Professional Documents
Culture Documents
Algorithms of Estimation For Nonlinear Systems A Differential and Algebraic Viewpoint 1st Edition Rafael Martínez-Guerra
Algorithms of Estimation For Nonlinear Systems A Differential and Algebraic Viewpoint 1st Edition Rafael Martínez-Guerra
https://textbookfull.com/product/algebraic-and-differential-
methods-for-nonlinear-control-theory-elements-of-commutative-
algebra-and-algebraic-geometry-rafael-martinez-guerra/
https://textbookfull.com/product/algebraic-and-differential-
methods-for-nonlinear-control-theory-martinez-guerra-r/
https://textbookfull.com/product/systems-of-nonlinear-partial-
differential-equations-applications-to-biology-and-engineering-
mathematics-and-its-applications-a-w-leung/
https://textbookfull.com/product/nonlinear-differential-
equations-in-micro-nano-mechanics-application-in-micro-nano-
structures-and-electromechanical-systems-1st-edition-ali-koochi/
Introduction to Data Science Data Analysis and
Prediction Algorithms with R 1st Edition By Rafael A.
Irizarry
https://textbookfull.com/product/introduction-to-data-science-
data-analysis-and-prediction-algorithms-with-r-1st-edition-by-
rafael-a-irizarry/
https://textbookfull.com/product/numerical-solutions-of-three-
classes-of-nonlinear-parabolic-integro-differential-
equations-1st-edition-jangveladze/
https://textbookfull.com/product/dynamic-estimation-and-control-
of-power-systems-1st-edition-abhinav-k-singh/
https://textbookfull.com/product/observer-design-for-nonlinear-
systems-pauline-bernard/
https://textbookfull.com/product/aspects-of-integrability-of-
differential-systems-and-fields-a-mathematical-primer-for-
physicists-costas-j-papachristou/
Understanding Complex Systems
Rafael Martínez-Guerra
Christopher Diego Cruz-Ancona
Algorithms
of Estimation
for Nonlinear
Systems
A Differential and Algebraic Viewpoint
Springer Complexity
Springer Complexity is an interdisciplinary program publishing the best research and
academic-level teaching on both fundamental and applied aspects of complex systems –
cutting across all traditional disciplines of the natural and life sciences, engineering,
economics, medicine, neuroscience, social and computer science.
Complex Systems are systems that comprise many interacting parts with the ability to
generate a new quality of macroscopic collective behavior the manifestations of which are
the spontaneous formation of distinctive temporal, spatial or functional structures. Models
of such systems can be successfully mapped onto quite diverse “real-life” situations like
the climate, the coherent emission of light from lasers, chemical reaction-diffusion systems,
biological cellular networks, the dynamics of stock markets and of the internet, earthquake
statistics and prediction, freeway traffic, the human brain, or the formation of opinions in
social systems, to name just some of the popular applications.
Although their scope and methodologies overlap somewhat, one can distinguish the
following main concepts and tools: self-organization, nonlinear dynamics, synergetics,
turbulence, dynamical systems, catastrophes, instabilities, stochastic processes, chaos, graphs
and networks, cellular automata, adaptive systems, genetic algorithms and computational
intelligence.
The three major book publication platforms of the Springer Complexity program are the
monograph series “Understanding Complex Systems” focusing on the various applications
of complexity, the “Springer Series in Synergetics”, which is devoted to the quantitative
theoretical and methodological foundations, and the “SpringerBriefs in Complexity” which
are concise and topical working reports, case-studies, surveys, essays and lecture notes of
relevance to the field. In addition to the books in these two core series, the program also
incorporates individual titles ranging from textbooks to major reference works.
Algorithms of Estimation
for Nonlinear Systems
A Differential and Algebraic Viewpoint
123
Rafael Martínez-Guerra Christopher Diego Cruz-Ancona
CINVESTAV-IPN CINVESTAV-IPN
Departamento de Control Automatico Departamento de Control Automatico
Mexico, Distrito Federal Mexico, Distrito Federal
Mexico Mexico
To my family,
Evaristo, Carmen, Junior, and Robin,
and specially to
Alejandra, and Quico and Max, of course
Christopher Diego Cruz-Ancona
Preface
The purpose of this book is to acquaint the reader with current developments in
nonlinear systems theory and its applications. This book is directed particularly
to graduate students in the fields of control of physical processes and computing,
biophysics, and physical information science. A great deal of literature related
to the control problems of nonlinear systems has been developed over the past
decades. This book proposes several algorithms motivated by systems models that
can approximate a wide class of nonlinear control systems. They can be represented
either as a state space model or an input–output equation. There exists a class of
nonlinear systems called input–affine systems or bilinear systems containing the
simplest nonlinear systems, which can be used to represent a wide range of physical,
chemical, biological, and social systems as well as manufacturing processes that
cannot be effectively modeled under linearity assumptions. The concept of bilinear
systems was introduced in the 1960s. In automatic control theory, these are control
systems whose dynamics are jointly linear in the state and multiplicative control
variables. Their theory can be developed from the theories of time-variant linear
systems and Lie groups; they have been applied to many areas of service and
technology. The purpose of this book is to discuss the development of these ideas.
We emphasize the role of several disciplines that modified our outlook on nonlinear
systems theory, among them differential-algebraic theory, the modern control theory
of dynamical systems, observers for nonlinear systems, generalized canonical
forms, and parametric and state estimation. Affine systems can approximate a wide
class of nonlinear control systems. Input–affine systems can be represented as state
space models or systems in the form
vii
viii Preface
y D Kx.t/;
1 Introduction .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 1
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 5
2 Differential Algebra in Control Theory . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 7
2.1 Elements of Differential Algebra .. . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 8
2.1.1 Transcendental Fields and Algebraic Fields,
Subfields, Differential Field Extensions .. . . . . . . . . . . . . . . . . . 8
2.2 Differential Rings and Differential Fields . . . . . . .. . . . . . . . . . . . . . . . . . . . 9
2.2.1 Differential Rings . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 10
2.2.2 Notion of a Differential Field Extension:
Algebraic and Transcendental.. . . . . . . . .. . . . . . . . . . . . . . . . . . . . 12
2.2.3 Differential Primitive Element, Differential
Transcendence Basis, Differential
Transcendence Degree, and Tower
of Differential Field Extensions.. . . . . . .. . . . . . . . . . . . . . . . . . . . 12
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 13
3 Algebraic Observability for Nonlinear Systems . . . . .. . . . . . . . . . . . . . . . . . . . 15
3.1 Introduction .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 15
3.2 Algebraic Observability .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 16
3.2.1 Nonlinear Dynamics .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 16
3.2.2 Algebraic Observability Property
in Nonlinear Systems . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 17
3.2.3 Examples .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 18
3.3 On the External Behavior of a Universally
Observable System .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 19
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 24
4 Generalized Observability Canonical Form .. . . . . . . .. . . . . . . . . . . . . . . . . . . . 25
4.1 Introduction .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 25
4.2 Differential Primitive Element .. . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 27
4.3 Single-Variable Case . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 28
xi
xii Contents
Index . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 195
List of Figures
xv
xvi List of Figures
_ Disjunction, or conditional OR
, By definition
Approximately equal
WD Defined as
Equivalent to
¤ Different from
< .>/ Less (greater) than
./ Less (greater) than or equal to
8 For all
2 Belongs to
Subset of
[ Union of sets
fg Set
! Tends to
) Implies
”
P Equivalent to, if and only if
Summation
R The set of real numbers
RC Positive real numbers
C The set of complex numbers
C Complex numbers with negative real part
Rmn The set of all m n matrices with elements from R
Rmn
C The set of all m n matrices with elements from RC
Œaij m n matrix with entries aij , 1 i m, 1 j n
Amn Matrix with m rows and n columns
det A The determinant of a square matrix A 2 Rnn
jAj The determinant of a matrix A
Tr A The trace of a matrix A
AT The transpose of a matrix obtained by interchanging the rows
and columns of A
xix
xx Symbols and Acronyms
deal of interest shown with regard to the observation problem, and examples of the
literature devoted to this topic from different perspectives for nonlinear systems and
linear systems include [1, 2, 5, 17]. Here, our perspective is different, since we adopt
a differential-algebraic approach to overcome this problem.
Given a input–affine system class, the observer synthesis is not obvious. In this
case, according to the input, we can distinguish two types of systems: those that are
observable for any input (universally observable system) and those with singular
inputs (namely, inputs that make the system unobservable, i.e., “bad inputs”). In
designing an observer, bad inputs are clearly inconvenient, because they add a
singularity problem. In the case of linear systems, the absence of singularities
explains the design and use of asymptotic observers (with arbitrary convergence
estimation speed), such as Kalman and Luenberger observers.
The first results obtained for observable systems for any input lead us to the case
in which the class of input–affine systems may be transformable to an observable
time-varying linear system with an output injection.
The observability property for nonlinear systems is associated with inputs u.t/.
Hence, the concept of “singular” inputs is implicit. That is to say, for a nonlinear
observable system, certain inputs u.t/ make unobservable all pairs of points in the
state space and complicate the synthesis of an observer. Then it is of interest to know
and describe these inputs (topological properties) with the aim of avoiding them.
Bilinear systems, besides being the simplest nonlinear systems, have great
practical importance, since many systems are modeled in this form (ecological
systems, nuclear reactors, petrochemistry, etc.). In [28], a model order reduction
algorithm for a given SISO bilinear system based on optimality conditions is
proposed. It is shown in [4, 12] that under a nonlinear transformation (e.g., via
immersion [20, 23]; output injection, etc.), it is possible to describe a nonlinear
system by an input–affine system (it may have greater dimension than the initial
system in the case of an immersion) and to reconstruct the nonlinear system state
from an observer for the input–affine system. Consequently, the study of input–
affine systems is fundamental. Now, a logical approach is to design observers for
the simplest case (input–affine systems ˙1 ) before tackling the problem of a more
general class of nonlinear systems.
Consider the class of input–affine systems that can be modeled by means of the
following relationships:
˙2 W zP D Au z C '.u; y/;
y D Cz;
where system ˙2 is observable independently of the value of the input u if and only
if z is observable with respect to fu; yg , i.e., z is algebraic over K hu; yi (z satisfies a
differential-algebraic equation on K hu; yi). Note that the system ˙2 can be viewed
as a time-varying linear system with output injection; the time-variant character
is due to matrix Au having its entries in K hui(elements of matrix Au depend on
input). Notice that the synthesis of an observer is simpler for the system ˙2 than
for the system ˙1 . So if we design an observer for the transformed system, we can
obtain estimates for system ˙1 . Therefore, a question that arises is this: what kind
of observers are valid for ˙2 ? For instance, we can think of the following observers.
Considering that ˙2 is observable, we can find Lu such that the matrix expression
(Au Lu C) has its spectrum in the complex left half- plane (.Au Lu C/ C ).
Then we get a differential-algebraic polynomial in j , u, and their time derivatives:
with a constant gain ki > 0 for 2 i n. Since zi cannot be obtained from the
model, an auxiliary system
Pi D ki i ki2 zNi1 ;
zOi D i C ki zNi1 ;
is a stable observer that can be regarded as a reduced-order observer for the ith state
of system ˙2 , where zN D .y; zO2 ; : : : ; zOn1 / is the output and the estimates of the
output derivatives. In this case, the estimation error zi zOi converges asymptotically
to a closed ball centered at the origin (compact set).
Some authors have worked with the development of state estimators for restricted
classes of nonlinear systems, either from a geometric viewpoint (theory mainly
attributed to Isidori, Hermann, Krener, Van der Schaft, Nijmeijer, Gauthier, Bornard,
Hammouri, Zeitz, Williamson, Levine and Marino, Khalil) or from a differential-
algebraic viewpoint (introduced by Fliess in 1986). Among other approaches, we
can mention the work in [26], where a method to synthesize an observer for an
uncertain Takagi–Sugeno fuzzy system is proposed; the case of observers for time
delay input–affine systems is studied in [9]; in [15], a predictor input–affine observer
is developed for a class of nonlinear systems subject to delayed outputs; [24]
considers the problem of optimal input signal estimation for input–affine systems
under output measurements and parametric estimation.
The foundations of the algebraic approach are based on differential algebra [7],
from which it has been attempted to capture the algebraic nature of the observability
concept. Preliminary work in this direction has been published [6, 13, 25].
In [18] and [19], the authors characterized input–affine systems and in general
nonlinear systems. These are carried out by means of a diffeomorphism into some
nonlinear canonical form of the type given in ˙2 .
For input–affine systems, asymptotic observers are found in some works such as
Hara–Furuta [14] and Funahashi [8], where some sufficient conditions are given in
order to guarantee asymptotic convergence of the estimation error.
In [27], Williamson shows that input–affine systems without singular inputs have
an observability canonical form, and from that structure he gives an appropriate
observer. A generalization to the former result is given by J.P. Gauthier and G.
Bornard [10] for observable nonlinear systems with respect to any input.
Using the nonlinear canonical form, it is shown in [11] that those systems have
a high-gain observer. An extension for observable nonlinear systems regarding any
input are found in [3].
In general, input–affine systems, modulo an output injection, have singular
inputs. Despite this fact, there are observers that work for some classes of inputs.
These observers are a modification of the Kalman filter in the deterministic case.
References 5
References
1. Balatif, et al.: Optimal control problem for a class of bilinear systems via block pulse functions.
IMA J. Math. Control Inform. (2016), dnw005
2. Besançon, G.: Nonlinear Observers and Applications. Lecture Notes in Control and Informa-
tion Sciences. Springer, New York (2007)
3. Bornard, G., Hammouri, H.: A high gain for a class of nonlinear systems under locally regular
inputs. In: IEEE Conf. on Dec. and Control CDC Brighton G.B. (1991)
4. Bornard, G., Couenne, N., Celle, F.: Regularly persistent observers for bilinear systems.
In: Proceedings of the 29 International Conference on nonlinear Systems. New Trends in
Nonlinear System Theory, Vol 122. Springer, New York (1988)
5. Fan, L., Wang, W.: A GPI based sliding mode control method for boost DC-DC converter.
IEEE Int Conference on Industrial Technology, pp. 1826–1831, March, 2016
6. Fliess, M.: Quelques Remarques sur les observateurs nonlinéaires. Proc. Colloque Gretsi Trait.
Signal Images, Nice pp. 169–172, 1987
7. Fliess, M.: Automatique et Corps différentiel. Forum Math. 1, 227–238 (1989)
8. Funahashi, Y.: Stable state estimator for bilinear systems. Int. J. Control 29(2), 181–188 (1979)
9. Gamiochipi, J., Ghanes, M., Aggome, W., De Leon, J., Barbot, J-P.: Observer design for a class
of nonlinear systems under a persistent excitation. In: 10th IFAC Symposium on Nonlinear
Control Systems, NOLCOS, 2016
10. Gauthier, J.P., Bornard, G.: Observbility for any u(t) of a class of nonlinear systems. IEEE
Trans. Automatic Control AC-26(4), 922–926 (1981)
11. Gauthier, J.P., Hammouri, H., Othman, S.: A simple observer for nonlinear systems applica-
tions to bioreactors. IEEE Trans. Automatic Control 36(6), 875–880 (1992)
12. Gauthier, J.P., Kazakos, D.: Observabilite et observateurs de systemes non lineaires. RAIRO
APII Analyse des Systemes 21, 201–212 (1987)
13. Glad, S.T.: Differential algebraic modelling of nonlinear systems. In: Kaashoek, M.A., Van
Schuppen, J.H., Ran, A.C.M. (eds.), Realization and Modelling in System Theory MTNS-89,
pp. 97–105. Birkhäuser, Boston (1989)
14. Hara, S., Furuta, K.: Minimal order state observers for bilinear systems. Int. J. Control. 24(5),
705–718 (1976)
15. Ibrir, S., Hunte, K., Bettayeb, M.: State observation of a class of bilinear system with large
sensor delay. IFAC Procc. 47(3), 5023–5028 (2014)
16. Kalman, R.E., Bucy, R.: New results in linear filtering and prediction theory. J. Basic Eng. 82
D, 35–40 (1960)
17. Korovin, S.K., Fomichev, V.V.: State observers for linear systems with uncertainty. De Gruyter
Expositions in Mathematics 51, p. 242 (2009)
18. Krener, A.J., Isidori, A.: Linearization by output injection and nonlinear observers. Syst.
Control Lett. 3, 47–52 (1983)
19. Krener, A.J., Respondek, W.: Nonlinear observer with linear error dynamics. SIAM J. Control
Optim. 23, 197–216 (1985)
20. Levine, J., Marino, R.: Nonlinear systems immersion observers and finite dimension filters.
Syst. Control Lett. 7, 133–142 (1986)
21. Luenberger, D.G.: Observers for multivariable systems. IEEE Trans. Automatic Control 11,
190–197 (1966)
22. Martínez-Guerra, R.: Observer synthesis for a class of bilinear systems. In: 32nd Annual
Allerton Conference on Communication Control and Computing, pp. 523–524, 1994
23. Martínez-Guerra, R.: Inmersión de un sistema no lineal y construcción de un observador no
lineal para un sistema TPA. In: Congreso Latinoamericano de Control Automático, Habana,
Cuba, pp. 251–255, 1992
24. Pardalos, P.M., Knopov, P.S., Urysev, S.P., Yatsenko, V.S.; Optimal estimation of signal
parameters using bilinear observation. In: Rubinov, A., Glover, B. (eds.), Optimization and
Related topics. Kluwer Academic, Boston (2001)
6 1 Introduction
25. Pommaret, J.F.: Géométrie différentielle algébrique et théorie du contrôle. C.R. Acad. Sc. Paris,
t. 302(Serie I, No.15), 547–550 (1986)
26. Vu, V.P., Wang, W.J.: Observer synthesis for uncertain Takagi–Sugeno fuzzy systems with
multiple output matrices. IET Control Theory Appl. 10(2), 151–161 (2016)
27. Williamson, D.: Observation of bilinear systems with application to biological control.
Automatica 13, 243–254 (1977)
28. Yang, P., Xu, K.L., Jiang, Y.L.: H2 model order reduction for bilinear systems based on the
cross gramian. IMA J. Math. Control Inform. 32, 1–16 (2016)
Chapter 2
Differential Algebra in Control Theory
This chapter presents some elementary definitions and ideas from differential
algebra on which this book is based.
In automatic control, the use of techniques from differential algebra are increas-
ingly important. This is the main reason why it is not only necessary to remember
the useful elements of this language. In addition, such techniques will serve as a
descriptive language to prove various lemmas and corollaries related to this theory.
Some contributions to theory can be found in [3, 6, 11].
Roughly speaking, differential algebra is a theory of structures that generalizes
the structures that have been used in traditional mathematics and its applications
for a long time. Such algebraic structures include rings, fields, field extensions, and
vector spaces (see, for example, [2]).
Fundamental differential-algebraic structures are essentially differential rings,
differential fields, differential field extensions, etc. These have corresponding
underlying structures that are usual algebraic structures. Their construction is given
generally by adding operations to the underlying structures, that is, additional laws
such as derivation, subject to consistency axioms yielding richer structures that are
also more complex than the original structures.
The key concepts in this chapter are differential polynomials and their handling
as well as elementary structures from differential algebra such as differential rings,
differential fields, and differential field extensions.
Definition 2.3 Let L be a field extension of the field K and let .xi /i2I be a family of
elements of L with I a set of indices. The family .xi /i2I is said to be algebraically
dependent over K if there exist a finite subset fi1 ; : : : ; in g of I and a nonzero
polynomial P of AŒxi1 ; : : : ; xin , the ring of polynomials, such that P .xi1 ; : : : ; xin / D
0. Such a family is said to be algebraically independent over K if every nonzero
polynomial in AŒ.xi /i2I is zero for .xi /i2I (this is the exact analogue of vectorial
linear independence in vector spaces).
Definition 2.5 An extension L of a field K it is called algebraic if all its elements are
algebraic over K, and it is called transcendental if there exists at least one element
that is transcendental over K.
a0 C a1 s C C an sn
f .s/ D ; (2.1)
b0 C b1 s C C bm sm
C W AA!A
.a; b/ 7! a C b
10 2 Differential Algebra in Control Theory
and
W AA!A
.a; b/ 7! a b
also satisfy
@ .a C b/ D @a C @b
and
@ .a b/ D .@a/ b C a .@b/ :
Definition 2.9 A differential ring A is a commutative ring with a unit element that is
provided with a finite set of differentiation operators over A such that 8@1 ; @2 2 ,
8a 2 A,
@1 @2 a D @2 @1 a:
Definition 2.10 A differential ring is called ordinary or partial if the finite set
respectively contains a single or more than one differentiation operators. These are
useful in the study of ordinary algebraic differential equations and partial algebraic
differential equations, respectively.
Definition 2.11 A differential subring of a differential ring A is a subring of A such
that with the differential operator in , A is a differential ring.
Definition 2.12 A differential subring is a differential ring that is also a differential
field.
2.2 Differential Rings and Differential Fields 11
a C 0 D a:
@ .a C 0/ D @a C @ .0/ D @ .a/ :
Hence
@ .0/ D 0:
a D a 1;
we obtain
@a D @ .a 1/ D 1 @a C a@1 D @a C a@1:
Hence
@.1/ D 0:
12 2 Differential Algebra in Control Theory
Definition 2.14 If L and K are two differential fields such that K is a differential
subfield of L (K L), then L is called a differential field extension of K.
Definition 2.15 The intersection of a set of differential subfields of L is also a
differential subfield of L.
Definition 2.16 An element a 2 L is called differentially algebraic over K if
there exists an algebraic differential equation with coefficients in K, i.e., there
exists a polynomial with coefficients in K, P x; xP ; : : : ; x.˛/ , with a finite number
of derivatives of x such that P a; aP ; : : : ; a.˛/ D 0.
To illustrate the above definition, consider the following examples.
Example 2.7 K D Q, a D et 2 L, satisfies the differential equation xP x D 0.
Example 2.8 K D R. Let L be the field R.x/, and xP D e , xR D xP e . The element x
x x
531-533
Symptoms,
514-521
515-517
521
General,
520
Lock-spasm in,
517
519
Paresis and paralysis, characters and seat,
518
Tremor in,
518
520
Treatment,
533
543
Arsenic, use,
537
538
Atropia, hypodermically,
537
Electricity, use,
534-536
536
537
538
537
540-542
Rest in,
533
534
Strychnia, use,
537
538
538
540
Z.
in hysteria,
278
in hystero-epilepsy,
313
1232
GENERAL INDEX.
A.
Alcoholism, v. 573
rheumatoid, ii. 78
Athetosis, v. 457
Atrophy, v. 1266
B.
Beriberi, i. 1038
inflammation of, acute and chronic, iv. 123, 126, 128, 341
C.
Caisson Disease, the, iii. 855
renal, iv. 42
Cardiac murmurs and their relation to valvular heart disease, iii. 651
Catalepsy, v. 314