Appendix

Selected Problems of Controllability and Observability

of Linear Systems

A.1 Reachability
Consider the following discrete-time linear system

xi 1 Axi  Bui , i 0, 1, ... , (A.1a)

yi Cxi  Dui , (A.1b)

where xi n is the state vector, ui m

the input vector, yi p
the output vector;
A nun, B num, C pun, D pum.

Definition A.1. The system (A.1) (or the pair (A,B)) is called reachable, if for
every vector xf n there exists an integer q > 0 and a sequence of inputs {ui,
i=0,1,…,q1} such that for x0 = 0, xq = xf.

Theorem A.1. The system (A.1) is reachable if and only if one of the following
conditions is met
1.

rank [B, AB,..., A n1B] n , (A.2)

2.

rank [I n z  A, B] n, for all finite z  , (A.3)

474 Appendix

3.

[I n z  A] and B are left coprime matrices. (A.4)

Proof. Using the solution

i 1
xi A i x0  ¦ A i k 1Buk
k 0

to equation (A.1a) for i = n, x0= 0 and taking into account that xn = xf, we obtain

ªun1 º
« »
n 1 u
xf ¦A n  k 1
Buk ª¬B, A, B,..., A n1B º¼ « n2 » . (A.5)
k 0
«# »
« »
¬ u0 ¼

From (A.5) it follows that for every xf there exits {ui, i=0,1,…,q1} if and only
if the condition (A.2) is met.
Let v n be a vector such that vTB = 0 and vTA = zvT for a certain complex
variable z. In this case,

vT AB zvT B 0, vT A 2 B zvT AB 0, ..., vT A n1B 0,

that is,

vT ª¬B, AB, ..., A n1B º¼ 0.

The condition (A.2) thus implies v = 0. Hence from

vT > I n z  A, B @ 0

(A.3) follows.
From (A.3) it follows that there exists a unimodular matrix

ª U1 U 2 º n  mu( n  m )
U «U U »   [ z] ,
¬ 3 4¼

such that

> Iz  A , B @ U > I n 0 @
 Appendix 475

and

>Iz  A@ U1  BU3 In . (A.6)

Thus [Iz – A] and B are left coprime matrices.

Let

u1i u10i  u11i z  u12i z 2    u1ni2 z n2 (i 1, ..., n),

(A.7)
u3i u30i  u31i z  u32i z 2    u3ni1 z n1

be the i-th columns of the polynomial matrices U1 nun[z] and U3 mun[z],
respectively. Substituting (A.7) into (A.6) and comparing the coefficients by the
same powers of the variable z , we obtain

Bu30i  Au10i ei
Bu  u  Au11i
1
3i
0
1i 0
.......................................... for i 1, ..., n , (A.8)
n2 n 3 n2
Bu 3i u 1i  Au 1i 0
n 1 n 2
Bu 3i u 1i 0

where ei is the i-th column of the identity matrix In.

Pre-multiplying the equations in (A.8) successively by A0,A1,A2,….,An1 and
adding them up, we obtain

Bu30i  ABu31i    A n1Bu3ni1 ei (i 1, ..., n)

and

ªu30i º
« 1 »
u
ª¬B, AB,..., A n1B º¼ « 3i » ei (i 1, ..., n). (A.9)
« »
« »
«¬u3ni1 »¼

(A.9) implies the condition (A.2). The conditions (A.2), (A.3), and (A.4) are thus
equivalent.

476 Appendix

If the system (A.1) is not reachable, then the set of reachable states from the
point x0 = 0 is given by the image of the matrix [B,AB,…,An-1B].

Example A.1.
Show that the pair

ª0 1 0 " 0 º ª0º
«0 0 1 " 0 »» «0»
« « »
A «# # # % # », B « #» (A.10)
« » « »
«0 0 0 " 1 » «0»
«¬ a0 a1 a2 " an1 »¼ «¬1 »¼

is reachable for arbitrary values of the coefficients a0,a1,…,an1.

Using (A.3), we obtain

ª z 1 0 " 0 0º
« 0 z 1 " 0 0 »»
«
rank > I n z  A, B @ rank « # # # % # #» n, (A.11)
« »
« 0 0 0 " 1 0»
«¬ a0 a1  a2 " z  an1 1 »¼

for all finite z .

The last n columns of (A.11) are linearly independent whatever the values of
the coefficients a0,a1,…,an-1.

Now consider the following continuous-time linear system

§ dx ·
x Ax  Bu ¨ x ¸, (A.12a)
© dt ¹
y Cx  Du , (A.12b)

where x = x(t) n is the state vector, u = u(t) m the input vector, y = y(t) p
the
output vector; A nun, B num, C pun, D pum.

Definition D.2. The system (A.12) (or the pair (A,B)) is called reachable if for
every vector xf n there exists a time tf > 0 and an input u(t) over the interval
[0, tf] such that for x0 = 0, x(tf) = xf.

Theorem D.2 The system (A.12) is reachable if and only if one of the following
conditions is satisfied:
 Appendix 477

1.

rank [B, AB, ..., A n1B] n , (A.13)

2.

rank [I n s  A, B] n, for all finite s  , (A.14)

3.

>I n s  A @ and B are left prime. (A.15)

The proof is similar to that of Theorem A.1.



A.2. Controllability
Definition A.3. The system (A.1) (or the pair (A,B)) is called controllable to zero
if for an arbitrary initial state x0 z 0 there exists an integer q > 0 and a sequence of
inputs {ui, i=0,1,…,q1} such that xq = 0.

Theorem A.3. The system (A.1) is controllable to zero if and only if one of the
following conditions is met:
1.

Im A n  Im [B, AB,..., A n1B] , (A.16)

2.

rank [I  dA, B] n, for all finite d  , (A.17)

3.

[I  dA] and B are left coprime. (A.18)

Proof. Using the solution to (A.1), for i = n, xn = 0 we obtain

ªun1 º
« »
n 1 u
A n x0 ¦ A nk 1Buk  ª¬B, AB,..., A n1B º¼ « n2 » . (A.19)
k 0
«# »
« »
¬ u0 ¼
478 Appendix

From (A.19) it follows that there exists a sequence of inputs {ui, i=0,1,…,q1}
for an arbitrary x0 if and only if the condition (A.16) is met.
Let v n be a vector such that vTB = 0 and vTA = zvT for a certain variable z. In
the same manner as in the proof of Theorem A.1, we obtain vT[B,AB,…,An-1B] = 0.
The condition (A.16) implies

0 vT A n O n vT and thus O 0 or v 0.

Hence the matrix [Inz – A, B] has full row rank n for all finite z z 0, which is
equivalent to the conditions (A.17).
Analogously to the proof of Theorem A.1 one can show that the condition
(A.17) implies (A.18), and the condition (A.18) in turn implies (A.16).


Remark A.1.
Each of the conditions (A.13), (A.14) and (A.15) for the system (A.1) with singular
A is only a sufficient condition, but not a necessary one for the controllability of
the system. If det A z 0, then these conditions are also necessary conditions for the
controllability of (A.1). For the system (A.1) with nonsingular A, the conditions of
its controllability are equivalent to the conditions of its reachability.

Example A.2.
The pair of matrices

ª0 a º ª1 º
A «0 0 » , B «0 » (A.20)
¬ ¼ ¬ ¼

is not reachable, since

ª z a 1 º
rank > Iz  A, B @ rank « » 1, for z 0.
¬0 z 0¼

On the other hand, using (A.17), we obtain

ª1 da 1 º
rank > I  dA, B @ rank « » 2 for arbitrary a and d .
¬0 1 0 ¼

The pair (A.20) is thus controllable for arbitrary a.

Note that in this case the state

ª0 º
xf «1 »
¬ ¼

is not reachable from the state x0 = 0, since x0 does not belong to

 Appendix 479

ª1 º
Im[B, AB] Im « » .
¬0¼

On the other hand, the state

ª0 º
x0 «1 »
¬ ¼

can be brought to zero by the zero input sequence u0 = u1 = 0, since A2 = 0 for

arbitrary a.

Definition A.4. The system (A.12) (or the pair (A,B)) is called controllable to zero
if for an arbitrary initial state x0 there exists a time tf > 0 and an input u = u(t) over
the interval [0, tf]] such that x(tf) = 0.

Theorem A.4. The system (A.12) is controllable to zero if and only if one of the
conditions (A.16), (A.17), (A.18) of Theorem A.3 is met.

The proof of this theorem follows similarly to that of Theorem A.3.



Using the solution to (A.12a), for x(0) = x0, x(tf) = 0, we obtain

tf tf
At f A (t f W )
xf e x0  ³e Bu (W ) dW 0 and x0  ³ e AW Bu (W ) dW
0 0

since eAt is a nonsingular matrix regardless of the matrix A. Hence the

controllability of a continuous-time system is equivalent to its reachability for
every A.

Example A.3.
We choose as the state variable x the voltage uc on the capacity of the electrical
circuit in Fig. A.1, and as the input the source voltage u. Note that the voltage uc on
the capacity is zero for an arbitrary value of the source voltage u. Therefore
changing u we cannot reach any desired nonzero value of the voltage uc = xf z 0.
Thus this circuit is an example of an uncontrollable system.
480 Appendix

Fig. A.1. Uncontrollable electrical circuit

A.3 Observability
First consider the discrete system (A.1).

Definition A.5. The system (A.1) (or the pair (A,C)) is called observable if there
exists an integer q > 0 such that for given sequences of inputs {ui, i=0,1,…,q-1}
and outputs {yi, i=0,1,…,q-1} one can determine the initial state x0 of this system.

Theorem A.5. The system (A.1) is observable if and only if one of the following
conditions is met:
1.

ª C º
« CA »
rank « » n, (A.21)
« # »
« n 1
»
¬«CA ¼»

2.

ªI z  A º
rank « n » n for all finite z   , (A.22)
¬ C ¼

3.

>I n z  A @ and C are right coprime. (A.23)

Proof. Substituting the solution of (A.1a) into (A.1b), we obtain

 Appendix 481

i 1
yic yi  Dui  ¦ CA i k 1Buk CA i x0 . (A.24)
k 0

Using (A.24), for i 0,1,..., n  1 , we have

ª y0c º ª C º
« c» « »
« y1 » « CA » x . (A.25)
« # » « # » 0
« » « n 1
»
¬ ync 1 ¼ ¬«CA ¼»

For the given sequences {ui, i=0,1,…,q1},{yi, i=0,1,…,q1} the sequence {y’i,
i=0,1,…,n1} is known. From (A.25) we can determine x0 if and only if the
condition (A.21) is met.

Equivalence of the remaining conditions can be proved similarly (dually) as in

Theorem A.1.


Example A.4.
Show that the pair

ª A1 0º
A «A A 3 »¼
, C >C1 0@
¬ 2

is not observable for arbitrary submatrices A1rur, A2(n-r)ur, A3(n-r)u (n-r), C1pur.
It is easy to verify that

ª A1k 0 º
Ak « », (A.26)
¬« * A 3k ¼»

where * denotes a submatrix insignificant in the following considerations.

Using (A.21) and (A.26), we obtain

ª C º ª C1 0º
« CA » «CA »
« » « 1 1 0» . (A.27)
« # » « # #»
« » « n 1 »
¬C1A1 0¼
n 1
«¬CA »¼

From (A.27) it follows that the condition (A.21) is not met for arbitrary
A1,A2,A3 and C1.
482 Appendix

Definition A.6. The system (A.12) (or the pair (A,C)) is called observable if there
exists a time tf > 0 such that for given u(t) and y(t) for 0 d t d tf, one can determine
the initial state x0 of this system.

Theorem A.6. The system (A.12) is observable if and only if one of the conditions
(A.21), (A.22), (A.23) of Theorem A.5 is met.

The proof of this theorem follows similarly to that of Theorem A.5.



Example A.5.
We take as the state variables in the circuit in Fig. A.2 the voltage uC on the
capacity and the current in the coil iL; as the input u we take the source current i,
and as the output y we take the voltage uR on the resistance R, uR = Ri. The circuit
is described by the following equations

ª 1º
«0  ª1º
d ª uc º C » ª uc º « » ª uc º
« » « » « »  C i, y [0 0] « »  Ri
dt ¬iL ¼ «1 i « » ¬iL ¼
0» ¬ L ¼ ¬ 0 ¼
¬« L ¼»

Fig. A.2. Unobservable electrical circuit

The circuit is not observable, since

ªC º
C [0 0] and thus « » 0.
¬CA ¼

Note that with both the source current i and the voltage uR known, we cannot
determine the initial state

ªuc (0) º
« »
¬iL (0) ¼

of this circuit.
 Appendix 483

It is easily verifiable that if we choose the voltage on the capacity as the output
y, then the circuit is observable.

A.4 Reconstructability
First consider the discrete-time system (A.1).

Definition A.7. The system (A.1) (or the pair (A,C)) is called reconstructable if
there exists an integer q > 0 such that for the two given sequences: input {ui,
i=0,1,…,q1} and output {yi, i=0,1,…,q1} one can determine the state vector xq
of this system.

Theorem A.7. The system (A.1) is reconstructable if and only if one of the
following conditions is met
1.

ª C º
« CA »
Ker « »  Ker A n , (A.28)
« # »
« n 1
»
¬«CA ¼»

2.

ªI  dA º
rank « n » n for all finite d  , (A.29)
¬ C ¼

3.

> I n  dA @ and  are right coprime. (A.30)

Proof of this theorem is analogous (dual) to that of Theorem A.5.

Example A.6.
The pair

ª1 1º
A «2 , C [1 1] (A.31)
¬ 2 »¼

is not observable, since

484 Appendix

ª Cº ª1 1º
rank « » rank « 1.
¬CA ¼ ¬3 3 »¼

One cannot determine the vector x0 = [x01, x02]T with y0 and y1 known for
u0 = u1 = 0, since y0 = Cx0 = x01 + x02, y1 = 3(x01 + x02), that is, we know only the
sum x01 + x02.
The pair (A.31) is reconstructable, since

ª1  d d º
ª I  dA º
rank « n » rank «« 2d 1  2d »» 2, for all finite d  .
¬ C ¼
«¬ 1 1»¼

From the equation

ª 3 y0 º
x2 A 2 x0 « »
¬ 2 y1 ¼

we can compute x2 with y0 and y1 known.

Remark A.2.
Each of the conditions (A.21), (A.22), (A.23) for the system (A.1) with A singular
is only a sufficient condition and not a necessary one for the reconstructability of
this system. If det Az 0, then these conditions are also necessary ones of the
reconstructability of the system (A.1). For the system (A.1) with A nonsingular, the
conditions of observability are equivalent to those of reconstructability.

Definition A.8. The system (or the pair (A,C)) is called reconstructable if there
exists a time tf > 0 such that with u(t) and y(t) given for 0 d t d tf one can determine
the state vector xf = x(tf) of this system.

Theorem A.8. The system (A.12) is reconstructable if and only if one of the
conditions (A.21), (A.22), (A.23) of Theorem A.5 is satisfied.

The proof of this theorem is analogous to that of Theorem A.5.

The reconstructability of the continuous-time system (A.12) is equivalent to its

observability.
 Appendix 485

A.5 Dual System

Definition A.9 The system

xi 1 AT xi  CT ui ,
(A.32)
yi BT xi

is called dual with respect to the system

xi 1 Axi  Bui ,
(A.33)
yi Cxi .

By virtue of Theorem A.1 and A.5 the following result ensues.

Theorem A.9. The system (A.33) is reachable (observable) if and only if its dual
(A.32) is observable (reachable).

The same theorem applies to the continuous-time system (A.12).

6 Stabilizability and Detectability

Consider the discrete-time system (A.1). Let z1,z2,…,zn be the eigenvalues of the
matrix A of this system.

Definition A.10. The eigenvalue zi of the system (A.1) is called controllable if

rank > I n zi  A, B @ n (i 1, ..., n) , (A.34)

The system (A.1) is reachable if and only if all the eigenvalues z1,z2,…,zn are
controllable.

Definition A.11. The system (A.1) is called stabilizable if all the unstable
eigenvalues |zi| t 1 of this system are controllable.

Theorem A.10. The system (A.1) is stabilisable if and only if

rank > I n z  A, B @ n, for all z t 1. (A.35)

The proof of this theorem is analogous to that of Theorem A.1.

486 Appendix

Definition D.12. An eigenvalue zi is called observable if

ªI z  A º
rank « n i » n (i 1,..., n) . (A.36)
¬ C ¼

The system (A.1) is observable if and only if all the eigenvalues z1,z2,…,zn are
observable.

Definition A.13. The system (A.1) is called detectable if all the unstable
eigenvalues (|zi| t 1) of this system are observable.

Theorem A.11. The system (A.1) is detectable if and only if

ªI z  A º
rank « n » n, for all z t 1. (A.37)
¬ C ¼

The proof of this theorem is analogous to that of Theorem A.5.

Note that reachability (observability) always implies stabilisability

(detectability) of the system (A.1).
With merely slight modifications, the foregoing considerations apply to
continuous-time systems of the form (A.12).
References

[1] AFRIAT S. N.: The quadratic form positive defined on a linear manifold. Proc.
Cambridge Phil. Soc. 1951, vol. 47, pp. 16
[2] ALIZADEH F.: Combinatorial optimization with semidefinite matrices. Proc.
of 2nd Annual Integer Programming and Combinatorial Optimization
Conference, Carnegie–Mellon University, 1992.
[3] ANDERSON B.: A system theory criterion for positive real matrices, SIAM
J. Control, 5, 1967, pp. 171182.
[4] ANTSAKLIS P. J., GAO Z.: Polynomial and rational matrix interpolation:
theory and control applications. Int. J. Control, 1993, vol. 58, No 2,
pp. 349404.
[5] ANTSAKLIS P. J., WOLOVICH W. A.: The canonical diophantine equations
with applications, SIAM J. Contr. Optimiz., vol. 22, No 5, 1984, pp. 77787.
[6] ARNOLD W. F., LAUB A. J.: Generalized eigenproblem algorithms and soft-
ware for algebraic Riccati equations, Proc. IEEE, vol. 72, No 12, 1984, pp.
17461754.
[7] AYRES F.: Theory and problems of matrices. Schaum, New York 1962.
[8] BACHMANN W., HAACKE R.: Matrizenrechnung für Ingenieure, Springer,
Berlin Heidelberg New York 1982.
[9] BALAKRISHNAN V., HUANG Y., PACKARD A., DOYLE J.: Linear matrix
inequalities in analysis with multipliers, In Proc. American Control Conf.,
1994. Invited session on Recent Advances in the Control of Uncertain
Systems.
[10] BANASZUK A, KOCIECKI M., Observability with unknown input and dual
properties for singular systems, J.C. Baltzer AG, Scientific Publishing Co,
IMACS, 1991, pp. 125129.
[11] BARNETT S.: Matrices in control theory. Van Nostrand Reinhold Company,
London 1960.
[12] BARNETT S.: Polynomials and linear control systems. Marcel Dekker, New
York, 1983.
[13] BELLMAN R.: Introduction to matrix analysis. McGraw–Hill, New York
1960.
488 References

[14] BELLMAN R., FAN K.: On systems of linear inequalities in Hermitian matrix
variables, In V. L. Klee, editor, Convexity, vol. 7 of Proc. of Symposia in Pure
Mathematics, American Mathematical Society, 1963, pp. 1–11.
[15] BENALLOU A., MELLICHAMP D. A., SEBORG D. E.: On the number of
solutions of multivariable polynomial systems. IEEE Trans. Auto. Control
AC-28, No 2, 1983, pp. 224–227.
[16] BENVENUTI L. FARINA L., A tutorial on the positive realization problem.
IEEE Trans. Autom. Control, vol. 49, No 5 pp. 651–664.
[17] BENVENUTI L. FARINA L., A tutorial on the positive realization problem. IEEE
Trans. Autom. Control, vol. 49, No 5, 2004, pp. 651–664.
[18] BERMAN A., NEUMANN M., STERN R.: Nonnegative Matrices in Dynamic
Systems, Wiley–Interscience, New York, 1989.
[19] BERMAN A., PLEMMONS R. J.: Nonnegative matrices in the mathematical
sciences. Computer Science and Applied Mathematics, Academic Press,
1979.
[20] BODEWIG E.: Matrix calculus. North–Holland, Amsterdam, 1965.
[21] BORNE P., TZAFESTAS S.: Multidimensional polynomial matrix equations.
Applied Modelling and Simulation of Technological Systems, Proc. of the 1st
IMACS, Lille–France, 3-6 June 1986, pp. 647–653.
[22] BOSÁK M., GREGOR J.: On generalized difference equations. Aplikace
Matematiky, vol. 32, No 3, 1987, pp. 224–239.
[23] BOYD S., EL GHAOUI L., FERON E., BALAKRISHAN V.: Linear Matrix
Inequalities in System and Control Theory. In Proc. Annual Allerton Conf.
on Communication, Control and Computing, Allerton House, Monticello,
Illinois, October, 1993.
[24] BOYD S., EL GHAOUI L., FERON E., BALAKRISHAN V.: Linear Matrix
Inequalities in System and Control Theory. SIAM, Philadelphia, 1994.
[25] BRIERLEY S., LEE E.B.: Solution of the equation A(z)X(z)+X(z)B(z)=C(z) and its
application to the stability of generalized linear systems. Int. J. Control, vol. 40, No 6,
1984, pp. 1065–1075.
[26] BRUNOWSKY P.: A classification of linear controllable systems. Kybernetika
cislo 3. 1970, vol. 6, pp. 173–187.
[27] BUSàOWICZ M., Explicit solution of discrete-delay equations. Foundations
of Control Engineering, vol. 7, No. 2, 1982, pp. 67–71.
[28] BUSàOWICZ M., Odporna stabilnoĞü ukáadów dynamicznych liniowych
stacjonarnych z opóĨnieniami. Warszawa–Biaáystok, 2000.
[29] BUSàOWICZ M., KACZOREK T.: Reachability and minimum energy control
of positive linear discrete-time systems with one delay. 12th Mediterranean
Conference on Control and Automation, June 6-9, 2004, Kusadasi, Izmir,
Turkey.
[30] BUSàOWICZ M., KACZOREK T.: Stability and robust stability of positive linear
discrete-time systems with pure delay. 10th IEEE Int. Conf. on Methods and Models
in Automation and Robotics, 2004, 30.08-2.09.2004, MiĊdzyzdroje, pp.105–108
[31] BUSàOWICZ M., KACZOREK T.: Robust stability of positive discrete-time interval
systems with time-delays. Bull. of the Poli. Acad. Sci, Techn. Sci, vol. 52, No 2,
2004, pp. 99–102
[32] BUSàOWICZ M., KACZOREK T.: Recent developments in theory of positive
discrete-time linear systems with delays–stability and robust stability. Pomiary,
Automatyka, Kontrola, No 9, 2004, pp. 9–12
 References 489

[33] CAMPBELL S. L., MEYER C. D. Jun., ROSE N. J.: Applications of the Drazin
inverse to linear systems of differential equations with singular constant
coefficients. SIAM J. App. Math. 1976, vol. 31, No 3, pp. 411–425.
[34] CAMPBELL S. L., NICHOLS N. K., TERRELL W. J.: Duality, observability and
controllability for linear time-varying descriptor systems. Circuits Systems
Signal Process, vol. 10, No 4, 1991, pp. 455–470.
[35] CAMPBELL S. L.: Singular systems of differential equations. Pitman
Advanced Publishing Program, pp. 138–143.
[36] CHANG F. R., CHEN H. C.: The generalized Cayley–Hamilton theorem for
standard pencils. Systems and Control Letters 18, 1992, pp. 179–182.
[37] CHOLEWICKI T.: Macierzowa analiza obwodów liniowych. PWN, Warszawa
1958.
[38] ÇIFTÇIBASI T., YĥKSEL ė.: On the Cayley-Hamilton theorem for two-
dimensional systems. IEEE Trans. Auto. Control AC-27, 1982, pp. 193–194.
[39] COBB D.: Controllability, observability and duality in singular systems. IEEE
Trans. Automat. Contr., vol. AC-29, No 12, 1984, pp. 1076–1082.
[40] COXON P. G.: Positive input reachability and controllability of positive
systems. Linear Algebra and its Applications 94, 1987, pp. 35–53.
[41] DAI L.: Singular Control Systems. Lectures Notes in Control and Information
Sciences, Springer. Berlin Heidelberg New York, 1989.
[42] DATA B. N.: Numerical methods for linear control systems: design and
analysis. Elsievier Academic Press, 2004.
[43] DIETRICH G., STAHL H.: Grudzüge der Matricenrechnung.VEB, Leipzig
1974.
[44] DORF R.C.: Modern control systems. Reading, Mass, Addison-Wesley 1967.
[45] DUAN G. R.: Solution to matrix equation AV+BW=EVF and eigenstructure
assignment for descriptor systems. Automatica, vol. 28, No 3, 1992, pp. 639–
643.
[46] DZIELIēSKI A.: Approximate Model Application to Geometric Nonlinear
Control. 7th IEEE International Conference Methods and Models in
Automation and Robotics, 2001, MiĊdzyzdroje (Poland), August 28-31,2001.
[47] DZIELIēSKI A.: MATLAB Package for N-D Sampling based Modeling of
Nonlinear Systems. W: Multidimensional Signals, Circuits and Systems. Eds.
K. Gaákowski and J. Wood, Taylor and Francis, London, 2001, pp. 29–44.
[48] DZIELIēSKI A.: New Models for Geometric Non-linear Control. 13th
International Conference on Process Control, PC’01, Štrbské Pleso
(Slovakia), June 11-14, 2001.
[49] DZIELIēSLI A., GRANISZEWSKI W. Optimal neurons number method for
nonlinear systems modeling. Proceedings of IEEE International Conference
on Computational Cybernetics, ICCC’03, Siofok (Hungary), August 29-31,
2003.
[50] EL BOUTHOURI A., PRITCHARD A.: A Riccati equation approach to
maximizing the stability radius of a linear system under structured stochastic.
Lipschitzian perturbations, Syst. Control Letter, January 1994.
[51] FARINA L., RINALDI S.: Positive Linear Systems; Theory and Applications.
J. Wiley, New York 2000.
490 References

[52] EMRE E., SILVERMAN L. M.: The equation XR+XY=); a characterization of

solution. SIAM J. Contr. Optimiz., vol. 19, No 1, 1981, pp. 33–38.
[53] EMRE E.: The polynomial equation QQc+RPc=) with application to
dynamic feedback. SIAM J. Control and Optimization. November 1980, vol.
18, No 6, pp. 611–620.
[54] FADDIEJEWA W. N.: Metody numeryczne algebry liniowej. PWN, Warszawa
1955.
[55] FANG Ch.-H.: A simple approach to solving the diophantine equation. IEEE
Trans. Auto. Control vol. 37, No 1, 1992, pp. 152–155.
[56] FANTI M. P., MAIONE B, TURCHIANO B.: Controllability of multi-input
positive discrete-time systems. Int. J. Control, 1990, vol. 51, No 6, pp. 1295–
1308.
[57] FANTI M. P., MAIONE B, TURCHIANO B.: Controllability of linear single-
input positive discrete-time systems. Int. J. Control, 1989, vol. 50, No 6, pp.
2523–2542.
[58] FEINSTEIN J., BAR-NESS Y.: The solution of the matrix polynomial equation
A(s)X(s)+B(s)Y(s)=C(s). IEEE Trans. Auto. Control, vol. AC-29, No 1, 1984,
pp. 75–77.
[59] FIEDLER M., PTAK V.: Diagonally dominant matrices. Czechoslovak
Mathematical Journal, 92, 1967, pp. 420–433.
[60] FIEDLER M., PTAK V.: On matrices with non-positive off-diagonal elements
and positive principal minors. Czechoslovak Mathematical Journal, 87, 1962,
pp. 382–400.
[61] FIEDLER M., PTAK V.: Some generalizations of positive definiteness and
monotonicity. Numerische Mathematik, 9, 1966, pp. 163–172.
[62] FRANKLIN J.N.: Matrix theory. Prentice-Hall, Englewood Cliffs NY 1968.
[63] GANTMACHER F. R., KLIEIN M. G.: Oscilliacionnyje matricy jadra i malye
kolebanija mechaniczeskich system. Gostizgatielstwo, Moskwa 1950.
[64] GANTMACHER F. R.: Teoria matric. Nauka, Moskwa 1967.
[65] GILLE J. Ch., CLIQUE M.: Rachunek macierzowy i wprowadzenie do analizy
funkcjonalnej. Skrypt Politechniki ĝląskiej, Gliwice 1977.
[66] GOLUB G., VAN LOAN C.: Matrix Computations. John Hopkins Univ. Press,
Baltimore, second edition, 1989.
[67] GÓECKI H., FUKSA S., GRABOWSKI P., KORYTKOWSKI A.: Analysis and
Synthesis of Time delay Systems. PWN and J. Wiley. Warszawa 1989.
[68] GRONE R., JOHNSON C. R., SÁ E. M., WOLKOWICZ H.: Positive definite
completions of partial Hermitian matrices. Linear Algebra and Appl., 58,
1984, pp. 109–124.
[69] HU H.: An algorithm for rescaling a matrix positive definite. Linear Algebra
and its Applications, 96, 1987, pp. 131–147.
[70] JOHNSON C. D.: A unified canonical form for controllable and uncontrollable
linear dynamical system. Int. J. Control 1971, vol. 12, No 3, pp. 497–517.
[71] KACZOREK T.: A new design method of minimal order functional observers
for linear discrete-time systems. 8th IEEE International Conference on
Methods and Models in Automation and Robotics, 2002, 2-5 Sept. 2002,
Szczecin, pp. 375–380.
 References 491

[72] KACZOREK T.: A novel approach to design of minimal order functional

observers for singular linear systems. Bull. Pol. Acad. Techn. Sci., vol. 51,
No 2, 2003, pp. 181–188.
[73] KACZOREK T.: A novel method for computation of realization in singular
systems. XXVI IC-SPETO, International Conference of Fundamentals of
Electronics and Circuit Theory, Niedzica 28-31.05.2003, pp. 189–192.
[74] KACZOREK. T.: A new method for computation and realizations in singular
systems. 14th Int. Conference Pocess Control, June 8-11.2003, Stribske Pleso,
Slovakia, pp. 165-1–165-5.
[75] KACZOREK T.: Algebraic operations on two-dimensional polynomials. Bull.
Acad. Pol. Sci., Ser. Sci. Techn., vol. 30, No 1-2, 1982, pp. 73–83.
[76] KACZOREK T.: Algorithms for finding 2-D polynomial matrices A,B
satisfying the condition AB = BA. Bull. Pol. Acad. Sci., Ser. Sci. Techn., vol.
30, No 9-10, 1982, pp. 509–516.
[77] KACZOREK T.: Algorithm for solving 2-D polynomial matrix equations.
Bull. Acad. Pol. Sci., Ser. Sci. Techn., vol. 31, No 1-2, 1983, pp. 51–57.
[78] KACZOREK T.: An extension of the Cayley-Hamilton theorem for non-square
block matrices and computation of the left and right inverses of matrices,
Bull. Pol. Acad. Techn. Sci., vol. 43, No 1, 1995, pp. 39–46.
[79] KACZOREK T., STAJNIAK A.: An extension of the Cayley-Hamilton theorem
for singular 2-D linear systems with non-square matrices. Bull. Pol. Acad.
Techn. Sci., vol. 43, No 1, 1995, pp. 47–56.
[80] KACZOREK T.: Badanie ekstremalnych wartoĞci mocy czynnej w liniowych
obwodach rozgaáĊzionych. Archiwum Elektrotechniki, 1962, t. XI, z.1, pp.
25–36.
[81] KACZOREK T.: Canonical forms of singular 1D and 2D linear systems. The
2nd Int. Workshop on Multidimensional (nD) Systems (NDS-2000), June 27-
30, 2000 Czocha Castle, Poland, pp. 189–196
[82] KACZOREK T.: Cechy i wáaĞciwoĞci singularnych ukáadów liniowych.
Przegląd Elektrotechniczny, R. LXXII 09’96 , pp. 225–230.
[83] KACZOREK T.: Computation of fundamental matrices and reachability of
positive singular discrete linear systems. Bull. Pol. Acad. Techn. Sci., vol. 46,
No 4, 1998, pp. 501–511.
[84] KACZOREK T., CICHOCKI A., STAJNIAK A.: Computation of the Drazin
inverse of a singular matrix making use of neural networks. Bull. Pol. Acad.
Techn. Sci., vol. 40, Nr 4, 1992, pp. 387–394.
[85] KACZOREK T.: Realization Problem for Positive 2D Systems with Delays.
Mechine Intelligence & Robotic Control, Vol. 6, No 2, 2004, pp. 61–68.
[86] KACZOREK T.: Decomposition of time-varying implicit linear systems into
dynamical and static parts. Bull. Pol. Acad. Techn. Sci. vol.40,No 2, 1992,
pp. 117–123.
[87] KACZOREK T.: Determination of solutions to singular 2-D continuous-
discrete linear systems with singular matrix pencils. Bull. Pol. Acad. Sc.
Techn. Sc., vol. 43, No 2, 1995, pp. 203–225.
[88] KACZOREK T.: Dynamics assignment problem of linear systems.
Konferencja Naukowo-Techniczna „Automatyzacja – NowoĞci i
Perspektywy” AUTOMATION 2001, 28-30.03.2001, Warszawa, pp. 53–58.
492 References

[89] KACZOREK T.: Dynamics assignment problem of linear systems. Bull. Pol.
Acad. Techn. Sci., vol. 49, No 3, 2001, pp. 461–466.
[90] KACZOREK T., Electrical circuits as examples of positive singular
continuous-time systems. SPETO’98, UstroĔ 20-22.05.98, pp. 37–43.
[91] KACZOREK T.: Elementary operations approach to dynamics assignment of
singular linear systems. Int. Workshop on Polynomial Methods in Signal,
Systems and Control, August 28, 2001 Wysokie Tatry.
[92] KACZOREK T.: Elimination of finite eigenvalues of strongly singular systems
by feedbacks in linear systems. Int. Conf. on “Mathematical Modeling as
Means of Power Consumption”, Lwów, 18-23.06.2001, No 421, pp. 73–77.
[93] KACZOREK T.: Elimination of finite eigenvalues of the matrices by
feedbacks linear systems. 10th Int. Conf. “Systems, Modeling Control”,
Zakopane 21-25.05.2001, pp. 345–350.
[94] KACZOREK T.: Elimination of finite eigenvalues of strongly singular the 2D
Roesser model by state-feedbacks. Int. J. Appl. Math. Comput. Sc., 2001,
vol. 11, No 2, pp. 369–376.
[95] KACZOREK T.: Existence and uniqueness of solutions and Cayley-Hamilton
theorem for general singular model of 2-D systems. Bull. Pol. Acad. Techn.
Sci., vol. 37, No 5-6, 1989, pp. 275–278.
[96] KACZOREK T.: Extension of the method of continuants for n-order linear
difference equations with variable coefficients. Bull. Acad. Pol. Sci. Ser., Sci.
Techn., vol. 33, No 7-8, 1985, pp. 395–400.
[97] KACZOREK T.: Extension on Sylvester's theorem to two-dimensional
systems. Bull. Acad. Pol. Sci., Ser. Sci. Techn.,vol. 30, No 1-2, 1982, pp. 109–114.
[98] KACZOREK T.: Extensions of the Cayley-Hamilton theorem for 2-D
continuous-discrete linear systems. Appl. Math. and Com. Sci., vol. 4 No 4,
1994, pp. 507–515.
[99] KACZOREK T.: Floquet-Lyapunov transformation for singular 2-D linear
systems. Bull. Pol. Acad. Sc. Techn. Sc., vol. 40, No 4, 1992, pp. 355–360.
[100] KACZOREK T.: Full-order perfect observer for continuous-time linear
systems. Pomiary, Automatyka, Kontrola, Nr 1, 2001, pp. 3–6.
[101] KACZOREK T.: SàAWIēSKI M., Functional observer for continuous-time
linear systems. Pomiary, Automatyka, Kontrola PAK 9’2003, pp. 4–9.
[102] KACZOREK T.: Generalization of the Cayley-Hamilton theorem for
nonsquare matrices. Prace XVII SPET0-1995, pp. 77–83.
[103] KACZOREK T.: Infinite eigenvalue assignment by output-feedbacks for
singular systems. Inter. J. Appl. Math. And Comp. Sci., vol. 14, No 1, 2004,
pp. 19–23.
[104] KACZOREK T.: Influence of the state-feedback on cyclicity of linear systems.
Materiaáy Konferencji Naukowo-Technicznej “Automation 2002”, 20-22
marca 2002, Warszawa, pp. 81–93.
[105] KACZOREK T.: Linear Control Systems, vol. 1 and 2. Research Studies Press,
J. Wiley, New York 1992-1993.
[106] KACZOREK T.: Methods for computation of solutions to regular discrete-time
linear systems. Appl. Math. and Comp. Science, vol. 5, no 4, 1995, pp. 635–
656.
 References 493

[107] KACZOREK T.: Minimal order perfect functional observers for singular linear
systems. Machine Intelligence & Robotic Control, vol. 4, No 2, 2002, pp. 71–
74.
[108] KACZOREK T.: Minimal order deadbeat functional observers for singular 2D
linear systems. Control & Cybernetics, vol. 32, No 2, 2003, pp. 301–311
[109] KACZOREK T.: Minimal order perfect functional observers for singular linear
systems. IV International Workshop “Computational Problems of Electrical
Engineering”, Zakopane, Sept. 2-5, 2002, pp. 52–55.
[110] KACZOREK T.: Minimalny rząd obserwatorów funkcjonalnych liniowych
ukáadów ciągáych. Pomiary, Automatyka, Kontrola, No 9, 2002, pp. 5–8.
[111] KACZOREK T., CICHOCKI A.: Neural-type structured networks for solving
algebraic Riccati, equations. Archives of Control Sciences, vol. 1, No 3-4,
1992, pp. 153–165.
[112] KACZOREK T.: New algorithms of solving 2-D polynomial equations. Bull.
Acad. Pol. Sci., Ser. Techn. Sci., vol.30, No 5-6, 1982, pp. 263–269.
[113] KACZOREK T.: Obserwatory doskonaáe liniowych ukáadów jedno i dwu-
wymiarowych. XIV Krajowa Konferencja Automatyki, Zielona Góra, 24-27
czerwca 2002, pp. 3–12.
[114] KACZOREK T.: On the solution of linear inhomogeneous matrix difference
equations of order n with variable coefficients. Zastosowanie Matematyki
(Applicationes Mathematicae), vol. 19, No 2, 1987, pp. 285–287.
[115] KACZOREK T.: Perfect functional observers of singular continuous-time
linear systems. Machine Intelligence & Robotic Control, vol. 4, No 1, 2002,
pp. 77–82.
[116] KACZOREK T.: Perfect observers for singular continuous-time linear systems.
Int. Scientific Conf. “Energy Savings Electrical Engineering”, Warsaw, 14-
15 May 2001, pp. 247–251.
[117] KACZOREK T.: Perfect observers for singular 2D Fornasini-Marchesini
models. IEEE Transactions on Automatic Control, vol. 46, No 10, 2001, pp.
1671–1675.
[118] KACZOREK T.: Perfect observers of standard linear systems. Bull. Pol. Acad.
Techn. Sci., vol. 50, No 3, 2002, pp. 237–245.
[119] KACZOREK T., SàAWIēSKI M.: Perfect observers for standard linear
systems. 8th IEEE International Conference on Methods and Models in
Automation and Robotics, 2002, 2-5 Sept. 2002, Szczecin, pp. 399–404.
[120] KACZOREK T.: Perfect observers for singular 2D linear systems. Bull. Pol.
Acad. Techn. Sci., vol. 49, No 1, 2001, pp. 141–147.
[121] KACZOREK T., KRZEMIēSKI S.: Perfect reduced-order unknown-input
observer for descriptor systems. 7th World Multiconference on Systems,
Cybernetics and Informatics, July 27-30,2003, Orlando, Florida, USA.
[122] KACZOREK T.: Perfect reduced-order unknown-input observer for 2D linear
systems. V Int. Workshop Computational Problems of Electrical
Engineering, Jazleevets, Ukraina, 26-29.08.2003, pp. 64–69.
[123] KACZOREK T.: PodzielnoĞü minorów stopnia drugiego macierzy licznika
transmitancji prze jej mianownik. Przegląd Elektrotechniczny, 12, 2001, pp.
297–302.
494 References

[124] KACZOREK T.: Polynomial and rational matrix interpolations for 2-D
systems. Bull. Acad. Pol. Sci. 1994, vol. 42, No2, pp.45-51.
[125] KACZOREK T.: Polynomial approach to pole shifting to infinity in singular
systems by feedbacks. Bull. Pol. Acad. Techn. Sci., vol. 50, No 2 , 2002, pp.
134–144.
[126] KACZOREK T.: Polynomial matrix equations in two indeterminates. Bull.
Acad. Pol. Sci., Ser. Sci. Techn. vol. 30, No 1-2, 1982, pp. 95–100.
[127] KACZOREK T.: Positive 1D and 2D Systems. Springer. Berlin Heidelberg
New York, 2002.
[128] KACZOREK T.: Positive 2D linear systems. Proc. of 13th Int. Conf. Systems
Science, 15-18 Sept. 1998, Wrocáaw, vol. 1, pp. 50–67.
[129] KACZOREK T.: Positive descriptor discrete-time linear systems. International
Journal: Problems of Nonlinear Analysis in Engineering Systems, No 1(7) ,
1998, pp. 38–54.
[130] KACZOREK T.: Positive singular discrete linear systems. Bull. Pol. Acad.
Techn. Sci., vol. 45, No 4, 1997, pp. 619–631.
[131] KACZOREK T.: Proper rational solution to 2-D two-sided polynomial matrix
equation. IEEE Trans. Auto. Control, vol. AC-32, No 9, 1987, pp. 826–828.
[132] KACZOREK T.: Rational and polynomial solutions to 2-D polynomial matrix
equations. Bull. Pol. Acad. Techn. Sci., vol. 39, No 1, 1991, pp. 105–109.
[133] KACZOREK T.: Real solutions to two-sided polynomial matrix equations.
Bull. Pol. Acad. Techn. Sci., vol. 35, No 11-12, 1987, pp. 673–677.
[134] KACZOREK T.: Reachability and controllability of non-negative 2-D Roesser
type models. Bull. Pol. Acad. Techn. Sci., vol. 44, No 1, 1998, pp. 408–410.
[135] KACZOREK T.: Realization problem for singular 2D linear systems. Bull. Pol.
Acad. Techn. Sci., vol. 6, No.3 , 1998, pp. 317–330.
[136] KACZOREK T.: Realization problem for weakly positive linear systems. Bull.
Pol. Acad. Techn. Sci., vol. 48, No 4, 2000, pp. 595–606.
[137] KACZOREK T., KRZEMIēSKI S.: Reduced-order unknown-input perfect
observer for descriptor. 9th IEEE International Conference on Methods and
Models in Automation and Robotics, MiĊdzyzdroje, 25-28.08.2003, pp. 375–
380.
[138] KACZOREK T., DĄBROWSKI W.: Reduction of singular 2-D periodic systems
to constant 2-D linear systems by the use of state feedback and Floquet-
Lyapunov transformation. Bull. Pol. Acad. Techn.Sci., vol. 41, No 2, 1993,
pp. 133–138.
[139] KACZOREK T.: Relationship between infinite eigenvalue assignment for
singular systems and solvability of polynomial matrix equations. 11th
Mediterranean Conference on Control and Automation MED’03, June 18-20,
2003, Rhodes, Greece.
[140] KACZOREK T.: Singular Compartmental Systems and Electrical Circuits.
Elektronoje Modeliowanie, Vol. 26, 2004, pp. 81–98.
[141] KACZOREK T.: Shifting to infinity of the eigenvalues in singular linear
systems. Przegląd Elektrotechniczny, R. LXXIX 6/2003, pp. 439–443.
[142] KACZOREK T.: Sáabo dodatnie ukáady w elektrotechnice. Przegląd
Elektrotechniczny RLXXIV, 11’98, pp.277–281.
 References 495

[143] KACZOREK T.: Sáabo dodatnie singularne ukáady dyskretne. Zeszyty Naukowe
Politechniki ĝląskiej, Seria Automatyka z. 123, 1998, pp. 233–248.
[144] KACZOREK T.: Solving of 2-D polynomial matrix equations. Functional-
Differential Systems and Related Topics, Proc. of III Int. Conference,
BáaĪejewko, May 22-29 1983, pp. 127–132.
[145] KACZOREK T., CICHOCKI A.: Solving of algebraic matrix equations by use
of neural networks. Bull. Acad. Techn. Sci. vol. 40, No 1, 1992. pp. 61–68.
[146] KACZOREK T., CICHOCKI A.: Solving of real matrix equations AX - YB = C
making use of neural networks. Bull. Pol. Acad. Techn. Sci., vol. 40, No 3,
1992, pp. 273–279.
[147] KACZOREK T.: Some Recent Developments in Perfect Observers. Triennial
International Conference on Applied Automatic Syst. Orhid, Macedonia,
Sept. 18-20, 2003, pp. 3–24.
[148] KACZOREK T.: Some recent developments in positive systems. Proc. 7th
Conference of Dynamical Systems Theory and Applications, àódĨ 2003, pp.
25-35.
[149] KACZOREK T.: Special canonical form of matrices A,B,C of multivariable
linear systems. Foundations of Control Engineering, 1979, vol. 4, No 1, pp.
27–34.
[150] KACZOREK T.: Straight line reachability of Roesser model. IEEE Trans. on
Autom. Contr., vol. AC-32, No 7, 1987, pp. 637–639.
[151] KACZOREK T.: Teoria Sterowania i Systemów. PWN, Warszawa 1997.
[152] KACZOREK T.: The relationship between the infinite eigenvalue assignment
for singular systems and the solvability of polynomial matrix equations. Int.
J. Appl. Math. and Comp. Sci., vol. 13, No 2, 2003, pp. 161–167.
[153] KACZOREK T.: Transformation of matrices A,B,C of multivariable linear
time-invariant systems to special canonical form. Bull. Acad. Pol. Sc. 1978,
vol. 26, No 7.
[154] KACZOREK T.: Transformation of time-varying implicit linear systems to
their Weierstrass canonical form. Bull. Pol. Acad. Techn. Sci. vol.40, No 2,
1992, pp. 109–116.
[155] KACZOREK T.: Two-Dimensional Linear Systems. Springer. Berlin
Heidelberg New York 1985.
[156] KACZOREK T.: Two-sided polynomial matrix equations. Bull. Acad. Pol.
Techn. Sci., vol. 35, No 11-12, 1987, pp. 667–671.
[157] KACZOREK T.: Weakly positive continuous-time linear systems. Bull. Pol.
Acad. Techn. Sci., vol. 46, No. 2, 1998, pp. 233–245.
[158] KACZOREK T.: Wektory i macierze w automatyce i elektrotechnice. WNT
Warszawa 1998.
[159] KACZOREK T.: Zero-degree solutions to AX + BY = C and invariant factors
assignment problem. Bull. Acad. Pol. Sci., Ser. Sci. Techn.,vol.34, No 9-10,
1986, pp. 553–558.
[160] KACZOREK T.: Zero-degree solutions to the bilateral polynomial matrix
equations. Bull. Acad. Pol. Sci., Ser. Sci. Techn., vol. 34, No 9-10, 1986, pp.
547–552.
496 References

[161] KACZOREK T., TRACZYK T.: Partial derivative of the Vandermonde

determinant and their application to the synthesis of linear control systems.
Zastosowania Matematyki, 1973, t. XIII, z. 3, pp. 329–337.
[162] KACZOREK. T.: Singular compartmental linear systems. Bull. Pol. Acad.
Science, Vol. 51, No 4, 2003, 409–419.
[163] KACZOREK. T.: Some recent developments in positive systems. 7th Conference on
„Dynamical Systems Theory and Applications”, àódĨ 8-11.12.03. , pp. 23–35.
[164] KACZOREK. T.: Minimal realization for positive multivariable linear systems with
delay. Proc. CAITA 2004, Conference on Advances in Internet Technologies and
Applications, July 8-11, 2004, Purdue, USA, pp. 1–12
[165] KACZOREK. T.: Stability of positive discrete-time systems with time-delay. 8th
World Muliconference on Systems, Cybernetics and Informatics, July 18-21, 2004
Orlando, Florida USA, pp. 321–324
[166] KACZOREK. T.: Structure decomposition and computation of minimal realization
of normal transfer matrix of positive systems. 10th IEEE Int. Conf. on Methods and
Models in Automation and Robotics, MMAR 2004, 30.08-2.09.2004, MiĊdzyzdroje,
pp.93–100
[167] KACZOREK. T.: Realization problem for positive multivariable linear systems with
time-delay. Int. Workshop “Computational Problems of Electrical Enginnering”,
Zakopane, 1-4.09. 2004, pp. 186–192
[168] KACZOREK. T.: Realization problem for positive discrete- time systems with delays.
Systems Science Vol. 30, Nr 4, 2004, pp. 17–30.
[169] KACZOREK. T.: Positive 1D and 2D Systems. Springer. Berlin Heidelberg New
York 2002.
[170] KACZOREK. T.: Realization problem for positive discrete-time systems with delay.
System Science, vol. 30, No. 4, 2004, pp. 117–130.
[171] KACZOREK. T.: Realization problem for positive continuous-time systems with
delays. Proc. Int. Conf. On Systems Engineering. Aug. 16-18, 2005, Las Vegas, USA.
[172] KACZOREK. T.: Positive minimal realizations for singular discrete-time systems
with one delay in state and one delay in control. Bull. Pol. Acad. Sci. Techn., vol 52,
No 3, 2005.
[173] KACZOREK. T.: Minimal realization for positive multivariable linear systems with
delay. Int. J. Appl. Math. Comput. Sci., 2004, vol. 14, No 2, pp. 181–187.
[174] KACZOREK. T., BUSàOWICZ M.: Recent developments in theory of positive
discrete-time linear systems with delays – reachability, minimum energy control and
realization problem. Pomiary, Automatyka, Kontrola, No 9, 2004, pp. 12–15.
[175] KACZOREK. T., BUSàOWICZ M.: Determination of Positive Realization of
Singular Continuous-Time Systems with Delays by State Variable Diagram Method.
Proc. Conf. Automation, 22-24 March, Warsaw, 2006 pp. 352–361.
[176] KACZOREK. T., BUSàOWICZ M.: Determination of realizations od composite
positive systems with delays (Wyznaczanie realizacji záoĪonych ukáadów dodatnich z
opóĨnieniami). XXVIII IC SPETO, MiĊdzynarodowa Konferencja z Podstaw
Elektrotechniki i Teorii Obwodów, UstroĔ 11-14 maja 2005, pp. 349–353
[177] KACZOREK. T., BUSàOWICZ M.: Minimal realization for positive multivariable
linear systems with delay. Int. Journal Applied Mathematics and computer Science,
No 2, vol. 14, 2004, pp. 181–187
[178] KACZOREK. T., BUSàOWICZ M.: Minimal realization for positive multivariable
linear systems with delay. Int. Journal Applied Mathematics and computer Science,
No 2, vol. 14, 2004, pp. 181–187
[179] KAILATH T.: Linear Systems. Prentice Hall, Englewood Cliffs NY, 1980.
 References 497

[180] KARCANIAS N.: Regular state-space realizations of singular system control

problems. Proc. IEEE Conf. Decision and Control, 1987, pp. 1144–1146.
[181] KIEàBASIēSKI A., SCHWETLICK H.: Numeryczna Algebra Liniowa. WNT,
Warszawa 1992.
[182] KLAMKA J.: Controllability of Dynamical Systems. Kluwer Academic Publ,.
Dordecht, 1991.
[183] KLAMKA J.: Function of 2-D matrix. Foundations of Control Engineering,
vol. 9, 1984.
[184] KLAMKA J.: SterowalnoĞü ukáadów dynamicznych, Warszawa-Wrocáaw,
WNT 1990.
[185] KOWALCZYK B.: Macierze i ich zastosowania. WNT, Warszawa 1976.
[186] KUýERA V.: A contribution to matrix equations. Trans. Autom. Control.
1972, vol. AC-17, pp. 344–347.
[187] KUýERA V.: Constant solutions of polynomial equations. Int. J. Control, vol.
53, No 2, 1991, pp. 495–502.
[188] KUýERA V.: Discrete linear control: The polynomial equation approach.
Academia, Prague 1979.
[189] KUREK J. E.: Genericness of solution to N-dimensional polynomial matrix
equation XA = I. IEEE Trans. on Circuits and Systems, vol. 37, No 8, 1990,
pp. 1041–1043.
[190] LAMPE B., ROSENWASSER E.: Algebraische Methoden zur Theorie der
MehrgrĘȕen – Abtastsysteme. Universität Rostock, 2000.
[191] LAMPE B., ROSENWASSER E.: Algebraic properties of irreducible transfer
matrices. Avtomatika i Telemekhanika, N 7, 2000, pp. 31–43 (in Russian)
(English translation: Automation and Remote Control, vol. 61, N 7, Pt. I,
2000, pp. 1091–1102.
[192] LANG S.: Algebra. PWN Warszawa 1984.
[193] LANG S.: Algebra. Reading, Mass, Addison–Wesley 1971.
[194] LANGENHOP C. E.: The Laurent Expansion for a nearly singular matrix.
Linear Algebra and its Applications 4, 1971, pp. 329–340.
[195] LANKASTER P.: Theory of matrices. Academic Press, New York 1969.
[196] LAUB A. J.: A Schur method for solving algebraic Riccati equations. IEEE
Trans. Auto Control, AC-24(6). 1979, pp. 913–921.
[197] LEVIN J.: On the matrix Riccati equation. IEEE Trans. Auto Control, 1967,
vol. AC-12, pp. 746–749.
[198] LEWIS F. L.: A survey of linear singular systems. Circuits Systems Signal
Process, vol. 5, no 1, 1986, pp. 3–36.
[199] LEWIS F. L.: Descriptor systems: Decomposition into forward and backward
subsystems. IEEE Trans. Auto Control, vol. AC-29, 1984, pp. 167–170.
[200] LEWIS F. L., MERTZIOS B. G.: Fundamental matrix of discrete singular
systems, Circuits. Syst., Signal Processing, vol. 8, no 3. 1989, pp. 341–355.
[201] LEWIS F. L.: Fundamental, reachability and observability matrices for
discrete descriptor systems. IEEE Trans. Automat. Contr., vol. AC-30, 1985,
pp. 502–505.
[202] LEWIS F. L.: Further remarks on the Cayley-Hamilton theorem and
Leverrier’s method for the matrix pencil (sE-A). IEEE Transactions on
Automatic Control, vol. AC-31, No. 9, 1986, pp. 869–870.
498 References

[203] LEWIS F. L.: Techniques in 2-D implicit systems. Control and Dynamic
Systems, vol. 69, pp. 89–131.
[204] LUENBERGER D. G.: Canonical forms for linear multivariable systems.
Trans. Autom. Control. June 1967, vol. AC-12, pp. 290–293.
[205] LUENBERGER D. G.: Dynamic equations in descriptor form. IEEE Trans.
Auto Control, vol. AC-22, No 3, 1977, pp. 312–321.
[206] LUENBERGER D. G.: Time-invariant descriptor systems. Automatica, vol. 14,
1978, pp. 473–480.
[207] MacDUFFEE C.C.: The theory of matrices. Chelsea, New York 1946.
[208] MacLANE S., BIRKHOFF G.: Algebra. MacMillan, New York 1967.
[209] MARCUS H., MINC H.: A survey of matrix theory and matrix inequalities.
Allyn and Bacon, Boston 1964.
[210] MINC H.: Nonnegative Matrices. New York, J. Wiley 1988.
[211] MIN-YEN WU: A new concept of eigen-values and eigen-vectors and its
applications. Trans. Autom. Control. 1980, vol. AC-25, No 4, pp. 824–826.
[212] MISRA P., VAN DOOREN P., VARGA A.: Computation of structural invariants
of generalized state-space systems. Automatica, vol. 30, no 12, 1994,
pp. 1921–1936.
[213] MITKOWSKI W.: Stabilizacja systemów dynamicznych. AGH, Kraków 1996.
[214] MOSTOWSKI A., STARK M.: Elementy Algebry WyĪszej. PWN Warszawa
1958.
[215] MUFTI I.H.: On the reduction of a system to canonical (phase-variable) form.
Trans. Autom. Control. April 1965, vol. AC-10, pp. 206–207.
[216] MURTHY D. N. P.: Controllability of a linear positive dynamic system. Int.
J. Systems Sci., 1986, vol. 17, No 1, pp. 49–54.
[217] NEWMAN M.: Integral matrices. Academic Press, New York 1972.
[218] ORTEGA J. M.: Matrix Theory. Plenum Press, New York and London, 1976.
[219] OHTA Y., MADEA H. and KODAMA S.: Reachability, observability and
realizability of continuous-time positive systems. SIAM J. Control and
Optimization, vol. 22, No 2, 1984, pp. 171–180.
[220] PACE I. S., BARNETT S.: Efficient algorithms for linear system calculation.
Smith form and common divisor of polynomial matrices. Int. J. System Sc.
1974, vol. 5, No 6, pp. 403–411.
[221] PENROSE R.: A generalized inverse for matrices. Proc. Cambridge Philos.
Soc. 1955, vol. 51, pp. 406–413.
[222] PERNELLA R.: Le calcul matriciel. Eyrolles Ed. Paris 1969.
[223] PETERSEN I. R., HOLLOT C. V.: A Riccati equation approach to the
stabilization of uncertain linear systems. Automatica, vol. 22, No 4, 1986, pp.
397–411.
[224] POPOV V. M.: Invariant description of linear time-invariant controllable
systems. SIAM J. Control. 1972, vol. 10, No 2, pp. 252–264.
[225] PORTER V. A.: On the matrix Riccati equation. Trans. Autom. Control. 1967,
vol. AC-12, pp. 746–749.
[226] PRONGLE R. M., RAYNER A. A.: Generalized inverse matrices. Griffin,
London 1971.
[227] REID W. T.: A matrix differential equation of Riccati type. Amer. J. Math.
1946, vol. 68, pp. 237–246.
 References 499

[228] ROTH W. E.: The equations AX-BY=C and AX-XB=C in matrices. Proc. Am.
Math. Soc.1952, vol. 3, pp. 392–396.
[229] SILVERMAN L. M.: Transformation of time-variable systems to canonical
(phase-variable) form. Trans. Autom. Control. 1966, vol. AC-11, April, pp.
300–303.
[230] SINHA N. K., ROZSA P.: Some canonical form for linear multivariable
systems. Int. J. Control. 1976, vol. 23, No 6, pp. 865–883.
[231] SMART N. M., BARNETT S.: The Algebra of Matrices in N-Dimensional
Systems. IMA Jour. of Math. Contr. and Inform. 6, 1989, pp. 121–133.
[232] SMART N. M., BARNETT S.: The Cayley-Hamilton Theorem in Two- and N-
Dimensional Systems. IMA Jour. of Math. Contr. and Inform. 2, 1985, pp.
217–223.
[233] SOLAK M. K.: Differential representations of multivariable linear systems
with disturbances and polynomial matrix equations PX+RY=W and
PX+YN=V. IEEE Trans. on Automatic Contr., vol. AC-30, No 7, 1985,
pp. 687–690.
[234] SOMMER R.: Entwurf nichlinear, zeitwarianter Systems durch Polynom. Re-
gelungstechnik 27, 1979, H. 12, pp. 393–399.
[235] ŠEBEK M.: 2-D polynomial equations. Kybernetika, vol. 19, No 3, 1983, pp.
212–224.
[236] ŠEBEK M.: Characteristic polynomial assignment for delay-differential
systems via 2-D polynomial equations. Kybernetika, vol. 23, No 5, 1987, pp.
345–359.
[237] ŠEBEK M., KUýERA V.: Matrix equations arising in regulator problems.
Kybernetika, vol. 17, No 2, 1981, pp. 128–139.
[238] ŠEBEK M.: n-D polynomial matrix equations. IEEE Trans. Auto Control, vol.
33, No 5, 1988, pp. 499–502.
[239] ŠEBEK M.: Two-sided equations and skew primness for n-D polynomial
matrices. System & Control Letters 12, 1989, pp. 331–337.
[240] THEODORU N. J.: M-dimensional Cayley-Hamilton theorem. IEEE Trans.
Auto Control, AC-34, No 5, 1989, pp. 563–565.
[241] TRZASKA Z., MARSZAàEK W.: Inversion of matrix pencils for generalized
systems. Journal of the Frankllin Institute, Pergamon Press Ltd., 1992, pp.
479–490.
[242] TUROWICZ A., MITKOWSKI W.: Teoria macierzy. Wydawnictwa AGH
Kraków 1995.
[243] TWARDY M., KACZOREK T.: Observer-based fault detection in dynamical
systems – part I. Pomiary, Automatyka, Kontrola, No 7/8, 2004, pp. 5–11.
[244] TWARDY M., KACZOREK T. Observer-based fault detection in dynamical
systems – part II. Pomiary, Automatyka, Kontrola, No 7/8, 2004, pp. 11–14.
[245] VAN DOOREN P.: The computation of Kronecker’s canonical form of a
singular pencil. Linear Algebra and Its Applications, 1979, vol. 27, pp. 103–
140.
[246] VARDULAKIS A. I. G.: Proper rational matrix diophantine equations and the
exact model matching problem. IEEE Trans. Auto. Control AC-29, No 5,
1984, pp. 475–477.
500 References

[247] VAUGHAM D. R.: A negative exponential solution for the matrix Riccati
equation. Trans. Autom. Control. 1969, vol. AC-14, pp. 72–79.
[248] VIDYASAGAR M.: On matrix measures and convex Liapunov functions. J.
Math Anl. and Appl., vol. 62, 1978, pp. 90–103.
[249] WARMUS W.: Wektory i macierze. PWN Warszawa 1981.
[250] WARWICK K.: Using the Cayley-Hamilton theorem with N-partitioned
matrices. IEEE Trans. Automat. Contr. AC-28, 1983, pp. 1127–1128.
[251] WILLEMS J. L.: On the existence of a nonpositive solution to the Riccati
equation. IEEE Trans. Aut. Control, AC-19 (5), October 1974.
[252] WOLOVICH W. A.: Linear multivariable systems. Springer. Berlin Heidelberg
New York 1974.
[253] WOLOVICH W. A.: Skew prime polynomial matrices. IEEE Trans. Auto.
Control AC-23, No 5, 1978, pp. 880–887.
[254] WOLOVICH W. A., ANTSAKLIS P.J.: The canonical diophantine equations
with applications. SIAM J. Control and Optimization, vol. 22, No 5, 1984,
pp. 777–787.
[255] XIE G., LONG WANG L, Reachability and controllability of positive linear
discrete-time systems with time-delays. in L. Benvenuti, A. De Santis and L.
Farina (eds): Positive Systems, LNCIS 294, Springer. Berlin Heidelberg New
York 2003, pp. 377–384.
[256] YAKUBOVICH V. A.: Solution of certain matrix inequalities encountered in
nonlinear control theory. Soviet Math. Dokl., vol. 5, 1964, pp. 652–656.
[257] YAKUBOVICH V. A.: The solution of certain matrix inequalities in automatic
control theory. Soviet Math. Dokl., vol. 3, 1962, pp. 620–623. (in Russian).
[258] YIP E. L. and SINOVEC E. F., Solvability, controllability and observability of
continuous descriptor systems. IEEE Trans. Auto Control, vol. AC-26, 1981,
pp. 702–707.
[259] YOKOYAMA R., KINNEN E.: Phase-variable canonical forms for linear,
multi-input, multi-output systems. Int. J. Control. 1973, vol. 17, No 6, pp.
1297–1312.
[260] ZIĉTAK K.: The lp solution of the nonlinear matrix equation XY=A. BIT 23,
1983, pp. 248–257.
[261] ZURMÜHL.: Matrizen. Eine darstellung für Ingenieure. Springer. Berlin
Heidelberg New York 1958.
[262] ĩAK S. H.: On the polynomial matrix equation AX+YB=C. IEEE Trans. Auto.
Control AC-30, No 12, 1985, pp. 1240–1242.
Index

Algebraic matrix equation 358 Cyclicity 264, 267

Asymptotic stability 423
Division on polynomial matrices 9
Canonical form Decomposition
Frobenius 45 Kalman 291
Jordan 45,231 normal matrices 182
McMillan 152 rational function 116
Smith 32 rational matrices 128
Computation of regular pencil 87
cyclic realization 231 singular pencil 95
equivalent standard systems 272 singular systems 299
Frobenius canonical form 45 structural 185, 305
fundamental matrices 276 Weierstrass 91
general solution of polynomial Weierstrass–Kronecker 299
equations 319 Dual system 482
Jordan canonical form 45, 48
minimal deree solution 322 Electrical circuit 200, 286
minimal realisation for singular linear fourth order 210
systems 367 general case 210
normal transfer matrix 244 RC 288
particular solution of polynomial RL 286
equations 313 second-order 200
rational solution 332 third-order 203
similarity transformation matrices 50 Elementary
Computing divisors 37
greatest common divisors 77, 79 operation 20
smallest common multiplication 79 operations method 54
Controllability 475
Cyclic pairs 255 Eigenvector method 57
Cyclic realization 220 Equivalence 42
existence 224 Equivalent standard systems 279
computation 226 Eigenvalues of matrix polynomial 345
502 Index

Fraction description of Normality of matix 164

normal matrices 170
rational matrices 136 Observability 478
Functional observers 391 Output-Feedback 197
Operations on
Generalised Bezoute identity 84, 86 polynomial 5
Generalization of Sylvester equation rational function 107
357 rational matrices 124

Kronecker Perfect observers

indices 102 for standard systems 384
product 340 for systems with unknown inputs 400
full-order 375
Lyapunov equation 361 of singular systems 367
Linear independence 23 reduced-order 375, 378, 408
2D systems 396
Matrices Polynomial 1
column reduced 30 Polynomial operations 5
cyclic 68, 69 Polynomial matrix equations 313, 336
decomposition of regular pencil 87 bilinear with two unknown 325
diagonalisation 60, 62 rational solution 332
Frobenius canonical form 45 unilateral with two variables 313
irreducible transfer 305 Polynomial matrices
Jordan canonical form 45 division 9
left equivalent 27 equivalents 27
normal 163 first degree 42
normalisation 191 greatest common divisors 75
rational normal 168 inverse matrix 132
right equivalent 27 lowest common divisors 75
row reduced 30 pairs 75
simple 68 rank 23
simple structure 60 reduction 32
Matrix relative prime 84
arbitrary square 65 simple 68
equation 313 upon 20
normal inverse 175, 180, 257 zeros 37, 39
normal transfer 260 Problem of realisation 219
pair method 50
variable elements 65 Reachability 264, 435
Minimum energy control 435, 440 Realisation
Normal matrices minimal 220
fraction description 170 cyclic 220
product 175 Realisation problem for
sum 175 positive discrete-time systems 444
Normal systems positive continuous-time systems 453
Cyclic 255 singular multi-variable discrete-time
singular 255 systems with delays 461
 Index 503

Rational System
Reachability 471 continuous-time 282, 422
function 107 discrete-time 419
matrices 107, 124 linear singular 272, 367
Reconstructability 480 positive linear with delays 419
Robust stability 432 singular discrete-time 255, 272
Similarity 42 Synthesis of regulators 155
Space basis 23
Stability of positive linear discrete-time Theorem
systems with delay 423 Bezoute16
Structural stability 244 Cayley–Hamilton 16
Sylvester equation 347 Weierstrass–Kronecker 95