the army—and the people suffering from an
intolerable load of burdens, and all the horrors of social anarchy. It is
unnecessary for our purpose, however, to dilate on these particulars;
and we do not intend to enter on them further than is absolutely
necessary for illustrating the relative movements in Scotland, which
it is our more immediate object to record.
In the Acts of the Assembly 1647,363 our readers will find a
Declaration and brotherly Exhortation to their brethren of England, in
which they deplore the many obstructions to the triumph of the
Covenant and Presbytery, in the variety of pestilent sectaries which
had sprung up; and, referring to the Declaration of the Scottish