Lecture 1

Signal and System Norms

2.1 Signal Norms

We consider real valued signals1 that are piecewise continuous functions of time
t ∈ [0, ∞). In this section we introduce some different norms for these signals.

Definition 2.1 (Norm on a Vector Space) Let V be a vector space, a given non-
negative function φ : V → R+ is a norm on V if it satisfies

φ(v) ≥ 0, φ(v) = 0 ⇔ v = 0
φ(αv) = |α|φ(v) (2.1)
φ(v + w) ≤ φ(v) + φ(w)

for all α ∈ R and v, w ∈ V.

A norm is defined on a vector space. To apply this concept to the case of signals,
it is necessary to define sets of signals that are vector spaces. This is the case of the
signal spaces described below.

25
26 2 Signal and System Norms

2.1.1 L1 -Space and L1 -Norm

The L1 -space is defined as the set of absolute-value integrable signals, i.e., L1 =

R +∞
{u(t) ∈ R : 0 |u(t)| dt < ∞}. The L1 -norm of a signal u ∈ L1 , denoted kuk1 , is
given by
Z +∞
¯ ¯
kuk1 = ¯u(t)¯ dt (2.2)
0
this norm can be used, for instance, to measure a consumption. In the case of
multidimensional signals u(t) = (u1 (t), . . . , unu (t))T ∈ Ln1 u with ui (t) ∈ L1 i =
1, . . . , nu , the norm is given by
Z nu
+∞ X nu
X
¯ ¯ ° °
kuk1 = ¯ui (t)¯ dt = °ui (t)°
1
(2.3)
0 i=1 i=1

2.1.2 L2 -Space and L2 -Norm

The L2 -space is defined as the set of square integrable signals, i.e., we have L2 =
R +∞
{u(t) ∈ R : 0 u(t)2 dt < ∞}. The L2 -norm of a signal u ∈ L2 , denoted kuk2 , is
given by
µZ +∞ ¶1/2
2
kuk2 = u(t) dt (2.4)
0

the square of this norm represents the total energy contained in the signal. According
to Parseval’s theorem,2 the L2 -norm of a signal u ∈ L2 can be calculated in the
frequency-domain as follows:
µ Z +∞ ¯ ¶1/2
1 ¯U (j ω)¯2 dω
¯
kuk2 = (2.5)
2π −∞

where U (j ω) is the Fourier transform of the signal u(t).

2 Parseval’s theorem states that for a causal signal u ∈ L2 , we have

Z +∞ 1
Z +∞ 1
Z +∞ ¯
2 ∗ ¯U (j ω)¯2 dω
¯
u(t) dt = U (j ω)U (j ω) dω =
0 2π −∞ 2π −∞

where U (j ω) represents the Fourier transform of u(t)

Z +∞
u(t)e−j ωt dt
¡ ¢
U (j ω) = F u(t) =
−∞
2.1 Signal Norms 27

In the case of multidimensional signals u(t) = (u1 (t), . . . , unu (t))T ∈ Ln2 u with
ui (t) ∈ L2 i = 1, . . . , nu , the norm is given by
¶1 ÃZ nu
!1 Ãn !1
µZ +∞ 2 +∞ X 2 Xu 2
2
kuk2 = T
u(t) u(t) dt = ui (t) dt = kui k22
0 0 i=1 i=1
(2.6)

2.1.3 L∞ -Space and L∞ -Norm

The L∞ -space is defined as the set of signals bounded in amplitude, i.e., L∞ =

{u(t) ∈ R : supt≥0 |u(t)| < ∞}. The L∞ -norm of a signal u ∈ L∞ , denoted kuk∞ ,
is given by
¯ ¯
kuk∞ = sup¯u(t)¯ (2.7)
t≥0

this norm represents the maximum value that the signal can take. In the case of
multidimensional signals u(t) ∈ Ln∞u (u(t) = (u1 (t), . . . , unu (t))T with ui (t) ∈ L∞ ),
the norm is given by
³ ¯ ¯´
kuk∞ = max sup ui (t)¯ = max kui k∞
¯ (2.8)
1≤i≤nu t≥0 1≤i≤nu

2.1.4 Extended Lp -Space

The Lp -space, p = 1, 2, ∞, only includes bounded signals. For instance, the L2 -

space only includes signals with bounded energy. In order to also include in our
study unbounded signals as well, it is necessary to introduce extended versions of
the standard Lp -spaces. For this purpose, consider the projection function denoted
PT (.) defined as
(
¡ ¢ u(t), t ≤ T
PT u(t) = uT (t) = (2.9)
0, t >T
where T is a given time interval over which the signal is considered. The extended
Lpe -space, p = 1, 2, ∞, is then defined as the space of piecewise continuous signals
u : R+ → Rm such that uT ∈ Lp .
 (1) One Norm v 1
The one-norm (also known as the L1 -norm, ℓ1 norm, or mean norm) of a vector v is denoted
v 1 and is defined as the sum of the absolute values of its components:
n
X
v 1 = |vi | (1)
i=1

for example, given the vector v = (1, −4, 5), we calculate the one-norm:

(1, −4, 5)1 = |1| + | − 4| + |5| = 10

(2) Two Norm v 2

The two-norm (also known as the L2 -norm, ℓ2 -norm, mean-square norm, or least-squares
norm) of a vector v is denoted v 2 and is defined as the square root of the sum of the
squares of the absolute values of its components:
v
u n
uX
v 2 = t |vi |2 (2)
i=1

for example, given the vector v = (1, −4, 5), we calculate the two-norm:
p √
(1, −4, 5)2 = |1|2 + | − 4|2 + |5|2 = 42

(3) Infinity Norm v ∞

The infinity norm (also known as the L∞ -norm, ℓ∞ -norm, max norm, or uniform norm) of
a vector v is denoted v ∞ and is defined as the maximum of the absolute values of its
components:

v ∞ = max{|vi | : i = 1, 2, . . . , n} (3)

for example, given the vector v = (1, −4, 5), we calculate the infinity-norm:

(1, −4, 5)∞ = max{|1|, | − 4|, |5|} = 5

(4) p-Norm v p

In general, the p-norm (also known as Lp -norm or ℓp -norm) for p ∈ N, 1 ≤ p < ∞ of a vector
v is denoted v p and is defined as:
v
u n
uX
v p = t
p
|vi |p (4)
i=1

There is an important qualitative difference between the 1- and 2-norms:

4
 • Small entries in a vector contribute more to the 1-norm of the vector than to the 2-norm.
For example, if v = (.1, 2, 30), the entry .1 contributes .1 to the 1-norm v1 but contributes
roughly .12 = .01 to the 2-norm v2 .

• Large entries in a vector contribute more to the 2-norm of the vector than to the 1-norm.
In the example v = (.1, 2, 30), the entry 30 contributes only 30 to the 1-norm v1 but
contributes roughly 302 = 900 to the 2-norm v2 .

Thus in our setting, when we want to compare the sizes of error vectors, we should use the 1-norm if
we prefer a few large errors to an accumulation of small errors - if we minimize error in the 1-norm,
most entries in the error vector will be 0, and a few may be relatively large. In the 2-norm, there
will be lots of small errors but few large ones. If we want to avoid outliers and don’t mind having
many small (but nonzero) errors, we should use the 2-norm.

3 Matrix Norms
We will also find it useful to measure the size matrices. Thus, we define the matrix norm as the
“size” of a matrix. We do not measure matrices in the same way we measure vectors, however.
That is, given a matrix (aij ) we do not take the one-norm by adding up the absolute values of
the individual components aij . Instead, we consider the effect the matrix has when multiplied by
vectors.
A matrix A can be seen as a map from vectors to vectors, for example the matrix
· ¸
1 2 3
A=
−4 −5 6

can be seen as a map from R3 to R2 that takes a vector v to another vector Av:
 
v(1) · ¸
v(1) + 2v(2) + 3v(3)
v = v(2) %→ Av =
  .
−4v(1) − 5v(2) + 6v(3)
v(3)

We would like to measure how much a matrix A can amplify a vector v (that is, increase its
norm), but notice that a given matrix A will have different effects on different vectors v , for example:
· 6
10 106
¸ · ¸ · ¸
1 0
6 6 · =
10 10 −1 0
· 6 6 2 · 106
¸ · ¸ · ¸
10 10 1
· =
106 106 1 2 · 106

When we take the norm of a matrix A, we would like to find the maximum expansion it can
cause to any vector v . More formally:

Definition 3.1 (Norm of a Matrix). Given a particular vector norm  , and M × N matrix A,
the norm of A is defined as follows:

A = max {Av  : v  = 1}

5
 Av
Equivalently, A = max{ v : v (= 0}.

Note that this definition does not provide an explicit formula for calculating the norm, but, in some
cases, calculating the matrix norm is not too difficult. To illustrate, consider the following example.
Example 3.1. Determine the one-norm of the matrix:
· ¸
1 −3
A =
2 8

To solve this problem, consider the effect that matrix A has on the simplest unit vectors, (1, 0)T
and (0, 1)T :
· ¸ · ¸ · ¸ °· ¸°
1 −3 1 1 ° 1 °
· = ° 2 ° = |1| + |2| = 3
° °
2 8 0 2 1
· ¸ · ¸ · ¸ °· ¸°
1 −3 0 −3 ° −3 ° = | − 3| + |8| = 11
° °
· =
2 8 1 8 ° 8 °1

It turns out that 11 is the maximum norm of any unit vector multiplied by A, so that A1 = 11.
In general, if one splits a matrix A into its column vectors.

AM ×N = [A1 , A2 , . . . AN ]

then the one-norm of A is the maximum of the one-norms of the column vectors Ai of A.

A1 = max{Ai  : Ai is a column vector of A}

Similarly, in general if one splits a matrix A into its row vectors

 
A1
 A2 
AM ×N =  . 
 
.
 . 
AM

then the ∞-norm of A is the maximum of the one-norms of the row vectors vectors Aj of A:

A∞ = max{Aj  : Aj is a row vector of A}

For example, let

· ¸
1 2 3
A=
−4 −5 6

It is easy to see that, for the purpose of getting the biggest Av∞ , the “best” unit vectors in the
∞-norm are those the form
 
±1
±1
±1

6
where the signs are chosen to avoid cancellation when multiplying by a row of A. In the case of
our matrix A above, the biggest we can get is
°  °
°· ¸ 1 ° °· ¸°
° 1 2 3  ° ° 6 °
° −4 −5 6 1 ° = ° 3 ° = 6 maximizing the first entry of the output vector
° ° ° °
° 1 ° ∞
°  °∞
°· ¸ −1 ° °· ¸°
° 1 2 3 ° ° 0 °
° −4 −5 6 −1 ° = ° 15 ° = 15 maximizing the second entry of the output vector
°  ° ° °
° 1 °
∞
∞

So A∞ is 15.

Algorithms for finding the two-norms (and higher p-norms) of a matrix are more complex, and
we do not discuss them here. We can get lower bounds on the two norm by taking any vector with
2-norm equal to 1, e.g.,
 
1
0 ,
0
multiplying our matrix by this vector, e.g.,
 
· ¸ 1 · ¸
1 2 3   1
0 = ,
−4 −5 6 −4
0
and finally computing the two norm of the output
√
°· ¸°
° 1 °
° −4 ° = 17.
° °
2
√
So A2 ≥ 17.

7
 2.2 LTI Systems 29

2.2 LTI Systems

Broadly speaking, a system can be seen as a device that associates to a given in-
put signal u(t ), an output signal y(t ). In this book, for tractability reasons, we
consider the particular class of linear time invariant finite-dimensional systems or
LTI-system

for short. The so-called state-space representation of this kind of system is defined
as follows:

ẋ(t) = Ax(t) + Bu(t)

(2.14)
y(t) = Cx(t) + Du(t)

where u ∈ Rnu is the input vector, y ∈ Rny is the output vector, x ∈ Rnx is the state
vector, and A, B, C, D are constant matrices of appropriate dimension. It can be
established that the solution of the state equation in (2.14), for a given initial state
vector x(t0 ), is as follows:

Z t
A(t−t0 )
x(t) = e x(t0 ) + eA(t−τ ) Bu(τ ) dτ (2.15)
t0

Note that this solution is the superposition of two terms, the first term eA(t−t0 ) x(t0 )
represents the state
R t evolution of the autonomous system, i.e., for u = 0, whereas
the second term t0 e A(t−τ ) Bu(τ ) dτ represents the state evolution of the system for
zero initial condition. This last term is written as the convolution product of the
quantity eAt B, called the input-to-state impulse matrix,5 by the input u(t). From
(2.14) and (2.15) we can see that the response y(t) of the system to a given input
vector u(t) is then given by

Z t
A(t−t0 )
y(t) = Ce x(t0 ) + CeA(t−τ ) Bu(τ ) dτ + Du(t) (2.16)
t0

An important question is to determine in which conditions the state remains bounded

(and therefore the output as well) when the system is driven by a bounded input
signal. This question is closely related to the ability of the autonomous system to
recover its equilibrium point6 starting from any initial state. This is the problem of
stability, which is briefly considered in the next section.
30 2 Signal and System Norms

2.2.1 System Stability

A fundamental property of any system is its stability. Stability is the ability of an

autonomous system7 to recover its equilibrium point after being disturbed from it.
More formally, the system described by (2.14) is stable if for every initial condition
x(t0 ) the following limit:

lim x(t) = 0 (2.17)

t→∞

holds when u = 0. From (2.15) we can see that the state vector solution of the au-
tonomous system is given by x(t) = eA(t−t0 ) x(t0 ). Therefore, the limit (2.17) holds
if and only if the matrix A, also called state matrix, has all its eigenvalues in the
open left-half plane C− . The eigenvalues of the matrix A ∈ Rnx ×nx are the nx roots
of the polynomial characteristic defined by

det(λI − A) = λnx + anx −1 λnx −1 + · · · + a1 s + a0 (2.18)

If the nx roots of the polynomial characteristic (2.18) are all in the open left-half
plane, then the matrix A is said to be Hurwitz. The set of n-by-n Hurwitz matrices
is defined as

H = H ∈ Rn×n : λi (H ) ∈ C− , i = 1, . . . , n
© ª
(2.19)

where λi (H ) represents the ith eigenvalue of H . Therefore, the autonomous system

ẋ(t) = Ax(t) is stable if and only if A ∈ Rnx ×nx is Hurwitz, i.e., A ∈ H. At this
point, it is important to note that the set of Hurwitz matrices is not convex.
• Non-convexity of the Set of Hurwitz Matrices. Given two Hurwitz matrices
A1 , A2 ∈ H, the convex combination A(α) = αA1 + (1 − α)A2 , α ∈ [0, 1] does
not necessarily belong to H for any α. To observe this, consider for instance the
matrices
· ¸ · ¸
a 2b a 0
A1 = , A2 =
0 b 2a b
with a, b < 0, the convex combination of A1 and A2 is given by
· ¸
a 2αb
A(α) =
2(1 − α)a b

It can be easily seen that A1 , A2 ∈ H whereas A( 21 ) ∈

/ H.
2.2 LTI Systems 31

Fig. 2.1 Example of a state

trajectory of a stable system.
From any initial state x(t0 ),
the state trajectory converges
to the equilibrium point of the
autonomous system i.e., the
origin of the state space

Lyapunov Method Another way to establish the stability of a given LTI au-
tonomous system
ẋ(t) = Ax(t) (2.20)
is the Lyapunov method. Consider a quantity related to the distance of the current
state vector x(t) to the origin of the state space8 e.g., its squared quadratic norm:
V (x(t)) = kxk2P = x(t)T P x(t), where P is a symmetric positive definite matrix.9
Under these conditions, it is clear that the limit (2.17) holds if and only if the dis-
tance of x(t) to the origin decreases as time increases (see Fig. 2.1). Therefore, we
can conclude that the system is stable if and only if there is a matrix P = P T ≻ 0
such that V (x(t)) = x T P x is a strictly decreasing function of time, i.e., V̇ (x(t)) < 0
for all x 6= 0. The time derivative of V is given by

V̇ x(t) = ẋ(t)T P x(t) + x(t)T P ẋ(t)

¡ ¢

= x(t)T AT P x(t) + x(t)T P Ax(t)

= x(t)T AT P + P A x(t)
¡ ¢
(2.21)

The quadratic form x(t)T (AT P + P A)x(t) is negative definite for all x 6= 0 if and
only if the symmetric matrix AT P + P A is negative definite, i.e., all its eigenvalues
32 2 Signal and System Norms

are negative, which is denoted by

AT P + P A ≺ 0 (2.22)

This expression is called a Lyapunov inequality on P , which is also a linear matrix

inequality (LMI). This LMI can be solved by taking any matrix Q = QT ≻ 0 and
by solving the following linear equation, also called the Lyapunov equation:

AT P + P A = −Q (2.23)

of unknown P . Thus, if the autonomous system (2.20) is stable then the matrix P
solution of the Lyapunov equation is definite positive.
The system stability issue can then be summarized as follows.
• System Stability Condition. The LTI system (2.14) is said to be stable if and
only if the state matrix A has all its eigenvalues in the open left-half plane C− ,
i.e., the eigenvalues of A have a negative real part. In this case, the state matrix
A is said to be a Hurwitz matrix.
Equivalently, the LTI system (2.14) is said to be stable if and only if there
exists a positive definite symmetric matrix P satisfying the Lyapunov inequality
AT P + P A ≺ 0.

Remark 2.1 The stability result given above is often referred to as the internal
stability. The notion of internal stability must be distinguished from the so-called
BIBO-stability. The LTI-system (2.14) is said to be BIBO-stable if a bounded input
produces a bounded output. From relation (2.16) it is clear that an internally stable
system is also BIBO-stable, the converse is false in general. This is because between
the input and output there can be unstable hidden modes, i.e., some unbounded state
variables which are not influenced by the inputs or have no influence to the outputs.
Therefore since these unstable modes are not input/output visible, the system can
be input/output stable but not internally stable. In this book, the notion of stability
always refers to internal stability.

2.2.2 Controllability, Observability

In Remark 2.1 we have introduced the notion of hidden modes. To illustrate this
notion, consider the LTI system (2.14) with
   
λ1 0 0 b11 b12
A = 0 λ2 0 ,
  B= 0
 0 
0 0 λ3 b31 b32
· ¸
c11 c12 0
C= , D=0
c21 c22 0
2.2 LTI Systems 33

Relation (2.15) makes it possible to calculate the evolution of the state vector for a
given initial state x(0) = (x1 (0), x2 (0), x3 (0)), and for a given input vector u(t) =
(u1 (t), u2 (t)), we have
Z t Z t
λ1 t λ1 (t−τ )
x1 (t) = e x1 (0) + b11 e u1 (τ ) dτ + b12 eλ1 (t−τ ) u2 (τ ) dτ
0 0

x2 (t) = eλ2 t x2 (0)

Z t Z t
λ3 t λ3 (t−τ )
x3 (t) = e x3 (0) + b31 e u1 (τ ) dτ + b32 eλ3 (t−τ ) u2 (τ ) dτ
0 0

we can see that the input vector u(t) has no influence on the evolution of the state
variable x2 . In this case we say that λ2 is an uncontrollable mode. The evolution of
the output vector is given by y(t) = Cx(t), we have

y1 (t) = c11 x1 (t) + c12 x2 (t), y2 (t) = c21 x1 (t) + c22 x2 (t)

we can see that the state variable x3 has no influence to the output vector y(t). In this
case we say that λ3 is a unobservable mode. Consider now the input/output relation
calculated for zero initial conditions; we have
µ Z t Z t ¶
λ1 (t−τ ) λ1 (t−τ )
y1 (t) = c11 b11 e u1 (τ ) dτ + b12 e u2 (τ ) dτ
0 0
µ Z t Z t ¶
λ1 (t−τ ) λ1 (t−τ )
y2 (t) = c21 b11 e u1 (τ ) dτ + b12 e u2 (τ ) dτ
0 0

we can see that the input/output relation, evaluated for zero initial conditions, only
involves modes that are both controllable and observable, in this example λ1 . Note
also that in the case where λ1 ∈ C− and λ2 , λ3 ∈ C+ , the system is BIBO-stable but
internally unstable.
The example given above suggests the following definitions about the notions of
controllability and observability of an LTI system.

Definition 2.2 (Controllability) An LTI system is controllable if every mode of A

is connected to the input vector u.

Definition 2.3 (Observability) An LTI system is observable if every mode of A is

connected to the output vector y.

The following results can be used to test the controllability and the observabil-
ity of a given LTI system. The notions of stabilizability and detectability are also
specified.
• Controllability, Stabilizability. The LTI system (2.14) is controllable if and only
if the controllability matrix

C = B AB A2 B · · · Anx −1 B
£ ¤
(2.24)
34 2 Signal and System Norms

is of full rank, i.e., rank(C) = nx . In this case, the pair (A, B) is said to be con-
trollable.
In the case where rank(C) = n < nx , the rank defect nx − n represents the
number of uncontrollable modes. The uncontrollable modes are the eigenvalues
λ of the state matrix A satisfying rank([λI − A B]) < nx . The LTI system (2.14)
is said to be stabilizable if and only if all uncontrollable modes are stable.
• Observability, Detectability. The LTI system (2.14) is observable if and only if
the observability matrix
 
C
 CA 
 CA2 
 
O=  (2.25)
 .. 
 . 
CAnx −1
is of full rank, i.e., rank(O) = nx . In this case, the pair (A, C) is said to be
observable.
In the case where rank(O) = n < nx , the rank defect nx − n represents the
number of unobservable modes. The unobservable modes are the eigenvalues λ
of the state matrix A satisfying rank λIC−A < nx . The LTI system (2.14) is said
£ ¤

to be detectable if and only if all unobservable modes are stable.

Physical Meaning of the Controllability and Observability Controllability is

related to the ability of a system to attain a given state under the action of an ap-
propriate control signal. If a state is not controllable, then it not possible to move
this state to another one by acting on the control input. If the dynamics of a non-
controllable state is stable, then the state is said to be stabilizable.
Observability is linked to the possibility of evaluating the state of a system
through output measurements. If a state is not observable there is no way to de-
termine its evolution. If the dynamics of a non-observable state is stable, then the
state is said to be detectable.

2.2.3 Transfer Matrix

The state space representation is often referred as an internal representation be-

cause it involves the state variables which are internal variables of the system. The
input/output representation, also called external representation, is obtained by elim-
inating the Laplace transform10 of the state vector, between the state equation and
the output equation for zero initial conditions. Taking the Laplace transform of the

R∞
10 The Laplace transform of a given signal u(t) is defined as U (s) = L(u(t)) = 0 x(t)e−st dt .
From this definition, it is easy to show that the Laplace transform of the derivative of a signal is
given by L(u̇(t)) = s L(u(t)) − u(0).
2.3 System Norms 35

Fig. 2.2 Block diagram of a closed-loop system, G(s) is the transfer matrix of the system to be
controlled, and K(s) is the controller which must be designed to obtain a low tracking error and
a control signal compatible with the possibility of the plant despite the external influences r, d,
and n

state equation in (2.14), we get X(s) = (sI − A)−1 BU (s). By substituting X(s) in
the output equation, we obtain the input/output relation

Y (s) = G(s)U (s), G(s) = C(sI − A)−1 B + D (2.26)

where G(s) is called the transfer matrix of the system. This transfer matrix repre-
sents the Laplace transform of the input to output impulse matrix. The elements of
the matrix G(s) are real rational transfer functions (i.e., ratios of polynomials in s
with real coefficients). A transfer matrix G(s) is proper if G(∞) = D, and strictly
proper if G(∞) = 0. We have seen that the input/output representation only involves
the eigenvalues that are both controllable and observable. These are called the poles
of the system. A proper transfer matrix G(s) is stable if the poles lie in the open
left-half plane C− . The set of proper and stable transfer matrices of size ny × nu
n ×n
is denoted RH∞y u . The set of strictly proper and stable transfer matrices of size
n ×n
ny × nu is denoted RH2 y u . It can be easily shown that these sets are convex. This
is in contrast to the non-convexity of the set of Hurwitz matrices (see Sect. 2.2.1).
36 2 Signal and System Norms

and u:
 
· ¸ r
e
= T (s) d 
u
n
Good performance is then obtained if the transfer matrix T (s) is small or, more
specifically, if the gain of T (s) is small. The word “gain” must be understood here
as a measurement of the size of the matrix T (s).
The gain of a system quantifies the amplification provided by the system between
the inputs and the outputs. This notion of gain needs to be defined more accurately,
this is the subject of the next section on H2 and H∞ norms of a system.

2.3.1 Definition of the H2 -Norm and H∞ -Norm of a System

Let G(s) be the transfer function of a stable single input single output (SISO) LTI-
system of input u(t) and output y(t). We know that G(s) is the Laplace transform
of the impulse response g(t) of the system, we define the H2 -norm of G(s) as the
L2 -norm of its impulse response:
µZ ∞ ¶1/2
2
kGk2 = g(t) dt = kgk2 (2.27)
0

Note that the previous norm is defined for a particular signal which is here the
Dirac impulse δ(t). According to Parseval’s theorem the H2 norm is defined in the
frequency domain as follows:
µ Z +∞ ¯ ¶1/2
1 ¯G(j ω)¯2 dω
¯
kGk2 = (2.28)
2π −∞

Remark 2.2 It is interesting to give an interpretation of the H2 -norm of a system. To

this end, recall that if Su (ω) is the power spectral density (PSD) of the signal applied
to the input of a stable system of transfer function G(s), the PSD of the signal output
Sy (ω) is given by Sy (ω) = |G(j ω)|2 Su (ω). Now, assume that the input u(t) is a
white noise signal, i.e. Su (ω) = 1 for all ω, in this case, the PSD of the signal output
is nothing but the square of the frequency gain of the system: Sy (ω) = |G(j ω)|2 .
Using (2.11) the RMS-value of the signal output is given by
µ Z +∞ ¶1/2 µ Z +∞ ¯ ¶1/2
1 1 ¯G(j ω)¯2 dω
¯
yrms = Sy (ω) dω = (2.29)
2π −∞ 2π −∞

which coincides with the definition of the H2 -norm of the system (see relation
(2.28)). In other words, the H2 -norm of a system represents the RMS-value of the
system response to a white noise input.
2.3 System Norms 37

We can define the gain provided by the system for a given particular input as
the ratio of the L2 -norm of the output signal to the L2 -norm of the input signal
kGkgain = kGuk2 /kuk2 , with kuk2 6= 0. For obvious reason, this gain is often re-
ferred to as the L2 -gain of the system. Instead of evaluating the L2 -gain for a par-
ticular input, one can also determine the greatest possible L2 -gain over the set of
square integrable signals, this is the definition of the H∞ -norm of a system

kGuk2
kGk∞ = sup (2.30)
u∈L2 kuk2
kuk2 6=0

This quantity represents the largest possible L2 -gain provided by the system. For a
MIMO system with nu inputs and ny outputs, the H∞ -norm is defined as

kGuk2 n
kGk∞ = sup with y ∈ L2 y (2.31)
u∈Ln2 u kuk2
kuk2 6=0

2.3.2 Singular Values of a Transfer Matrix

The usual notion of the frequency gain of a SISO system can be extended to the
MIMO case by considering the singular values of the transfer matrix G(s) of the
system. Let y(t) be the system response to a causal input u(t). In the frequency
domain, this response is written

Y (j ω) = G(j ω)U (j ω) (2.32)

where Y (j ω) = (Y (s))s=j ω , U (j ω) = (U (s))s=j ω , Y (s) = L(y(t)), U (s) =

L(u(t)), and the notation L(.) stands for the Laplace transform of the signal passed
in argument. In the SISO case, the gain of the system at frequency ω is given by
|G(j ω)|. This notion of frequency gain can be extended to the MIMO case by using
the singular values, denoted σi , of the matrix G(j ω) = (G(s))s=j ω . The singu-
lar values of the matrix G(j ω) are defined as the square roots of eigenvalues of
G(j ω)G(−j ω)T
¡ ¢ q ¡ ¢ q ¡ ¢
T
σi G(j ω) = λi G(j ω)G(−j ω) = λi G(−j ω)T G(j ω) (2.33)

with i = 1, . . . , min(nu , ny ). The matrix G(−j ω)T represents the conjugate trans-
pose of G(j ω), and is usually denoted G(j ω)∗ i.e., G(j ω)∗ = G(−j ω)T . The
matrices G(j ω)G(j ω)∗ and G(j ω)∗ G(j ω) are Hermitian11 positive semi-definite,
their eigenvalues are therefore non-negative.
38 2 Signal and System Norms

Fig. 2.3 Singular values and H∞ -norm of a transfer matrix. The frequency gain of the MIMO
system lies between the smallest and the largest singular values. The maximum over ω of the
largest singular value represents the H∞ -norm of the LTI system

We denote σ̄ (G(j ω)) the largest singular value of G and σ (G(j ω)) the smallest
¡ ¢ ¡ ¢ ¡ ¢ ¡ ¢
σ̄ G(j ω) = σ1 G(j ω) ≥ σ2 G(j ω) ≥ · · · ≥ σ G(j ω) ≥ 0 ∀ω (2.34)

we then have12
¡ ¢ ° ° ° ° ¡ ¢
σ G(j ω) ≤ °G(j ω)U (j ω)°2 /°U (j ω)°2 ≤ σ̄ G(j ω) (2.35)

This means that the frequency gain of the system lies between the smallest and the
largest singular values. Therefore, the singular values can be used to extend to the
MIMO case the usual notion of gain. The singular values are positive functions of
ω and can be represented in the frequency domain as shown Fig. 2.3.
In the case of a SISO system, G(s) is scalar, it is then easy to see that we have
only one singular value which is equal to the modulus of G(j ω)
¡ ¢ ¯ ¯
σ G(j ω) = ¯G(j ω)¯ (2.36)

It is worth noting that any complex matrix M ∈ Cny ×nu has a singular value
decomposition, see the Notes and References.

12 Indeed, it can be shown that for a complex matrix A ∈ Cp×m and a complex vector x ∈ Cm , we
have
kAxk2 kAxk2
σ̄ (A) = maxm and σ (A) = minm
x∈C kxk2 x∈C kxk2
kxk2 6=0 kxk2 6=0

To observe this, consider the first-order optimality condition of λ = kAxk22 /kxk22 =

(x ∗ A∗ Ax)/(x ∗ x). We have
∂λ ¡ ∗ ¢
= A A − λI x = 0
∂x
thus, λ represents the eigenvalues of the matrix A∗ A. Therefore, since λ = kAxk22 /kxk22 , the
∗
maximum p of kAxk2 /kxk2 is given by the square root of the largest eigenvalue of A A i.e.,
σ̄ (A) = λ̄(A∗ A), and the minimumpof kAxk2 /kxk2 is given by the square root of the small-
est eigenvalue of A∗ A i.e., σ (A) = λ(A∗ A). Note that the input vector for which the gain is
maximal (respectively, minimal) is given by the eigenvector associated to the largest (respectively,
smallest) eigenvalue of A∗ A.
2.3 System Norms 39

2.3.3 Singular Values and H2 , H∞ -Norms

Let G(s) be a stable and strictly proper transfer matrix13 of dimension p × m. The
n ×n
set of stable and strictly proper transfer matrices is denoted RH2 y u . For any trans-
n ×n
fer matrix G(s) ∈ RH2 y u , we define the H2 -norm as14
µ Z +∞ ¶1/2
1
Trace G(j ω)G∗ (j ω) dω
° ° ¡ ¢
°G(s)° = (2.37)
2 2π −∞

this norm can be also expressed using the singular values:15

!1/2
+∞ min(m,p)
à Z
1 X
σi2 G(j ω) dω
° ° ¡ ¢
°G(s)° = (2.38)
2 2π −∞ i=1

The square of the H2 -norm represents the area under the curve of the sum of squared
singular values.
Now, consider a stable and proper transfer matrix G(s). The set of stable and
n ×n n ×n
proper transfer matrices is noted RH∞y u . For any transfer matrix G(s) ∈ RH∞y u
the H∞ -norm is defined as
° ° ¡ ¢
°G(s)° = sup σ̄ G(j ω) (2.39)
∞
ω

This norm represents the largest possible frequency gain, which corresponds to the
maximum of the largest singular value of G(j ω) (see relation (2.35) and Fig. 2.3).
In the case of a SISO system, kG(s)k∞ is the maximum of |G(j ω)|
¯ ¯
kGk∞ = max¯G(j ω)¯ (2.40)
ω
Introduction to Multivariable Control
Example
 
5 4
G1 =
3 2
The singular value decomposition of G1 is
     H
0.872 0.490 7.343 0 0.794 −0.608
G1 =
0.490 −0.872 0 0.272 0.608 0.794
| {z } | {z } | {z }
U Σ VH

The largest gain of 7.343 is for an input in the direction

0.794
v̄ = , the smallest gain of 0.272 is for an input in the
0.608  
−0.608
direction v = .
0.794
 Elements of Linear System Theory
Example
1
G(s) =
s+a
H2 norm:
Z ∞
1 1
kG(s)k2 = ( |G(jω)|2 dω) 2
2π −∞ | {z }
1
ω 2 +a2
r
1 h −1 ω i∞ 1 1
=( tan ( ) )2 =
2πa a −∞ 2a

Alternatively: Consider the impulse response

 
1
g(t) = L−1 = e−at , t ≥ 0
s+a

Elements of Linear System Theory

to get
sZ r
∞
1
kg(t)k2 = (e−at )2 dt =
0 2a
as expected from Parseval’s theorem.
H∞ norm:
1 1
(3.103) ||G(s)||∞ = max |G(jω)| = max 1 =
ω ω (ω 2 + a2 ) 2 a
 Lecture 2
Stability
Bounded-Input Bounded-Output (BIBO) Stability
Asymptotic Stability
Lyapunov Stability

Bounded-Input Bounded-Output (BIBO) stablility

Definition: For any constant N, M >0

Any bounded input yields bounded output, i.e.

u (t ) N y (t ) M

For linear systems: p( s)

T ( s) C ( sI A) 1 B
q( s )
BIBO Stability All the poles of the transfer function lie in the LHP.

q(s ) 0 Solve for poles of the transfer function T(s)

Characteristic Equation

2
 Asymptotic stablility
When u(t ) 0, i. e. the system x Ax
x (t ) 0 as t

For linear systems: x Ax Bu

y Cx

Asymptotically stable All the eigenvalues of the A matrix

have negative real parts
(i.e. in the LHP)

p( s ) C adj[ sI A]B
T ( s) C ( sI A) 1 B
q( s ) sI A

sI A 0 Solve for the eigenvalues for A matrix

Note: Asy. Stability is indepedent of B and C Matrix

3

Asy. Stability from Model Decomposition

Suppose that all the eigenvalues of A are distinct. A Rn n

Let vi the eigenvector of matrix A with respect to eigenvalue i

i.e. i , satisfying Avi v , i 1,

i i ,n
Coordinate Matrix T [v1, v2 , ,vn ]

T 1 AT A T 1 AT
B T 1B
1 1 0 0 0 1

0 0 0 C CT
2 2 2

0 z T 1 ATz T 1 Bu
n
0 0 n n
y CTz Du

1
x (t ) T (t ) v1e 1t 1 (0) v2 e 2t 2 (0) vn e nt n (0), (0) T x (0)
Hence, system Asy. Stable all the eigenvales of A at lie in the LHP

4
 Instability in 1D

Stability Concept in 1D

Asymptotic Stability in 1D

Asymptotic Stablility versus BIBO Stability

In the absence of pole-zero cancellations, transfer function poles are
identical to the system eigenvalues. Hence BIBO stability is
equivalent to asymptotical stability.
Conclusion: If the system is both controllable and observable, then
BIBO Stability Asymptotical Stability

Methods for Testing Stability

Asymptotically stable
All the eigenvalues of A lie in the LHP
BIBO stable
Routh-Hurwitz criterion
Root locus method
Nyquist criterion
....etc.

6
Lyapunov Stablility

A state xe of an autonomous system is called an equilibrium state,

if starting at that state the system will not move
from it in the absence of the forcing input.
In other words, consider the system x f ( x (t ), u(t ))
equilibrium state xe must satisfy f ( xe ,0) 0, t t0
Example: Find the equilibrium point ?
x2
Equilibrium point
0 1 1
x x u (t )
2 3 1 x1

Definition: An equilibrium state xe of an autonomous system is

stable in the sense of Lyapunov if for every 0, exist a ( ) 0
such that x0 xe x (t , x0 ) xe for t t0

x2

x (t ) x1

xe
x0

8
Definition: An equilibrium state xe of an autonomous system is
asymptotically stable if
(i) it is stable
(ii) there exist a e 0 such that
x0 xe e x ( t ) xe 0, as t

x2

e
x1
xe
x0

x(t )
9

Lyapunov Theorem
Consider the system x f (x ) (1)

Eq. State : xe 0 f (0) 0

A function V(x) is called a Lapunov fuction V(x) if
(1) V ( x ) 0, x 0
(2) V (0) 0 for x 0
dV ( x ) dV ( x )
(3) f ( x) 0
dt dx
Then eq. state of the system (1) is stable.
Moreover, if the Lyapunov function satisfies
dV ( x) dV ( x )
0, x 0 and 0 x 0
dt dt
Then eq. state of the system (1) is asy. stable.

10
Explanation of the Lyapunov Stability Theorem

1. The derivative of the Lyapunov function along the trajectory is negative.

2. The Lyapunov function may be consider as an energy function of the system.

x2

V ( x (t ))

x1
0 x (t )

11

Lyapunov s method for Linear system: x Ax where A 0

The eq. state x 0 is asymptotically stable.

For any p.d. matrix Q , there exists a p.d. solution of the

Lyapunov equation T
A P PA Q

Proof: Choose V ( x) x T Px

V ( x) x T Px x T Px
x T AT Px x T PAx
x T ( AT P PA) x AT P PA Q
x T Qx 0, for x 0
Hence, the eq. state x=0 is asy. stable by Lapunov theorem.

12
Asymptotically stable in the large
( globally asymptotically stable)
(1) The system is asymptotically stable for all the initial states x(t0 ) .
(2) The system has only one equilibrium state.
(3) For an LTI system, asymptotically stable and globally
asymptotically stable are equivalent.
Lyapunov Theorem (Asy. Stability in the large)
If the Lyapunov function V(x) further satisfies

(i) x ,V ( x)
(ii) x , V ( x)

Then, the (asy.) stability is global.

13

A symmetric n n matrix Q is p.d. if and only if all its n

leading principle minors are positive.

Definition
The i-th leading principle minor Qi i 1,2,3, , n of an n n
matrix Q is the determinant of the i i matrix extracted from
the upper left-hand corner of Q.

q11 q12 q13

Q q21 q22 q23 Q1 q11
q31 q32 q33
q11 q21
Q2 Q3 Q
q21 q22

14
Remark:
(1) Q1 , Q2 , Qn are all negative Q is n.d.
(2) All leading principle minors of Q are positive Q is n.d.

Example:

Q1 2 0
Q2 6 0
Q3 24 0

Q is not p.d.

15

0 1
Example: Test the stability of the system x x
1 1

p11 p12
Let Q I , Assume P
p12 p22
Solve for AT P PA I
0 1 p11 p12 p11 p12 0 1 1 0
1 1 p12 p22 p12 p22 1 1 0 1
p11 p12 1 3 1
P
p12 p22 2 1 2

p11 3 0 P 5 0 P is p.d.
System is asymptotically stable

16
Q.1 Check the stability of the equilibrium state of the system
described by

Solution:
Select Lyapunov function
+
Which is positive definite function
Its derivative is

17

is always negative definite.

as Then system is asymptotically stable in-the-large.

18
Q.2 Consider the nonlinear system described by the equations

Find the region in the state plane for which the equilibrium state of
the systems asymptotically stable.
Solution:
Select Lyapunov function
+
Which is positive definite function
Its derivative is

19

Case1:
to be negative definite if
or
If these conditions are fulfil then system remain asymptotically
stable.
Case2:
= 0 if at this point rate of change of energy will
become zero. At this point there is no trajectory. But energy is not
zero at this point.

20
Case3:
Vo(x) is not negative definite if >1
In this region system will become unstable.

X2
Possibly unstable
Possibly unstable

X1

Stable Stable

21
 Lecture 3 : Sensitivity of Control Systems

!"#$%& '()*$""&+',-.#

!"# $%&'()*'(&'( +#,-($*%."$& +#&+#.#%(. ("# *-/"$0-.10$22$*# +#,-($*%."$& */ ("#

&+*0#..1 2"$0" $% ('+% +#&+#.#%(. - &+*0#..$%3 */ ("# $%&'( .$3%-, (* &+*4$5# -%
*'(&'( .$3%-, 4-+$-6,#77

#8"9 $8"9
!8"9
3.!4/ 54/!4/
%&'()""*+%,-./*+01"/)2
 3).#(),+ !"#$%"
; 0*%(+*, .<.(#= $. 5#/$%#5 -. -% $%(#+0*%%#0($*% */ 0*=&*%#%(. /*+=$%3 -
.<.(#= 0*%/$3'+-($*% ("-( 2$,, &+*4$5# - 5#.$+#5 .<.(#= +#.&*%.#7
>( -0"$#4#. ("# 5#.$+#5 +#.&*%.# 6< (2* /*+= */ .<.(#=. .(+'0('+#.?
4 56$. 7))6 3).#(), !"#$%
#8"9 88"9 $8"9
08"9 9 !8"9
;)"<&)= :(/4-,
&)"!'.") 7'./&',,)& :(/4-/'& %&'()""*+%,-./*+01"/)2 &)"!'.")

!4 3,)"$1 7))6 89$$1:-*;4 *).#(), "!"#$%

#8"9 68"9 88"9 $8"9
08"9 9 !8"9
;)"<&)= ;0('-(*+ :(/4-,
7'./&',,)& %&'()"" &)"!'.")
&)"!'.")
>8"9
?)-"4&)2)./+'4/!4/ @))=A-(B
0)."'&
@#0-'.# ("# 5#.$+#5 .<.(#= +#.&*%.#1 #8"91 $. A%*2%1 - .$3%-, &+*&*+($*%-, (*
("# #++*+1 68"91 6#(2##% ("# 5#.$+#51 #8"91 -%5 ("# -0('-, +#.&*%.#1 $8"91 $.
3#%#+-(#57 6 8 "9 ! # 8 "9 $ 8 " 9

56$.+7))6+ !"#$%"

56$. ,))6 "!"#$%< ;% *&#%),**& 0*%(+*, .<.(#= '($,$B#. -% -0('-($%3 5#4$0# (*

0*%(+*, ("# &+*0#.. 5$+#0(,< 2$("*'( '.$%3 /##56-0A7

#8"9 88"9 $8"9

08"9 9 !8"9
;)"<&)= :(/4-,
&)"!'.") 7'./&',,)& ;0('-(*+ %&'()""*+%,-./*+01"/)2 &)"!'.")

#8"9 88"9 $8"9

D*+C9E 08"9 !8"9

7'./&',,)& %&'()""*+%,-./*+01"/)2

!"#C.<.(#=C(+-%./#+C/'%0($*%C-%5C*'(&'(C-+#?
$ 8 "9
! C 8 "9 ! ( 8 "9 ! 8 "9
#8 " 9
$ 8 "9 ! ( 8 " 9 ! 8 " 9 # 8 "9

:
 9$$1:-*;+83,)"$1+7))64+ !"#$%"
9$$1:-*; 83,)"$1 ,))64 "!"#$%< !"# '.# */ #++*+ 86#(2##% ("# 5#.$+#5 -%5 ("#
-0('-, +#.&*%.#9 .$3%-, (* 0*%(+*, ("# &+*0#.. +#.',(. $% - 0,*.#5),**& .#G'#%0# */
*&#+-($*%. ("-( $. 0-,,#5 - /##56-0A .<.(#=7 !"$. 0,*.#5 ,**& .#G'#%0# */
*&#+-($*%. $. ."*2% $% D$3'+#7
#8"9 68"9 88"9 $8"9
08"9 9 !8"9
;)"<&)= ;0('-(*+ :(/4-,
7'./&',,)& %&'()"" &)"!'.")
&)"!'.")
>8"9
?)-"4&)2)./+'4/!4/ @))=A-(B
0)."'&

!"# $%(+*5'0($*% */ /##56-0A (* $=&+*4# ("# 0*%(+*, .<.(#= $. */(#% %#0#..-+<7 >( $.
$%(#+#.($%3 ("-( ("$. $. -,.* ("# 0-.# /*+ .<.(#=. $% %-('+#1 .'0" -. 6$*,*3$0-, -%5
&"<.$*,*3$0-, .<.(#=.? /##56-0A $. $%"#+#%( $% ("#.# .<.(#=.7 D*+ #H-=&,#1 ("#
"'=-% "#-+( +-(# 0*%(+*, .<.(#= $. - /##56-0A 0*%(+*, .<.(#=7 ;% *&#%),**&
0*%(+*, .<.(#= '($,$B#. -% -0('-($%3 5#4$0# (* 0*%(+*, ("# &+*0#.. 5$+#0(,< 2$("*'(
'.$%3 /##56-0A7

9$$1:-*;+83,)"$1+7))64+ !"#$%"
9$$1:-*; "!"#$%< ; 0,*.#5),**& .<.(#= '.#. - =#-.'+#=#%( */ ("# *'(&'( .$3%-,
-%5 - 0*=&-+$.*% 2$(" ("# 5#.$+#5 *'(&'( (* 3#%#+-(# -% #++*+ .$3%-, ("-( $. '.#5 6<
("# 0*%(+*,,#+ (* -5I'.( ("# -0('-(*+7

#8"9 68"9 88"9 $8"9

08"9 9E !8"9
;)"<&)= ;0('-(*+ :(/4-,
7'./&',,)& %&'()"" &)"!'.")
&)"!'.")
>8"9
?)-"4&)2)./+'4/!4/ @))=A-(B
0)."'&

!"#C.<.(#=C(+-%./#+C/'%0($*%C$.? $ 8 "9 ( 8 "9 ! 8" 9

! C 8 "9 !
#8 "9 " ( 8 "9 ! 8 "9 > 8 "9

!"#C.<.(#=C*'(&'(C$.? ( 8"9 ! 8"9

$ 8 "9 ! #8 " 9
" ( 8 "9 ! 8 "9 > 8 "9

F
 9$$1:-*;+83,)"$1+7))64+ !"#$%"

#8"9 68.9 88"9 $8"9

08"9 !8"9

7'./&',,)& %&'()""*+%,-./*+01"/)2
>8"9E
0)."'&
>/ ("# .#%.*+ =#-.'+#. ("# *'(&'( -00'+-(#,<1 >8"9E 1 ("#%?
$ 8 "9 ( 8 "9 ! 8 "9
!"#C.<.(#=C(+-%./#+C/'%0($*%C$.? ! C 8 "9 !
#8 "9 " ( 8 "9 ! 8 "9
( 8"9 ! 8 "9
!"#C.<.(#=C*'(&'(C$.? $ 8 "9 ! #8 " 9
" ( 8 "9 ! 8 "9

$/C 08"9 !8"9CKKC /*+C-,,C0*=&,#HC/+#G'#%0$#.C*/C$%(#+#.(1C("#CC.<.(#=C*'(&'(C$.

$ 8 " 9 # #8 " 9

9$$1:-*;+83,)"$1+7))64+ !"#$%"

#8"9 68"9 88"9 $8"9

08"9 !8"9
;)"<&)= :(/4-,
7'./&',,)& %&'()"" &)"!'.")
&)"!'.")
>8"9E
?)-"4&)2)./+'4/!4/ 0)."'& @))=A-(B

D*+ 08"9 !8"9 KK ("# .<.(#= *'(&'( $. $8"9 # #8"9

!"# *'(&'( $. -&&+*H$=-(#,< #G'-, (* ("# $%&'(7

L*2#4#+1 ("# 0*%5$($*% 08"9 !8"9 KK =-< 0-'.# ("# .<.(#= +#.&*%.# (* 6#
"$3",< *.0$,,-(*+< -%5 #4#% '%.(-6,#7
@'( ("# /-0( ("-( $%0+#-.$%3 ("# =-3%$('5# */ ("# ,**& 3-$% D8"9E 08"9 !8"9
+#5'0#. ("# #//#0( */ !8"9 *% ("# *'(&'( $. -% #H0##5$%3,< '.#/', +#.',(7
!"#+#/*+#1 *%# -54-%(-3# */ - /##56-0A .<.(#= $. ("-( ("# #//#0( */ ("#
4-+$-($*% */ ("# &-+-=#(#+. */ ("# &+*0#..1 !8"91 $. +#5'0#57

J
 3).#(),,=.>+?.*$(#-=.+)(+?.%)1$,,$1 !"#$%"
N*%(+*, .<.(#= #%3$%##+. -+# 0*%0#+%#5 2$(" '%5#+.(-%5$%3 -%5 0*%(+*,,$%3
.#3=#%(. */ ("#$+ #%4$+*%=#%(1 */(#% 0-,,#5 .<.(#=.1 (* &+*4$5# '.#/', #0*%*=$0
&+*5'0(. /*+ .*0$#(<7 !"# (2$% 3*-,. */ /.1$("#-.1=.> -%5 *).#(),,=.> -+#
0*=&,#=#%(-+< 6#0-'.# #//#0($4# .<.(#=. 0*%(+*, +#G'$+#. ("-( ("# .<.(#=. 6#
/.1$("#))1 -%5 %)1$,$17 D'+("#+=*+#1 0*%(+*, #%3$%##+$%3 ='.( */(#% 0*%.$5#+
("# 0*%(+*, */ 6))(,! /.1$("#))1 .<.(#=. .'0" -. 0"#=$0-, &+*0#.. .<.(#=.7

#8"9 $8"9
!8"9C$ % !8"9
3.!4/ 54/!4/
O#.$+#5C.<.(#=C+#.&*%.# %&'()""*+%,-./*+01"/)2 ;0('-,C.<.(#=C+#.&*%.#

P"#+#C !8"9C$.C("#C=*5#,C*/C("#C.<.(#=C-%5C$ % !8"9C'%0#+(-$%(<C*+C'%=*5#,,#5

&-+(C*/C("#C.<.(#=7

$."=#=@=#!+)2+3).#(),+ !"#$%"+#)+'-(-%$#$(+A-(=-#=)."
!"# .#%.$($4$(< */ - 0*%(+*, .<.(#= (* &-+-=#(#+ 4-+$-($*%. $. */ &+$=# $=&*+(-%0#
.$%0# $( $. 5#.$+-6,# (* =$%$=$B# ("# #//#0(. */ &-+-=#(#+ 4-+$-($*%. -%5
'%0#+(-$%($#.7
; &+*0#..1 +#&+#.#%(#5 6< ("# (+-%./#+ /'%0($*% &8"9* 2"-(#4#+ $(. %-('+#1 $.
.'6I#0( (*
- 0"-%3$%3 #%4$+*%=#%(1
-3$%31
$3%*+-%0# */ ("# #H-0( 4-,'#. */ ("# &+*0#.. &-+-=#(#+.1 -%5
*("#+ %-('+-, /-0(*+. ("-( -//#0( - 0*%(+*, &+*0#..7
!"#.# 0-% 6# +#&+#.#%(#5 6< % !8"9 2"#+# Q !8"9QKKQ% !8"9Q

#8"9 $8"9
!8"9C$ % !8"9
3.!4/ 54/!4/
O#.$+#5C.<.(#=C+#.&*%.# %&'()""*+%,-./*+01"/)2 ;0('-,C.<.(#=C+#.&*%.#

M
 $."=#=@=#!+)2+56$.+7))6+ !"#$%"+#)+'-(-%$#$(+A-(=-#=)."

>/ ("# .<.(#= 0"-%3#5 6< % !8"9 2"#+# Q !8"9QKKQ% !8"9Q ("#% ("# .<.(#= 6#0*=#.
!8"9$% !8"9?

#8"9 88"9 $8"9

08"9 !8"9$% !8"9

7'./&',,)& %,-./*+01"/)2

D*+C("#C*&#%C,**&C.<.(#=C("#C.<.(#=C(+-%./#+C/'%0($*%C-%5C*'(&'(C-+#?
$ 8 "9
! C 8"9 ! ( 8 "9T ! 8"9 $ % ! 8 " 9S
#8 " 9
$ 8 " 9 $ %$ 8 " 9 ! ( 8 "9T ! 8"9 $ % ! 8 "9S# 8 "9
!"# .<.(#= &-+-=#(#+ 0"-%3#.1 $% !8"91 5$+#0(,< -&&#-+. $% ("# .<.(#= *'(&'( -%5
$.
%$ 8 "9 ! $ ( 8 "9% ! 8 "9 # 8 "9

$."=#=@=#!+)2+56$.+7))6+ !"#$%"+#)+'-(-%$#$(+A-(=-#=)."
#8"9 88"9 $8"9
08"9 !8"9$% !8"9

7'./&',,)& %&'()""*+%,-./*+01"/)2
$ 8 "9 $ %$ 8" 9 ! ( 8" 9T ! 8"9 $ % ! 8" 9S#8 "9
N*%.$5#+ ("# /*,,*2$%3 *&#% ,**& 0*%(+*, .<.(#= 2$(" ("# 5$.('+6-%0# ;8"9 -%5
2$("*'( $% !8"97
U%&+#5$0(#5C5$.('+6-%0#
$%&'(1C;8"9
#8"9 88"9 $8"9
08"9 !8"9
7'./&',,)& %&'()""*+%,-./*+01"/)2
!"#C*'(&'(C$.? $# 8 "9 " $; 8 "9 ! 8"9 8 "9 #8 " 9 " 8 " 9 ;8 " 9
( ! !

!"# 5$.('+6-%0#1 ;8"91 5$+#0(,< $%/,'#%0#. ("# *'(&'(1 $8"97

!"# *&#% ,**& 0*%(+*, .<.(#= $. "$3",< .#%.$($4# (* 5$.('+6-%0#.1 ;8"91 -%5 (*
0"-%3#. $% !8"9 $% &-+-=#(#+. */ !8"9

R
 $."=#=@=#!+)2+9$$1:-*;+ !"#$%"+#)+'-(-%$#$(+A-(=-#=)."
W'&&*.# ("# .<.(#= 8*+ &,-%(9 &8"9 '%5#+3*#. - 0"-%3# .'0" ("-( ("# (+'# &,-%(
=*5#, $. !8"9$% !8"9 2"#+# Q !8"9QKKQ% !8"9Q7 !"# 0"-%3# $% ("# &,-%( =-< 6#
5'# (* - 0"-%3$%3 #H(#+%-, #%4$+*%=#%( *+ %-('+-, -3$%31 *+ $( =-< I'.( +#&+#.#%(
("# '%0#+(-$%(< $% 0#+(-$% &,-%( &-+-=#(#+.7E

#8"9 68"9 88"9 $8"9

08"9 !8"9$% !8"9
;)"<&)= :(/4-,
7'./&',,)& %&'()"" &)"!'.")
&)"!'.")
>8"9E
?)-"4&)2)./+'4/!4/ @))=A-(B
0)."'&
$ 8" 9 ( 8 " 9T ! 8 " 9 $ % ! 8 "9S
!"#C.<.(#=C(+-%./#+C/'%0($*%C$.? ! C 8"9 !
# 8" 9 " ( 8" 9T ! 8" 9 $ % ! 8 "9S
!"#C.<.(#=C*'(&'(C$.?

( 8" 9 ! 8" 9 ( 8" 9% ! 8" 9

$ 8" 9 $ %$ 8 " 9 ! #8" 9 " #8 " 9
" ( 8 " 9T ! 8"9 $ % ! 8 "9S " ( 8 " 9T ! 8" 9 $ % ! 8 " 9S

$."=#=@=#!+)2+9$$1:-*;+3).#(),+ !"#$%"+#)+'-(-%$#$(+A-(=-#=)."
#8"9 68"9 88"9 $8"9
08"9 !8"9$% !8"9
;)"<&)= 7'./&',,)& :(/4-,
%,-./*+01"/)2
&)"!'.") &)"!'.")
>8"9E
?)-"4&)2)./+'4/!4/ @))=A-(B
0)."'&
( 8"9 ! 8"9 ( 8 "9 % ! 8 "9
$ 8" 9 $ %$ 8 " 9 ! # 8 "9 $ #8 "9
" ( 8 "9T ! 8" 9 $ % ! 8 " 9S " ( 8 "9T ! 8 "9 $ % ! 8 " 9S

D*+ Q !8"9QKKQ% !8"9Q -%5 08"F &8"9GG ("#% ("# 0"-%3# -&&#-+. $% *'(&'( $.

( 8 " 9% ! 8 "9
%$ 8 " 9 ! $ #8 " 9
" ( 8 "9T ! 8" 9 " % ! 8" 9S
#X
;3-$% $/ 08"F &8"9GG /*+ -,, 0*=&,#H /+#G'#%0$#. */ $%(#+#.( ("# .<.(#= *'(&'(
$.
$ 8 " 9 & #8 "9

V
 $."=#=@=#!+)2+9$$1:-*;+3).#(),+ !"#$%"+#)+'-(-%$#$(+A-(=-#=)."

Z#( .('5< ("# #//#0( *% ("# (+-0A$%3 #++*+ 68"9 5'# (* % !8"97

#8"9 68"9 88"9 $8"9

08"9 !8"9$% !8"9
;)"<&)= :(/4-,
7'./&',,)& %,-./*+01"/)2 &)"!'.")
&)"!'.")
>8"9E
?)-"4&)2)./+'4/!4/ @))=A-(B
0)."'&

6 8 " 9 $ %6 8 " 9 ! #8 " 9

" ( 8 " 9T ! 8"9 $ % ! 8 "9S

!"#% ("# 0"-%3# $% ("# (+-0A$%3 #++*+ $.

( 8 " 9% ! 8 "9
%6 8 " 9 ! # 8 "9
T " ( 8 "9 ! 8 "9 $ ( 8 "9 % ! 8 " 9ST " ( 8"9 ! 8 " 9S

$."=#=@=#!+)2+9$$1:-*;+3).#(),+ !"#$%"+#)+'-(-%$#$(+A-(=-#=)."
#8"9 68.9 88"9 $8"9
08"9 !8"9$% !8"9

7'./&',,)& %&'()""*+%,-./*+01"/)2

( 8" 9% ! 8 "9
%6 8 " 9 ! #8" 9
T " ( 8 "9 ! 8 "9 $ ( 8 "9% ! 8 "9ST " ( 8" 9 ! 8 "9S

W$%0# 2# '.'-,,< /$%5 ("-( (8"9 !8"9 GG (8"9% !8"9 2# "-4#

( 8 " 9% ! 8" 9 ( 8 " 9% ! 8"9

%6 8 " 9 & :
#8 " 9 *+ %6 8 " 9 & #8"9
T " ( 8"9 ! 8 " 9S T " D8 " 9S:

2"#+# D8 " 9 ! ( 8"9 ! 8"9 $.C("#C,**&C(+-%./#+C/'%0($*%7

!"'. ("# (+-0A$%3 #++*+ $. +#5'0#5 6< ("# /-0(*+ [ D8"91 2"$0" $. 3#%#+-,,<
3+#-(#+ ("-% *4#+ ("# +-%3# */ /+#G'#%0$#. */ $%(#+#.(7

Y
 $."=#=@=#!+)2+9$$1:-*;+3).#(),+ !"#$%"+#)+'-(-%$#$(+A-(=-#=)."

( 8 " 9% !8"9 #8"9 68.9 88"9 $8"9

%6 8 " 9 & :
#8"9 08"9 !8"9$% !8"9
T " D8 " 9S
>8"9
2"#+# D8 " 9 ! ( 8"9 ! 8"9

D*+ ,-+3# D8"91 ("#% [ D8"9 & D8"91 -%5 ("# -&&+*H$=-(# ("# 0"-%3# $% ("# (+-0A$%3
#++*+ $.
% ! 8"9
%6 8 " 9 & # 8" 9
D 8 "9 ! 8"9

Z-+3#+ =-3%$('5# D8"9 (+-%.,-(#. $%(* .=-,,#+ 0"-%3#. $% ("# (+-0A$%3 #++*+ 8("-( $.1
+#5'0#5 .#%.$($4$(< (* 0"-%3#. $% % !8"9 $% ("# &+*0#..97
;,.*1 ,-+3#+ D8"9 $=&,$#. .=-,,#+ .#%.$($4$(<1 08"97
!"# G'#.($*% -+$.#.1
B)C 1) C$ 1$2=.$ "$."=#=@=#! "8#4 2)( #D$ 2$$1:-*; "!"#$%E

W$%0# *'+ 3*-, $. (* +#5'0# .<.(#= .#%.$($4$(<1 $( =-A#. .#%.# (* /*+=-,,< 5#/$%#
("# .#%.$($4$(< (#+=7

$."=#=@=#!+)2+3).#(),+ !"#$%"+#)+'-(-%$#$(+A-(=-#=)."
!"#$% "$."=#=@=#!& "8#4& $. ("# +-($* */ ("# 0"-%3# $% $%&'()*'(&'( .<.(#= (+-%./#+
/'%0($*%1 C8"91 (* ("# 0"-%3# */ - &+*0#.. (+-%./#+ /'%0($*%1 &8"9 8*+ &-+-=#(#+9 /*+
- .=-,, $%0+#=#%(-, 0"-%3#7

%C 8 " 9 ] C 8 " 9
0 C 8 "9 !
% ! 8"9 ] ! 8"9
>%C("#C,$=$(1C/*+C.=-,,C$%0+#=#%(-,C0"-%3#.1C$(C6#0*=#.?

'C 8 " 9 ] C 8 " 9 'C 8 " 9 ! 8 " 9

0 C 8 "9 ! !
' ! 8 "9 ] ! 8"9 ' ! 8"9 C 8 "9
' ,% C 8 "9
!
' ,% ! 8 " 9

\
 $."=#=@=#!+)2+56$.+7))6+ !"#$%"
N*%.$5#+ -3-$% ("# /*,,*2$%3 *&#%),**& 0*%(+*, .<.(#=
#8"9 88"9 $8"9
08"9 !8"9

7'./&',,)& %&'()""*+%,-./*+01"/)2

!"#C.<.(#=C(+-%./#+C/'%0($*%C$.? C 8 " 9 ! 8"9 8" 9

( !

!"#C.#%.$($4$(<C*/C("#C*&#%),**&C.<.(#=C(*C0"-%3#.C$%C("#C&,-%(C &8"9C$.C

'C 8 " 9 ! 8 " 9 ! 8" 9

0C ! ! ( 8 "9 !
' ! 8 "9 C 8 "9 ( 8" 9 ! 8" 9

!"# .#%.$($4$(< */ ("# *&#%),**& .<.(#= (* 0"-%3#. $% ("# &,-%( &8"9 $. #G'-, (* 7
;%< 0"-%3# $% ("# &,-%(1 &8"91 *+ ("# 5$.('+6-%0#1 ;8"91 5$+#0(,< $%/,'#%0#. ("#
*'(&'(1 $8"97

$."=#=@=#!+)2+9$$1:-*;+ !"#$%"
!"# .#%.$($4$(< */ - 0*%(+*, .<.(#= (* &-+-=#(#+ 4-+$-($*%. $. */ &+$=# $=&*+(-%0#7
; &+$=-+< -54-%(-3# */ - /##56-0A /##56-0A 0*%(+*, .<.(#= $. $(. -6$,$(< (* +#5'0#
("# .<.(#=^. .#%.$($4$(<
N*%.$5#+ -3-$% ("# /*,,*2$%3 /##56-0A 80,*.#5),**&9 0*%(+*, .<.(#=

#8"9 68.9 88"9 $8"9

08"9 !8"9$% !8"9

>8"9

( 8 "9 ! 8 "9
!"#C.<.(#=C(+-%./#+C/'%0($*%C$.? C 8 " 9 !
" ( 8" 9 ! 8 " 9 > 8 "9
!"#C.#%.$($4$(<C*/C/##56-0ACC.<.(#=C(*C0"-%3#.C$%C("#C&,-%(C &8"9C$.C
'C 8 " 9 ! 8 " 9
0 C 8 "9 ! !
' ! 8 " 9 C 8 "9 " ( 8 "9 ! 8 "9 > 8 " 9

X
 $."=#=@=#!+)2+9$$1:-*;+ !"#$%"

#8"9 68"9 88"9 $8"9

08"9 !8"9$% !8"9
;)"<&)= :(/4-,
7'./&',,)& %,-./*+01"/)2 &)"!'.")
&)"!'.")
?)-"4&)2)./+'4/!4/ >8"9E
@))=A-(B
0)."'&
'C 8 " 9 ! 8 " 9
0C ! !
' ! 8 "9 C 8 "9 " ( 8 "9 ! 8 "9 > 8" 9

P#C/$%5C("-(C("#C.#%.$($4$(<C*/C("#C.<.(#=C=-<C6#C+#5'0#5C6#,*2C("-(C*/C("#C*&#%C
,**&C.<.(#=C6<C$%0+#-.$%3CD8"9CEC 08"9 &8"9C*4#+C("#C/+#G'#%0<C+-%3#C*/C$%(#+#.(7

F(()(+ =>.-,+G.-,!"="+)2+9$$1:-*;+ !"#$%"

_H(#+%-,1CU%&+#5$0(#5C
5$.('+6-%0#C$%&'(1C;8"9C
#8"9 68"9 88"9 $8"9
08"9 !8"9
;)"<&)= :(/4-,
7'./&',,)& %,-./*+01"/)2 '4/!4/
<.!4/
?)-"4&)2)./+'4/!4/ >8"9E
0)."'& >%(#+%-,C5$.('+6-%0#C%*$.#C
W#%.*+C%*$.#1CI8"9

!"#C0,*.#5),**&C/##56-0AC0*%(+*,C.<.(#=C"-.C("+##C$%&'(.?
;)"<&)=+<.!4/+#8"9
6H/)&.-,+-.=+4.!&)=<(/)=+=<"/4&A-.()+<.!4/+;8"9CC
3./)&.-,+=<"/4&A-.()+.'<")+I8"9
-%5C*%#C*'(&'(1C$8"97C

O#/$%#C("#C#(-*;=.>+$(()(+-. 6 8"9 ! #8 "9 $ 8 "9

 F(()(+ =>.-,+G.-,!"="+)2+9$$1:-*;+ !"#$%"
U%&+#5$0(#5C5$.('+6-%0#C$%&'(1C;8"9C
#8"9 68"9 88"9 $8"9
08"9 !8"9
;)"<&)= :(/4-,
7'./&',,)& %,-./*+01"/)2 '4/!4/
<.!4/
>8"9
?)-"4&)2)./+'4/!4/
0)."'& W#%.*+C%*$.#1CI8"9
!"#C)/#6/#+$.
( 8 "9 ! 8 "9 ! 8 "9 ( 8 "9 ! 8 "9 > 8 "9
$ 8 "9 ! #8 "9 " ;8" 9 I 8 "9
" ( 8 "9 ! 8" 9 > 8 "9 " ( 8 "9 ! 8" 9 > 8 "9 " ( 8" 9 ! 8 "9 > 8" 9
( 8"9 ! 8 "9 ! 8 "9 D8 "9
! #8 "9 " ; 8 "9 I 8 "9
" D8 "9 " D 8"9 " D 8" 9
!"#C("#C#(-*;=.>+$(()( $.
! 8 "9 ( 8"9 ! 8"9 > 8"9
6 8" 9 ! # 8"9 ; 8"9 I 8"9
" ( 8" 9 ! 8" 9 > 8"9 " ( 8"9 ! 8"9 > 8"9 " ( 8" 9 ! 8" 9 > 8"9
! 8" 9 D 8"9
! #8 "9 ; 8 "9 I 8"9
" D8 "9 " D8" 9 " D 8"9 2"#+#CD8"9CEC 08"9 &8"9>8"9

F(()(+ =>.-,+G.-,!"="+)2+9$$1:-*;+ !"#$%"

;8"9C
#8"9 68"9 88"9 $8"9
D8"9CEC 08"9 &8"9>8"9 08"9 !8"9
;)"<&)= 7'./&',,)& :(/4-,
<.!4/ %,-./*+01"/)2 '4/!4/
!"#C"$."=#=@=#!+
>8"9
2/.*#=).+$. 0 8 " 9 !
" D8" 9 0)."'& I8"9
!"#C*)%6,$%$.#-(! D8 "9
"$."=#=@=#!+2/.*#=).+$. 7 8 " 9 ! " D8 " 9
>%C(#+=.C*/C("#C/'%0($*%.C08"9C-%5C78"9 ("#C(+-0A$%3C#++*+C$.
6 8"9 ! 0 8"9 #8"9 0 8"9 ! 8 " 9 ; 8 " 9 7 8 "9 I 8 "9
D*+ - 3$4#% &8"91 $/ 2# 2-%( (* =$%$=$B# ("# (+-0A$%3 #++*+1 68"9 2# 2-%( 6*("
08"9 -%5 78"9 (* 6# .=-,,7 `*($0# ("-( 08"9 -%5 78"9 -+# 6*(" /'%0($*%. */ ("#
0*%(+*,,#+1 08"91 2"$0" ("# 0*%(+*, 5#.$3% #%3$%##+ ='.( .#,#0(7 L*2#4#+1 ("#
/*,,*2$%3 .&#0$-, +#,-($*%."$& 6#(2##% 08"9 -%5 78"9 "*,5.7
0 8 "9 " 7 8"9 !

:
 F(()(+ =>.-,+G.-,!"="+)2+9$$1:-*;+ !"#$%"
;8"9C
#8"9 68"9 88"9 $8"9
D8"9CEC 08"9 &8"9>8"9 08"9 !8"9
;)"<&)= 7'./&',,)& :(/4-,
<.!4/ %,-./*+01"/)2 '4/!4/
>8"9
0 8"9 ! D8 "9
7 8 "9 !
" D8" 9 " D8 "9 0)."'& I8"9

!"#C(+-0A$%3C#++*+C$. 6 8 " 9 ! 0 8 " 9 # 8 " 9 0 8 "9 ! 8 " 9 ;8 "9 7 8 "9 I 8 "9

D*+ - 3$4#% &8"91 (* =$%$=$B# ("# (+-0A$%3 #++*+1 68"91 6*(" 08"9 -%5 78"9 (*
6# .=-,,7 W$%0# 08"9 -%5 78"9 -+# 6*(" /'%0($*%. */ ("# 0*%(+*,,#+1 08"91 -%5
("# /*,,*2$%3 .&#0$-, +#,-($*%."$& 6#(2##% 08"9 -%5 78"9 "*,5.7
0 8 "9 " 7 8 "9 !
P# 0-%%*( .$=',(-%#*'.,< =-A# 08"9 -%5 78"9 .=-,,7 a64$*'.,<1 5#.$3%
0*=&+*=$.#. ='.( 6# =-5#7

F(()(+ =>.-,+G.-,!"="+)2+9$$1:-*;+ !"#$%"

D8"9CEC 08"9 &8"9>8"9 ;8"9C
#8"9 68"9 88"9 $8"9
08"9 !8"9
0 8"9 ! ;)"<&)= :(/4-,
" D8" 9 7'./&',,)& %,-./*+01"/)2
<.!4/ '4/!4/
D 8" 9 >8"9
7 8 "9 !
" D8 "9 0)."'& I8"9
6 8 " 9 ! 0 8 "9 # 8 " 9 0 8 "9 ! 8 " 9 ;8 "9 7 8 "9 I 8 "9
0 8"9 " 7 8"9 !
!* -%-,<B# ("# (+-0A$%3 #++*+ #G'-($*%1 2# %##5 (* '%5#+.(-%5 2"-( $( =#-%. /*+ -
(+-%./#+ /'%0($*% (* 6# b,-+3#b *+ (* 6# b.=-,,7b !"# 5$.0'..$*% */ =-3%$('5# */ -
(+-%./#+ /'%0($*% $. ("# .'6I#0( */ /+#G'#%0< +#.&*%.# =#("*5.1 Z#( 5#.0+$6# ("#
=-3%$('5# */ ("# ,**& 3-$% D8"9 6< 0*%.$5#+$%3 ("# =-3%$('5# QD8J(9Q *4#+ ("#
+-%3# */ /+#G'#%0$#.1 (1 */ $%(#+#.(7

F
 H="#/(:-.*$+($I$*#=).+=.+#D$+9$$1:-*;+3).#(),+ !"#$%"
H="#/(:-.*$ J$I$*#=).& $8#4< P"#% #8"9 E I8"9 E X1 ("# .<.(#= $.
;8"9C
#8"9 68"9 88"9 $8"9 $8#4+ 68.9 $8"9
08"9 !8"9 !8"9
;)"<&)= 7'./&',,)& :(/4-,
<.!4/ %,-./*+01"/)2 '4/!4/
>8"9 08"9>8"9
0)."'& I8"9
! 8"9
!"#C(+-0A$%3C#++*+C#G'-($*%C$. 6 8 " 9 ! 0 8 " 9 ! 8"9 ; 8"9 ! ; 8"9
" D 8" 9
D*+ - /$H#5 &8"9 -%5 - 3$4#% ;8"91 -. ("# ,**& 3-$% D8"9 $%0+#-.#.1 ("# #//#0( */ ;8"9
*% ("# (+-0A$%3 #++*+ 5#0+#-.#.7 >% *("#+ 2*+5.1 ("# .#%.$($4$(< /'%0($*% 08"9 $. .=-,,
2"#% ("# ,**& 3-$% $. ,-+3#7 P# .-< ("-( ,-+3# ,**& 3-$% ,#-5. (* 3**5 5$.('+6-%0#
+#I#0($*%7 c*+# &+#0$.#,<1 /*+ 3**5 5$.('+6-%0# +#I#0($*%1 2# +#G'$+# - ,-+3# ,**&
3-$% *4#+ ("# /+#G'#%0$#. */ $%(#+#.( -..*0$-(#5 2$(" ("# #H&#0(#5 5$.('+6-%0#
.$3%-,.7

H="#/(:-.*$+($I$*#=).&+$8#4&+=.+#D$+9$$1:-*;+3).#(),+ !"#$%"
$8#4+
$8#4+ 68.9 $8"9
#8"9 68.9 88"9 $8"9 !8"9
08"9 !8"9

08"9>8"9
>8"9
D8"9CEC 08"9 &8"9>8"9
I8"9
0 8"9 !
!"#C(+-0A$%3C#++*+C#G'-($*%C$. " D8" 9

! 8"9
6 8"9 ! 0 8"9 ! 8"9 ; 8"9 ! ; 8 "9
" D 8" 9
>% &+-0($0#1 ("# 5$.('+6-%0# .$3%-,. ;8"9 -+# */(#% ,*2 /+#G'#%0<7 P"#% ("-( $. ("#
0-.#1 2# .-< ("-( 2# 2-%( ("# ,**& 3-$% (* 6# ,-+3# -( ,*2 /+#G'#%0$#.7 !"$. $.
#G'$4-,#%( (* .(-($%3 ("-( 2# 2-%( (* 5#.$3% ("# 0*%(+*,,#+ 08"9 .* ("-( ("#
.#%.$($4$(< /'%0($*% 08"9 $. .=-,, -( ,*2 /+#G'#%0$#.7

J
 H="#/(:-.*$+($I$*#=).&+$8#4&+=.+#D$+9$$1:-*;+3).#(),+ !"#$%"
D8"9CEC 08"9 &8"9>8"9 ;8"9C
#8"9 68"9 88"9 $8"9
08"9 !8"9
0 8"9 ! ;)"<&)= :(/4-,
" D8" 9 7'./&',,)& %,-./*+01"/)2
<.!4/ '4/!4/
D 8" 9 >8"9
7 8 "9 !
" D8 "9 0)."'& I8"9
0 8 "9 " 7 8 "9 !
! 8"9
6 8"9 ! 0 8"9 ! 8"9 ; 8"9 ! ; 8 "9
" D 8" 9
H="#/(:-.*$ J$I$*#=).& $8#4< N*%.$5#+$%3 ("# (+-0A$%3 #++*+ $% 68"91 $( $. #4$5#%(
("-(1 /*+ - 3$4#% &8"91 (* +#5'0# ("# $%/,'#%0# */ ("# 5$.('+6-%0#1 ;8"91 *% ("#
(+-0A$%3 #++*+1 68"91 2# 5#.$+# ("# =-3%$('5# QD8J(9Q *4#+ ("# +-%3# */ /+#G'#%0$#.1
(1 ("-( 0"-+-0(#+$B# ("# 5$.('+6-%0#.7
!"-( 2-<1 ("# (+-%./#+ /'%0($*% &8"9]8 [ D8"99 2$,, 6# .=-,,1 ("#+#6< +#5'0$%3 ("#
$%/,'#%0# */ ;8"91 W$%0# D8"9 E 08"9 &8"9>8"9 ("$. $=&,$#. ("-( 2# %##5 (* 5#.$3%
("# 0*%(+*,,#+ 08"9 (* "-4# - ,-+3# =-3%$('5# Q 08J(9Q7

K$-"/($%$.#+.)="$+-##$./-#=).+=.+#D$+9$$1:-*;+3).#(),+ !"#$%"
K$-"/($%$.# L)="$ G##$./-#=).& %8#4< P"#% #8"9 E ;8"9 E X1 ("# .<.(#= $.
;8"9C
#8"9 68"9 88"9 $8"9
08"9 !8"9
;)"<&)= 7'./&',,)& :(/4-,
<.!4/ %,-./*+01"/)2 '4/!4/
>8"9
$ 8"9 D8 " 9
0)."'& I8"9 !
D8"9CEC 08"9 I 8"9 " D8"9
&8"9>8"9
%8#4+ 68.9 $8"9
K 08"9 !8"9>8"9
D 8" 9
7 8 "9 !
" D8 "9
D 8" 9
!"#C(+-0A$%3C#++*+C#G'-($*%C$. 6 8 " 9 ! 7 8 " 9 I 8 " 9 ! I 8"9
" D8 " 9
;. ("# ,**& 3-$% QD8J(9Q 5#0+#-.#.1 ("# #//#0( */ I8"9 *% ("# (+-0A$%3 #++*+ 5#0+#-.#.7
>% *("#+ 2*+5.1 ("# 0*=&,#=#%(-+< .#%.$($4$(< /'%0($*% 78"9 $. .=-,, 2"#% ("# ,**&
3-$% QD8J(9Q $. .=-,,7 >/ 2# 5#.$3% 08"9.'0" ("-( QD8J(9Q dd 1 ("#% ("# %*$.# $.
-((#%'-(#5 6#0-'.#
7 8 "9 & D8 "9

M
 K$-"/($%$.#+.)="$+-##$./-#=).&+%8#4&+=.+#D$+9$$1:-*;+3).#(),+
!"#$%"

;8"9C

#8"9 68.9 88"9 $8"9 %8#4+ 68.9 $8"9

08"9 !8"9 08"9 !8"9>8"9

>8"9
%8#4
D 8" 9
!"#C(+-0A$%3C#++*+C#G'-($*%C$. 6 8 " 9 ! 7 8 " 9 I 8 " 9 ! I 8"9
" D8 " 9

P# .-< ("-( .=-,, ,**& 3-$%1 QD8J(9Q dd 1 ,#-5. (* 3**5 %*$.# -((#%'-($*%7 c*+#
&+#0$.#,<1 /*+ #//#0($4# =#-.'+#=#%( %*$.# -((#%'-($*%1 2# %##5 - .=-,, ,**& 3-$%
*4#+ ("# /+#G'#%0$#. -..*0$-(#5 2$(" ("# #H&#0(#5 %*$.# .$3%-,.7

K$-"/($%$.#+.)="$+-##$./-#=).&+%8#4&+=.+#D$+9$$1:-*;+3).#(),+
!"#$%"

;8"9C

#8"9 68.9 88"9 $8"9 %8#4+ 68.9 $8"9

08"9 !8"9 08"9 !8"9>8"9

>8"9
%8#4
D 8" 9
!"#C(+-0A$%3C#++*+C#G'-($*%C$. 6 8 " 9 ! 7 8 " 9 I 8 " 9 ! I 8"9
" D8 " 9

>% &+-0($0#1 =#-.'+#=#%( %*$.# .$3%-,. -+# */(#% "$3" /+#G'#%0<7 !"'. 2# 2-%( ("#
,**& 3-$%1 QD8J(9Q1 (* 6# ,*2 -( "$3" /+#G'#%0$#.7 !"$. $. #G'$4-,#%( (* - .=-,,
0*=&,#=#%(-+< .#%.$($4$(< /'%0($*% 78"9 -( "$3" /+#G'#%0$#.7

R
K$-"/($%$.#+.)="$+-##$./-#=).&+%8#4&+=.+#D$+9$$1:-*;+3).#(),+
!"#$%"
D8"9CEC 08"9 &8"9>8"9 ;8"9C
#8"9 68"9 88"9 $8"9
08"9 !8"9
0 8"9 ! ;)"<&)= :(/4-,
" D8" 9 7'./&',,)& %,-./*+01"/)2
<.!4/ '4/!4/
D 8" 9 >8"9
7 8 "9 !
" D8 "9 0)."'& I8"9
0 8 "9 " 7 8 "9 !
D 8" 9
6 8"9 ! 7 8"9 I 8" 9 ! I 8"9
" D8 " 9
K$-"/($%$.# L)="$ G##$./-#=).& %8#4< (* -((#%'-(# ("# =#-.'+#=#%( %*$.#1
I8"91 -%5 +#5'0# ("# $%/,'#%0# *% ("# (+-0A$%3 #++*+1 68"91 2# 5#.$+# ("# =-3%$('5#
QD8J(9Q *4#+ ("# +-%3# */ /+#G'#%0$#.1 (1 ("-( 0"-+-0(#+$B# ("# =#-.'+#=#%( %*$.#7
!"# (+-%./#+ /'%0($*% D8"9]8 [ D8"99 2$,, 6# .=-,,1 ("#+#6< +#5'0$%3 ("# $%/,'#%0#
*/ I8"97 ;3-$%1 .$%0# D8"9 E 08"9 &8"9>8"9 ("-( $=&,$#. ("-( 2# %##5 (* 5#.$3% ("#
0*%(+*,,#+ 08"9 (* "-4# - .=-,, =-3%$('5# Q 08J(9Q7

F(()(+ =>.-,+G.-,!"="+)2+9$$1:-*;+ !"#$%"

D8"9CEC 08"9 &8"9>8"9 ;8"9C
#8"9 68"9 88"9 $8"9
08"9 !8"9
0 8"9 ! ;)"<&)= :(/4-,
" D8" 9 7'./&',,)& %,-./*+01"/)2
<.!4/ '4/!4/
D 8" 9 >8"9
7 8 "9 !
" D8 "9 0)."'& I8"9

6 8"9 ! 0 8"9 #8"9 0 8 "9 ! 8"9 ;8 "9 7 8"9 I 8"9

H="#/(:-.*$ J$I$*#=).& L)="$ G##$./-#=).& %8#4

$8#4 %##5. (* 5#.$3% ("# %##5. (* 5#.$3% ("#
0*%(+*,,#+ &*8#4 (* "-4# - 0 8 "9 " 7 8 "9 ! 0*%(+*,,#+ &*8#4 (* "-4# -
,-(>$ =-3%$('5# Q 08J(9Q "%-,, =-3%$('5# Q 08J(9Q

V
 F(()(+ =>.-,+G.-,!"="+)2+9$$1:-*;+ !"#$%"
;8"9C
#8"9 68"9 88"9 $8"9
&*8#4 !8"9
;)"<&)= '()*+(,,-+ :(/4-,
<.!4/ %,-./*+01"/)2 '4/!4/
>8"9
D 8" 9
0 8"9 ! 7 8 "9 !
" D8" 9 0)."'& I8"9 " D8 "9

H="#/(:-.*$ J$I$*#=).& L)="$ G##$./-#=).& %8#4

$8#4 %##5. (* 5#.$3% ("# D8"9CEC 08"9 &8"9>8"9 %##5. (* 5#.$3% ("#
0*%(+*,,#+ &*8#4 (* "-4# - 0*%(+*,,#+ &*8#4 (* "-4# -
,-(>$ =-3%$('5# Q 08J(9Q 0 8 "9 " 7 8 "9 ! "%-,, =-3%$('5# Q 08J(9Q

!"# .#&-+-($*% */ 5$.('+6-%0#.1 $8#4 8-( ,*2 /+#G'#%0$#.9 -%5 =#-.'+#=#%( %*$.#1
%8#4 8-( "$3" /+#G'#%0$#.9 $. 4#+< /*+('%-(# 6#0-'.# $( 3$4#. ("# 0*%(+*, .<.(#=
5#.$3%#+ - 2-< (* -&&+*-0" ("# 5#.$3% &+*0#..e ("# 0*%(+*,,#+1 &*8#4 ."*',5 6#
"$3" 3-$% -( ,*2 /+#G'#%0$#. -%5 ,*2 3-$% -( "$3" /+#G'#%0$#.7

$."=#=@=#!+)2+3,)"$1+7))6+3).#(),+ !"#$%"+#)+'-(-%$#$(+
A-(=-#=)."
;8"9C
#8"9 68.9 88"9 $8"9
LM,'NOMQD8J(9Q

08"9 !8"9

>8"9
QD8J(9Q1C,-+3#C-(C (N(P N-<.+(&'""'Q)&+ I8"9
,*2C/+#G'#%0$#. /+#G'#%0<
#)+($I$*#+ (
1="#/(:-.*$"&+$8"4& QD8J(9Q1C.=-,,C-(C"$3"C
/+#G'#%0$#.C
D'R+N-<."+/'+&)=4()+")."</<Q</1+/'+ #)+-##$./-#$+
")."'&+.'<")+-.=+2'=),+4.()&/-<./1 %$-"/($%$.#+.)="$&+%8#4&

D*+('%-(#,<1 ("# -&&-+#%( 0*%/,$0( 6#(2##% 2-%($%3 (* =-A# Q 08J(9Q ,-(>$ #) ($I$*#
1="#/(:-.*$"& $8"4& -%5 ("# 2-%($%3 (* =-A# Q 08J(9Q "%-,, #) -##$./-#$
%$-"/($%$.# .)="$& %8#4& 0-% 6# -11($""$1 =. #D$ 1$"=>. 6D-"$ 6< %-;=.> #D$
,))6 >-=.& M.8/(4M& ,-(>$ -# ,)C 2($N/$.*=$" 83#%#+-,,< -..*0$-(#5 2$(" ("#
/+#G'#%0< +-%3# */ 5$.('+6-%0#.91 -%5 %-;=.> M.8/(4M& "%-,, -# D=>D 2($N/$.*=$"
83#%#+-,,< -..*0$-(#5 2$(" =#-.'+#=#%( %*$.#97

Y
 $."=#=@=#!+)2+3,)"$1+7))6+3).#(),+ !"#$%"+#)+'-(-%$#$(+
A-(=-#=)."
;8"9C
#8"9 68.9 88"9 $8"9
LM,'NOMQD8J(9Q 08"9 !8"9

>8"9
QD8J(9Q1C,-+3#C-(C (N(P N-<.+(&'""'Q)&+ I8"9
,*2C/+#G'#%0$#. /+#G'#%0<
(
QD8J(9Q1C.=-,,C-(C"$3"C
/+#G'#%0$#.
D'R+N-<."+/'+&)=4()+")."</<Q</1+/'+
")."'&+.'<")+-.=+2'=),+4.()&/-<./1
;% $=&*+(-%( -54-%(-3# */ /##56-0A 0*%(+*, .<.(#=. $. ("# -6$,$(< (* +#5'0# ("#
#//#0( */ ("# 4-+$-($*% */ &-+-=#(#+. */ - 0*%(+*, .<.(#= 6< -55$%3 - /##56-0A ,**&7
!"# 0,*.#5),**& .<.(#= -,,*2. &8"9 (* 6# ,#.. -00'+-(#,< .&#0$/$#51 6#0-'.# ("#
.#%.$($4$(< (* 0"-%3#. *+ #++*+. $% &8"9 $. +#5'0#5 6< ("# ,**& 3-$% D8"97!"$.
6#%#/$( */ 0,*.#5),**& .<.(#=. $. - &+*/*'%5 -54-%(-3# /*+ ("# #,#0(+*%$0 -=&,$/$#+.
*/ ("# 0*=='%$0-($*% $%5'.(+<7 ; .$=&,# #H-=&,# 2$,, $,,'.(+-(# ("# 4-,'# */
/##56-0A /*+ +#5'0$%3 .#%.$($4$(<7

$."=#=@=#!+)2+9$$1:-*;+3).#(),+ !"#$%"+#)+'-(-%$#$(+A-(=-#=)."
Z#( .('5< - .$=&,# #H-=&,# (* $,,'.(+-(# ("# 4-,'# */ /##56-0A /*+ +#5'0$%3
.#%.$($4$(<7
01-)2,((12 31,454-+< ;%C-=&,$/$#+C'.#5C$%C=-%<C-&&,$0-($*%.C"-.C-C3-$%CK9-1C-.C
."*2%C$%C("#C/*,,*2$%3CD$3'+#7 !"#C*'(&'(C4*,(-3#C$.

!"#C*'(&'(C4*,(-3#C$. QX ! 9 - Q<.
Q$% -<. QX
K9- !"#C(+-%./#+C/'%0($*%C$. C ! 9-

'C 9 -
!"#C.#%.$($4$(<C(*C0"-%3#.C$%C("#C-=&,$/$#+C3-$%C$. 0 9C - ! !
'9 - C

\
 $."=#=@=#!+)2+9$$1:-*;+3).#(),+ !"#$%"+#)+'-(-%$#$(+A-(=-#=)."
6--7! 892 31,454-+< a/(#%C-C&*(#%($*=#(#+C#& <.C-55#5C(*C-0"$#4#C/##56-0AC-"+
."*2%C$%CD$3'+#7
Q$% QX K9-
T<.8"9 6-8.9 $8"9
-<. # [9-8) [ 9
#&
K9-
#L

#:
#& S # [C#: -%5C ) !
#
!"#C0,*.#5),**&C(+-%./#+C/'%0($*%C*/C("#C/##56-0AC-=&,$/$#+C$.
9-
C! !
" " )9 -
!"#C.#%.$($4$(<C*/C("#C0,*.#5),**&C/##56-0AC-=&,$/$#+C$.
0 9C - ! 0 C 0 9 - !
" )9 -

$."=#=@=#!+)2+9$$1:-*;+3).#(),+ !"#$%"+#)+'-(-%$#$(+A-(=-#=)."

T<.8"9 6-8.9
$8"9
6--7! 892 31,454-+< K9-

9-
!"#C0,*.#5),**&C(+-%./#+C/'%0($*%C$. C!
" )9 -
!"#C.#%.$($4$(<C*/C("#C.<.(#=C(*C&-+-=#(#+C 0 9C - ! 0 C 0 9 - !
4-+$-($*%.CC$. " )9-
>/C9- $.C,-+3#1C("#C.#%.$($4$(<C$.C,*27CD*+C#H-=&,#1C$/C9-E XJ -%5C)E XC("#%C("#C
.#%.$($4$(<C$.
0 9C - ! !
" )9 - " XF
!"#C=-3%$('5#C$.C*%#)("*'.-%5("C*/C("#C=-3%$('5#C*/C("#C*&#%),**&C-=&,$/$#+Cf

:X
 $."=#=@=#!+)2+3,)"$1+7))6+3).#(),+ !"#$%"+#)+'-(-%$#$(+
A-(=-#=)."
;8"9C
#8"9 68.9 88"9 $8"9
08"9 !8"9

>8"9
I8"9
C
a/(#%1C2#C.##AC(*C5#(#+=$%#C 0* 8" 9 2"#+#C+* $.C-C&-+-=#(#+C2$("$%C("#C(+-%./#+
/'%0($*%C*/C-C6,*0AC&68#4E+U.$%3C("#C0"-$%C+',#1C2#C/$%5C("-(

0*C 8 " 9 ! 0 C 0*
g#+< */(#%1 ("# (+-%./#+ /'%0($*% */ ("# .<.(#= C8"9 $. - /+-0($*% */ ("# /*+=
I 8 "1 * 9
C 8 "1 * 9 !
; 8 "1 * 9
2"#+#C* $.C-C&-+-=#(#+C("-(C=-<C6#C.'6I#0(C(*C4-+$-($*%C5'#C(*C("#C#%4$+*%=#%(7

$."=#=@=#!+)2+3,)"$1+7))6+3).#(),+ !"#$%"+#)+'-(-%$#$(+
A-(=-#=)."
;8"9C
#8"9 68.9 88"9 $8"9
08"9 !8"9

>8"9
I8"9
!"#%C2#C=-<C*6(-$%C("#C.#%.$($4$(<C(*C+* 6<C+#2+$($%3C_G'-($*%
'C 8 " 9 ] C 8 " 9 'C 8 " 9 8 " 9 ' ,% C 8 " 9
0C ! ! !
' 8 " 9 ] ! 8 " 9 ' ! 8 " 9 C 8 " 9 ' ,% ! 8 " 9
-.
' ,% C ' ,% I ' ,% ;
0*C ! ! ! 0*I 0*;
' ,% * ' ,% * *X ' ,% * * X

2"#+#C*X $.C("#C%*=$%-,C4-,'#C*/C("#C&-+-=#(#+7

:
 !"#$%!&'( )!*+,#,-,#.&/*0&122#&32"$+&24&52*#%23

6!!07/"8&52*#%23&).+#!9+
!"# $% &'# ()*+# )#,-$"- %$) &'# .-# $% "#/,&*0# %##12,34 *" 3$"&)$5 -6-&#+- *- &'#
)#1.3&*$" $% &'# #%%#3& $% (,),+#&#) 0,)*,&*$"-7
&:!; 3:!; 1:!; ':!;
$3:!; " $%:!;
7,!8),9 =3&.,&$) 4+/560
-*./)*00,) ()*+,!! ),!%*.!,
),!%*.!,
2:!;
:,6!5),;,./<*5/%5/ =,,9>6+?
,.!*)
8'# #%%#3& $% (,),+#&#) 0,)*,&*$"- 3," 2# 1#-3)*2#1 26 , +#,-.)# $% &'# -#"-*&*0*&69
:!;9 $% &'# -6-&#+ (#)%$)+,"3# &$ +:!",4," :/%/9!#!% :5#& -,6 "; ";/*<!+7 8'#
32</%,#;9," +!*+,#,-,#. *- $)*/*",556 -.//#-&#1 7. =20! ,-

# 5" # : ! ; # : ! ;# : ! ; # : !; "
" :!; ! ! !
5" " "<" " # : !;
' : !; "$+ : !;$ % : !;
>'#)#?&'#?-6-&#+?&),"-%#)?%."3&*$"9?#:!;9?*- # : !; ! !
&: ! ; " "$+ : !;$ % : !; 2 : !;
,"1?&'#?(,),+#&#)?$%?*"&#)#-&?*-?"7
 6!!07/"8&52*#%23&).+#!9+
&:!; 3:!; 1:!; ':!;
$3:!; " $%:!;
7,!8),9 =3&.,&$) 4+/560
-*./)*00,) ()*+,!! ),!%*.!,
),!%*.!,
2:!;
:,6!5),;,./<*5/%5/ =,,9>6+?
,.!*) ;

' :!; "$+ :! ;$ % :! ;

" #!" A
@!
@
8'#?&),"-%#)?%."3&*$"?*-E ! ! .
&:! ; " "$+ :! ;$ % :! ; 2 : ! ;
8!
#!" ) 8
A'#)# " *- &'# 35$-#1 5$$( -6-&#+ /,*"7
B& ',- 2#3$+# .-#%.5 &$ 1#%*"# , -#"-*&*0*&6 +#,-.)# *" &#)+- $% &'# ($-*&*$"- $% &'#
%22#+ 24 #;! ";/%/"#!%,+#," !>$/#,2*7
(: ! ; ! " "$+ : ! ;$ % : ! ; 2 : ! ;
C#3,.-# &'# )$$&- $% &'# 3',),3&#)*-&*3 #D.,&*$" )#()#-#"& #;! 029,*/*# 920!+ $%
&),"-*#"& )#-($"-#9 &'# #%%#3& $% (,),+#&#) 0,)*,&*$"- $" &'# ($-*&*$" $% &'# )$$&- *-
," *+($)&,"& ,"1 .-#%.5 +#,-.)# $% &'# -#"-*&*0*&67

6!!07/"8&52*#%23&).+#!9+(&122#& 2"$+&?32#+
8'# )$$& 5$3.- *- &'# (,&' $% &'# )$$&- $% &'# 3',),3&#)*-&*3 #D.,&*$" &),3#1 $.& *"
&'# !F(5,"# ,- , -6-&#+ (,),+#&#) 0,)*#- %)$+ G#)$ &$ *"%*"*&69 H I"I$7

':!; K',),3&#)*-&*3?#D.,&*$"J
&:!; 3:!;
"$(:!;
(: ! ; ! " "$ % : ! ; 2 : ! ; ! H
2:!; "$ % : !; 2 : ! ; ! %

8'# 3',),3&#)*-&*3 #D.,&*$" 3," 2# -(5*& *"&$ &>$ #D.,&*$"- 26 #D.,&*"/ #;! /*<3!+
,"1 9/<*,#$0!+ $% 2$&' -*1#-9 )#-(#3&*0#569 &$ $2&,*" &'# %$55$>*"/J

"$ % : !; 2 : !; ! %
A*<3!&52*0,#,2*(& :6; '"$ % : !; 2 : !; ! & MHH :@? " ;9 :? ! H9 9@9777;
@/<*,#$0!&52*0,#,2*(& :>; L "$ % : ! ; 2 : ! ; L!

@
 122#&)!*+,#,-,#.&24&6!!07/"8&52*#%23&).+#!9+
8'#?#%%#3&?$%?(,),+#&#)?0,)*,&*$"-9 ?9?$"?&'#?($-*&*$"?$%?&'#?)$$&- $%?3',),3&#)*-&*3?
#D.,&*$"9?(:!;9?*-?.-#%.5?+#,-.)#?$%?&'#?-#"-*&*0*&67?

&:!; 3:!; ':!;

?$(:!;

2:!;

?$ % : !;
8'#?&),"-%#)?%."3&*$"?*-E # :! ; !
" ?$ % : !; 2 : ! ;
K',),3&#)*-&*3?#D.,&*$"J
(: ! ; ! " ?$ % : ! ; 2 : !; ! H
O$)?&'#?%*P#1?/,*"??E
(: !; ! : ! " ) ;: ! " )@ ;777:! " )8 ;777:! " ). ; ! H

122#&)!*+,#,-,#.&24&6!!07/"8&52*#%23&).+#!9+
O$)?&'#?%*P#1?/,*"?? &'#?3',),3&#)*-&*3?#D.,&*$"?*-E &:!; 3:!; ':!;
?$(:!;
(: !; ! " ?$ % : ! ; 2 : !; ! H
2:!;
! : ! " ) ;: ! " )@ ;777:! " )8 ;777:! " ). ; ! H
8'# %22# +!*+,#,-,#. $% , -6-&#+ #:!; 3," 2# 1#%*"#1 ,-

)8 )8 )8
? : !; ! ! ;
5" ? ?<?
>'#)# )* #D.,5- &'# 8&' )$$& ?$ % : ! ;
" #! " A
@!
@

# : !; ! ! .
$% &'# -6-&#+9 -$ &',& " ?$ % : !; 2 : ! ;
#! " )
8!
8

,"1?? *-?,?(,),+#&#)?,%%#3&*"/?&'#?)$$&-7?
8'#?)$$&?-#"-*&*0*&6?)#5,&#-?&'#?3',"/#-?*"?&'#?5$3,&*$"?$%?&'#?)$$&?*"?&'#?!F(5,"#?
&$?&'#?3',"/#?*"?&'#?(,),+#&#)9??7

N
 122#&)!*+,#,-,#.&24&6!!07/"8&52*#%23&).+#!9+
8'# %22# +!*+,#,-,#. $% , -6-&#+ #:!; *- )8 )8 )8
? : !; ! !
; 5" ? ? <?
)* #D.,5- &'# 8&' )$$& $% " #! " A
@!
@
&:!; 3:!; ':!;
&'# -6-&#+ -$ &',& # : !; ! .
?$(:!;
#! " ) 8
8! 2:!;
B;!&%22#&+!*+,#,-,#.&*-?)#5,&#1?&$?&'#?5$/,)*&'+*3?-#"-*&*0*&6?26?&'#?)#5,&*$"
.
5" " )
#
? :!; ! %( 8
5" ? 8 ! 5" ? ! " )8
A@
>'#" &'# G#)$- $% #:!; ,)# *"1#(#"1#"& $% &'# (,),+#&#) ?9 -$ &',& !H
5" ?
O$) &'*- (,)&*3.5,) 3,-#9 >'#" &'# /,*" $% &'# -6-&#+ *- *"1#(#"1#"& $% &'#
(,),+#&#) ?9 &'#"
.
)8 )8
#
? : ! ; ! %( ? >'#)# )8
? !
8! ! " )8 5" ?
,"1?&'#?&>$?-#"-*&*0*&6?+#,-.)#-?,)#?1*)#3&56?)#5,&#17

122#&)!*+,#,-,#.&24&6!!07/"8&52*#%23&).+#!9+
8'#?#0,5.,&*$"?$%?&'#?)$$&?-#"-*&*0*&6?%$)?,? &:!; 3:!; ':!;
"$(:!;
3$"&)$5?-6-&#+?3,"?2#?)#,1*56?,33$+(5*-'#1?
26?.&*5*G*"/?&'#?)$$&?5$3.-?+#&'$1-7 2:!;
8'#?)$$&?-#"-*&*0*&6

)8 )8 )8
? ! !
5" ? ? <?
+,6?2#?#0,5.,&#1?,&?)$$&?F)* 26?#P,+*"*"/?&'#?)$$&?3$"&$.)-?%$)?&'#?(,),+#&#)??7?
A#?3,"?3',"/#?? 26?,?-+,55?%*"*&#?,+$."&?)? ,"1?1#&#)+*"#?&'#?+$1*%*#1?)$$&
F:)* R?))*;?,&?? R?)?7
)8 ))8
? +
)? < ?
8'*- SD.,&*$" *- ," ,(()$P*+,&*$" &',& ,(()$,3'#- &'# ,3&.,5 0,5.# $% &'#
-#"-*&*0*&6 ,- )? * H7

Q
 𝛼
Example 4.01: A unity feedback system has transfer function 𝐺𝑝 (𝑠) =
𝑠(𝑠+𝛽)
1. Draw the effect +∆𝛼 on the root locus
2. Draw the effect −∆𝛼 on the root locus
3. Draw the effect +∆𝛽 on the root locus
4. Draw the effect −∆𝛽 on the root locus
𝑟1 𝑟2
5. Find the root sensitivity 𝑆𝛼+ , 𝑆𝛼+
𝑟1 𝑟2
6. Find the root sensitivity 𝑆𝛼− , 𝑆𝛼−
𝑟1 𝑟2
7. Find the root sensitivity 𝑆𝛽+ , 𝑆𝛽+

𝑟1 𝑟2
8. Find the root sensitivity 𝑆𝛽− , 𝑆𝛽−

with 𝛼 = 0.5 ± 0.1 𝛽 = 1 ± 0.2 ?

 122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9
CD/9:3!&'EFG(&K$"-*1#)?&'#?%##12,34 &:!; 3:!; ':!;
$(:!;
-6-&#+?>*&'? %$55$>*"/?25$34-E
- 2:!;
$ % : !; ! 9 2 : !; !
!: ! " , ;
8'#?-6-&#+?&),"-%#)?%."3&*$"?*-J
' :!; $ % : !; -
! !
&:! ; " $ % : !; 2 :!; !: ! " , ; " -
U#&?3$"-*1#)?&'#?/,*"??B - &'#"?&'#?3',),3&#)*-&*3?#D.,&*$"?*-J
(- : ! ; ! " $ % : ! ;$ : !; ! H
?
! " !H
!:! " , ;
! !:! " , ; " ? ! H
! : ! " ) ;: ! " )@ ; ! H

122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9
CD/9:3!&'EFG(&O$)9?"B - &'#?3',),3&#)*-&*3?#D.,&*$"?*-J
&:!; 3:!; ':!;
(- : ! ; ! !: ! " , ; " ? ! !: ! " , ; " - ! H $(:!;

! : ! " ) ;: ! " )@ ; ! H 2:!;

8'#" &'# #%%#3& $% , 3',"/# *" #,3' (,),+#&#) 3," 2# 1#&#)+*"#1 26 .&*5*G*"/ &'#
)#5,&*$"-7
; - ! - H & )- ,"1
@; , ! , H & ),
>'#)# -H ,"1 ,H ,)# &'# "$+*",5 $) 1#-*)#1 0,5.#- %$) &'# (,),+#&#)- - ,"1 ,
)#-(#3&*0#567
U#& 3$"-*1#) &'# 3,-# >'#" &'# "$+*",5 ($5# 0,5.# *- ,H V ,"1 &'# 1#-*)#1 /,*" *-
-H V ? V H7T7
; - ! H7T & )- ,"1
@; , ! & ),

T
 122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9
CD/9:3!&'EFG( &:!; 3:!; ':!;
- $(:!;
$ % : !; ! 9 2 : !; !
!: ! " , ;
2:!;
K',),3&#)*-&*3?#D.,&*$"J (- : ! ; ! ! : ! " , ; " - ! H

; - ! H7T & )- ,"1

@; , ! & ),
8'#" &'# )$$& 5$3.- 3," 2# $2&,*"#1 ,- , %."3&*$" $% - V ? 26 .&*5*G*"/ &'# )$$&
5$3.- #D.,&*$" ,- -'$>" *" O*/.)#J @/
? -H ?VH7W
(- : ! ; ! " $ % : ! ;$ : ! ; ! " ! " !H ) ?VH7T @<H7T
!: ! " , H ; ! :! " , H ; ?VH7Q
! !@ " ,H! " -H
C H .
8'# "$+*",5 0,5.# $% /,*" -H V ? V H7T ,"1 ,H V
)#-.5&- *" &>$ 3$+(5#P )$$&-9 F) 9@ V F H7T & @H7T ,- )@ ?VH7T C@<H7T
-'$>" *" O*/.)#7
!C%06.,

122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9
CD/9:3!&'EFG(
? - & )- &:!; 3:!; ':!;
(- : !; ! " ! " H !H $(:!;
!: ! " , H ; !: ! " , H ;
GH B;! !44!"# 24 ";/*<!+ ,* #;! </,*I &)- 42% 4,D!0 2:!;
,FJGK 8$ #0,5.,&# &'# #%%#3& $% .",0$*1,25# 3',"/#- *" @/
)-VH7 - VH7W
&'# /,*"9 &'# 3',),3&#)*-&*3 #D.,&*$" >*&' -V-H&)- ) - VH7T @<H7T
2#3$+#- )-VFH7 - VH7Q

(- : ! ; ! ! @ " , H ! " - H & )-

C H .
! ! @ " ! " H7T & )-
)@ - VH7T C@<H7T
8'#)#%$)#9 &'# #%%#3& $% 3',"/#- *" &'# /,*"9 ? V-H9 !C%06.,
3," 2# #0,5.,&#1 %)$+ &'# )$$& 5$3.- $% O*/.)#7 ) 9@ V?F H7T?& @H7T?%$), H V? 9
-HV?V?H7T
O$) , @HX 3',"/# *" -9 ># ',0# )- V YH7 7 8'# )$$& 5$3,&*$"- %$) , /,*" -VH7Q
,"1 -V H7W ,)# )#,1*56 1#&#)+*"#1 26 )$$& 5$3.- +#&'$1-9 ,"1 &'# )$$& 5$3,&*$"-
%$) )- V YH7 ,)# -'$>" *" O*/.)#7

W
 122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9
CD/9:3!&'EFG(& &:!; 3:!; ':!;
$(:!;
GH B;! !44!"# 24 ";/*<!+ ,* #;! </,*I ")- 42%
4,D!0 , FJGK 2:!;
(- : ! ; ! ! @ " , H ! " - H " )- ! ! @ " ! " H7T " )-
)- E)- @/
! "
! @ " ! " H7T )-VH7 - JFEL
A'#"?- J !J&FEL9?&'#?)$$&?*"?&'#?-#3$"1?D.,1),"&? ), V?H ) - VH7T @<H7T
$%?&'#?! (5,"#?*- )-VFH7 - VH7Q

:%) ; " )) ! %H7T " @)H7T[ C H .

&'#?3',"/#?*"?/,*"?)? V:H7WFH7T;VH7 ?3,.-#?&$?3',"/#?*"?
&'#?)$$&?2#?)) V@:H7T[FH7T;V@H7H[?,"17? )@ - VH7T C@<H7T
8'.-9 &'# )$$& -#"-*&*0*&6 %$) ) *- !C%06.,
) )) " @ H7H[
?" ! ! ! @ H7QT ! H7QT' " [HH ) 9@ V?F H7T?& @H7T?D*) , H V? 9
)? < ? " H7@ 6.9<-HV?V?H7T
42%&:2+,#,-!&";/*<!+&$%?/,*"I ")-K

122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9
CD/9:3!&'EFG(
GH B;! !44!"# 24 ";/*<!+ ,* #;! </,*I ) - 42% 4,D!0 , FJGK
&:!; 3:!; ':!;
(- : ! ; ! ! @ " ! " H7T " )- $(:!;
A'#"?- J !J&FE'9?)? V:H7QFH7T;VFH7 9?&'#?)$$&?
*"?&'#?-#3$"1?D.,1),"&?$%?&'#?! (5,"#?*- 2:!;
#JFEL @/
: % ) ; " )) ! %H7T " @) H7NMZ )-VH7
), V?H ) "VH7T @<H7T
,"1 &'# 3',"/# *" &'# )$$& )) V F@H7 7 )-VFH7 #JFE'
62% *!</#,-! ,*"%!9!*#+I M) -I $% /,*"E C)-

(- : ! ; ! ! @ " , H ! " - H % )- ! ! @ " ! " H7T % )-

C FH7T H .
)-
! % @
! " ! " H 7T
!C%06.,
&'# )$$& -#"-*&*0*&6 %$) ) *- )@ "VH7T C@<H7T
)) % @ H7 ) 9@ V?F H7T?& @H7T?D*)<, H V? 9
)
?% ! ! ! % @ H7TT ! H7TT' % [HH 6.9<-HV"V?H7T
)? < ? " H 7@

Z
 122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9
CD/9:3!&'EFG(
&:!; 3:!; ':!;
GH B;! !44!"# 24 ";/*<!+ ,* #;! </,*I &)- $(:!;
42% 4,D!0 , FJGK
2:!;
) )) @/
?% !
)? < ? )-VH7 JFEL
O$) *"%*"*&#-*+,556 -+,55 3',"/#- *" &'# ), V?H ) ?VH7T @<H7T
)-VFH7 ?VH7Q
(,),+#&#) &)- 9 &'# -#"-*&*0*&6 >*55 2# #D.,5 %$)
"#/,&*0# $) ($-*&*0# *"3)#+#"&- *" ?7 C H .
C)-
)@ ?VH7T C@<H7T
8'# ,"/5# $% &'# )$$& -#"-*&*0*&6 *"1*3,&#- &'#
1*)#3&*$" &'# )$$& +$0#- ,- &'# (,),+#&#) 0,)*#-7 !C%06., E)-
8'# ,"/5# $% +$0#+#"& %$) R)- *- ,5>,6- MH\
%)$+ &'# ,"/5# $% +$0#+#"& %$) F)- ,& &'# ($*"& ) 9@ V?F H7T?& @H7T?D*) , H V? 9
- V -HF 6.9<-HV?V?H7T

122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9
CD/9:3!&'EFG(
NH B;! !44!"# 24 ";/*<!+ ,* :23!I &),$ 42% 4,D!0 J -F J FEOK
8'# ($5# , 0,)*#- 1.# &$ #"0*)$"+#"&,5 3',"/#-9 &:!; 3:!; ':!;
$(:!;
,"1 *& +,6 2# )#()#-#"&#1 26 ,V,$&),9 >'#)# ,$V 7
8'#" &'# #%%#3& $% 0,)*,&*$" $% &'# ($5#- *- 2:!;
)#()#-#"&#1 26 &'# 3',),3&#)*-&*3 #D.,&*$"7
-
- $ % :! ; !
(, : !; ! " $ % : ! ; 2 : ! ; ! " ! :! " , H & ), H ;
!: ! " , ; , ! , H & ), H
2 : !; !
-
! " ! ! @ " : , H & ), H ; ! " -
!: ! " , H & ), H ;
O$)?,$V? ?,"1??V?-H?V?H7T?&'#"?*"?)$$&?5$3.-?%$)+9
(, : !; ! ! @ " : & ), H ; ! " H7T ! H
& ), !
! " !H
! @ " ! " H7T

M
 122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9
CD/9:3!&'EFG(
NH B;! !44!"# 24 ";/*<!+ ,* :23!I &),$ 4,D!0 J-F J FEOK
-
$ % : !; ! 9 2 :! ; !
! :! " , & ),! ;
&:!; 3:!; ':!;
), ! $(:!;
(, :!; ! & @ !H @/
! " ! " H7T
2:!; )- V?H )
), ! ,H V? 9 @<H7T
! & -HV?H7T
: ! " H7T " @H7T;: ! " H7T % @ H7T;
62% :2+,#,-! ";/*<!+ 24 P), K
O$) ,H V 9 ,"1 -HV"V H7T &'#" &'# $(#" 5$$(
-6-&#+ ',- &>$ 3$+(5#P )$$&- ,"1 $"# G#)$ ,& C H .
C%GIN J MFEO & &FEO ,"1 ,& ' J FI )#-(#3&*0#567

), !
(, : !; ! "
: ! " H7T " @ H7T;:! " H7T % @ H7T; !C%06., )@ C@<H7T

122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9
CD/9:3!&'EFG(
NH B;! !44!"# 24 ";/*<!+ ,* :23!I ), 42% 4,D!0 -F J FEOK
62% :2+,#,-! ";/*<!+ 24 P), K &'# 3',),3&#)*-&*3 #D.,&*$" *-
!
(, :! ; ! " ), !H @/
: ! " H7T " @ H7T;:! " H7T % @ H7T;
20 V@@TH @<H7T
)#D.*)#-?&',&?&'#?)$$&?5$3.-?-,&*-%6 )
&'#?#D.,&*$"-
), ( : ! ; ! % 1
:6 ; '),( : ! ; ! & MHH :@? " ;9 :? ! H9 9@9777;
C H .
:>; L ),( : ! ; L!
="/5#?$%?1#(,)&.)#?%)$+?,?3$+(5#P?($5#?) ?*- 20@
)- V?H
1 % 0 % 0 @ ! % MH )@ !C%06., C@<H7T
H H
0 ! MH " 1 % 0 @ ! MH " NT % [H ! @@T H H H ,H V? 9 -HVH7T9 6.9<R),

[
 122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9
CD/9:3!&'EFG(
NH B;! !44!"# 24 ";/*<!+ ,* :23!I ), 42% 4,D!0 -F J FEOK

? !- &:!; 3:!; ':!;

$ % : !; ! 9 2 : !; ! $(:!;
!: ! " , ; @/
2:!; ) @<H7T
,H V? 9
R), -HV?H7T
62% :2+,#,-! ";/*<!+ 24 P),K
&'# 3',),3&#)*-&*3 #D.,&*$" *-

!
(, :! ; ! " ), !H C H .
: ! " H7T " @ H7T;: ! " H7T % @H7T;

8'#?)$$&?5$3.-?$%?&'#?-6-&#+?*-
R),
)@ C@<H7T
!C%06.,
,H V? 9 -HVH7T9 6.9<R),

122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9
CD/9:3!&'EFG(
NH B;! !44!"# 24 ";/*<!+ ,* :23!I (), 42%
&:!; 3:!; ':!;
4,D!0 -F J FEOK $(:!;
? !-
$ % :! ; ! 9 2 : !; !
!:! " , % ), ; 2:!;
62% *!</#,-! ";/*<!+ 24 M),K =/,*" &'# $(#" 5$$( -6-&#+ ',- &>$ 3$+(5#P
)$$&- ,"1 $"# G#)$ ,& %GIN J MFEO & &FEO ,"1 ,& ' J F )#-(#3&*0#567 8'# 3',),3&#)*-&*3
#D.,&*$" *- ),!
(, :!; ! %
:! " H7T " @H7T;:! " H7T % @ H7T;
B&?)#D.*)#-?&',&?&'#?)$$&?5$3.-?-,&*-%6?&'#?#D.,&*$"-
(, : ! ; ! % ),( : !; ! H >'#)# !
( :! ; !
: ! " H7T " @ H7T;: ! " H7T % @ H7T;
), ( : ! ; !
:6 ; '),( : ! ; ! H H & ? NWHH 9 :? ! 9@9777;
:>; L ),( : ! ; L!

H
 122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9
CD/9:3!&'EFG(
NH B;! !44!"# 24 ";/*<!+ ,* :23!I (), 42% 4,D!0 -F J FEOK
? !- &:!; 3:!; ':!;
$ % :! ; ! 9 2 : !; ! $(:!;
!:! " , % ), ;
@/
2:!;
62% *!</#,-! ";/*<!+ 24 M), K )
,H V? 9 @<H7T
8'# 3',),3&#)*-&*3 #D.,&*$" *- (, : ! ; ! % ),( : !; ! H -HV?H7T

>'#)# !
( :! ; !
: ! " H7T " @ H7T;:! " H7T % @ H7T;
8'#?)$$&?5$3.-?)#D.*)#-?&$?-,&*-%6?&'#?%$55$>*"/? C H .
#D.,&*$"-

), (:! ; !
)@ C@<H7T
:6; '),(: !; ! H H & ? NWH H 9 : ? ! 9@9777;
!C%06.,
:>; L ),(: !; L!

122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9
CD/9:3!&'EFG(
NH B;! !44!"# 24 ";/*<!+ ,* :23!I ), 42% 4,D!0 -F J FEOK
? !- &:!; 3:!; ':!;
$ % :! ; ! 9 2 : !; ! $(:!;
!:! " , % ), ; @/
62% *!</#,-! ";/*<!+ 24 M), K 2:!; 20 VQTH
8'# 3',),3&#)*-&*3 #D.,&*$" *- @<H7T
)
(, : ! ; ! % ),( : !; ! H
!
>'#)# ( :! ; !
: ! " H7T " @ H7T;:! " H7T % @ H7T;
C .
O)$+?&'#?,"/5#?#D.,&*$" H
:6; '),(: ! ; ! H H & ? NWH H 9 : ? ! 9@9777;
8'#?,"/5#?$%?1#(,)&.)#?%)$+?,?3$+(5#P?($5#?) ?*-
)@ !C%06., C@<H7T
1 % 0 % 0 @ ! HH
,H V? 9 -HV?H7T9 6.9<F),
0 ! H H " 1 % 0 @ ! H H " NTH % [HH ! QTH
 122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9
CD/9:3!&'EFG(
NH B;! !44!"# 24 ";/*<!+ ,* :23!I ), 42% 4,D!0 -F J FEOK

&:!; 3:!; ':!; ? !-

$(:!; $ % :! ; ! 9 2 : !; !
!:! " , % ), ; @/
2:!;
F),
,H V? 9
62% *!</#,-! ";/*<!+ 24 M), K -HV?H7T 20 VQTH
&'# 3',),3&#)*-&*3 #D.,&*$" *- ) @<H7T

(, : ! ; ! " ), (: !; ! H
C H
.
>'#)#
)@ C@<H7T
! F),
( :! ; !
: ! " H7T " @ H7T;: ! " H7T % @ H7T; !C%06.,
,H V? 9 -HVH7T9 6.9<F),
8'#?)$$&?5$3.-?$%?&'#?-6-&#+?*-

122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9
CD/9:3!&'EFG(
NH B;! !44!"# 24 ";/*<!+ ,* :23!I ), 42% 4,D!0 -F J FEOK
@/
&:!; 3:!; ':!;
$(:!;
,H V? 9 ), V?FH7@
2:!; -HV?H7T 20 VQTH
? !- ) @<H7T
$ % :! ; ! 9 2 :!; ! ), V??H7@
! : ! " , & ), ;

(, : ! ; ! & ), (: !; ! H C H
.
R),
! )@
>'#)# ( :! ; ! C@<H7T
: ! " H7T " @ H7T;: ! " H7T % @ H7T; F),
!C%06.,
8$?%*"1?&'#?#%%#3&?$%?,?@HX?3',"/#?$%?&'#? ,H V? 9 -HVH7T9 6.9<F),
(,),+#&#)?, 9?>#?#0,5.,&#?&'#?"#>?)$$&-?%$)?
),V?YH7@H9?,-?-'$>"?*"?O*/.)#7

@
 122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9
CD/9:3!&'EFG(
NH B;! !44!"# 24 ";/*<!+ ,* :23!I ), 42% 4,D!0 -F J FEOK
&:!; 3:!; ':!;
? !- $(:!;
$ % :! ; ! 9 2 :!; ! @/
! : ! " , & ), ;
2:!;
,H V? 9 ), V?FH7@
8'#?)$$&?-#"-*&*0*&6?*-?)#,1*56?#0,5.,&#1?/),('*3,556? -HV?H7T 20 VQTH
,"19?%$)?,?($-*&*0#?3',"/#?*"?,9?*- ) @<H7T
), V??H7@
) )) H7 W' % @MH
," ! ! ! H7M' % @M
), < , H 7@ C H
.
8'#?)$$&?-#"-*&*0*&6?%$)?,?"#/,&*0#?3',"/# R),
)@ C@<H7T
*"?,9?*- F),
!C%06.,
) )) H7 @T'N[H
,% ! ! ! H7W@T'N[ ,H V? 9 -HVH7T9 6.9<F),
), < , H 7@

122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9
CD/9:3!&'EFG(
NH B;! !44!"# 24 ";/*<!+ ,* :23!I ), 42% 4,D!0 -F J FEOK
&:!; 3:!; ':!;
? !- $(:!;
$ % :! ; ! 9 2 :!; ! @/
! : ! " , & ), ;
2:!;
=-?&'#?(#)3#"&,/#?3',"/#?), G, ,H V? 9 ), V?FH7@
-HV?H7T 20 VQTH
1#3)#,-#-9?&'#?-#"-*&*0*&6?+#,-.)#-
) @<H7T
) ,"1 )
," ,% ), V??H7@
A*55?,(()$,3'?#D.,5*&6?*"?+,/"*&.1#?,"1?,?
1*%%#)#"3#?*"?,"/5#?$%? MH\7?8'.-9?%$)?-+,55? C H .
3',"/#-?>'#"?), G, 3 H7 H9?&'#?-#"-*&*0*&6? R),
)@ C@<H7T
+#,-.)#-?,)#?)#5,&#1?,- F),
)
! )
,"1 )
! MHH " ) !C%06.,
," ,% ," ,%
,H V? 9 -HVH7T9 6.9<F),

N
 122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9

8$ .&*5*G# &'# )$$& -#"-*&*0*&6 +#,-.)# %$) &'# ,",56-*- ,"1 1#-*/" $% 3$"&)$5
-6-&#+-9 , -#)*#- $% 3,53.5,&*$"- +.-& 2# (#)%$)+#1E &'#6 >*55 1#&#)+*"# &'#
0,)*$.- -#5#3&*$"- $% ($--*25# )$$& 3$"%*/.),&*$"- ,"1 &'# G#)$- ,"1 ($5#- $% &'#
$(#"F5$$( &),"-%#) %."3&*$"7
8'#)#%$)#9 &'# )$$& -#"-*&*0*&6 +#,-.)# ,- , 1#-*/" &#3'"*D.# *- -$+#>',&
5*+*&#1 26 &>$ &'*"/-J
6; &'# )#5,&*0#56 5,)/# ".+2#) $% 3,53.5,&*$"- )#D.*)#1 ,"1
>; &'# 5,34 $% ," $20*$.- 1*)#3&*$" %$) ,1].-&*"/ &'# (,),+#&#)- *" $)1#) &$
()$0*1# , +*"*+*G#1 $) )#1.3#1 -#"-*&*0*&67
^$>#0#)9 &'# )$$& -#"-*&*0*&6 +#,-.)# 3," 2# .&*5*G#1 ,- ," ,",56-*- +#,-.)#9
>'*3' (#)+*&- &'# 1#-*/"#) &$ 3$+(,)# &'# -#"-*&*0*&6 %$) -#0#),5 -6-&#+
1#-*/"- 2,-#1 $" , -.*&,25# +#&'$1 $% 1#-*/"7

122#&)!*+,#,-,#.&24&/&6!!07/"8&52*#%23&).+#!9

8'# )$$& -#"-*&*0*&6 +#,-.)# *- , .-#%.5 *"1#P $% &'# -6-&#+_- -#"-*&*0*&6 &$
(,),+#&#) 0,)*,&*$"- #P()#--#1 *" &'# !F(5,"#7 8'# >#,4"#-- $% &'# -#"-*&*0*&6
+#,-.)# *- &',& *& )#5*#- $" &'# ,2*5*&6 $% &'# )$$& 5$3,&*$"- &$ )#()#-#"& &'#
(#)%$)+,"3# $% &'# -6-&#+7
=- ># 4"$> &'# )$$& 5$3,&*$"- )#()#-#"& &'# (#)%$)+,"3# D.*&# ,1#D.,&#56
%$) +,"6 -6-&#+-9 2.& 1.# 3$"-*1#),&*$" +.-& 2# /*0#" &$ &'# 5$3,&*$" $% &'#
G#)$- $% &'# 35$-#1F5$$( &),"-%#) %."3&*$" ,"1 &'# 1$+*","36 $% &'# (#)&*"#"&
)$$&-7
8'# )$$& -#"-*&*0*&6 +#,-.)# *- , -.*&,25# +#,-.)# $% -6-&#+ (#)%$)+,"3#
-#"-*&*0*&6 ,"1 3," 2# .-#1 )#5*,256 %$) -6-&#+ ,",56-*- ,"1 1#-*/"7

Q
 Lecture 5: Introduction to Robust Control Systems

Feedback Control Systems

A feedback control systems are generally designed based on the assumption that
all blocks of the feedback system are well known and fixed.
The model of the process and controller, and constant parameters are well
known and do not changed.
External
disturbance
d(t)
Gp(s)
r(t) e(t) Gc(s) K Plant, y(t)
Controller Actuator System,
Reference, u(t) Output
(Process)
Desired manipulated
e(t)=r(t)-y(t)
input variable
Difference
y(t) Sensor
Controlled variable (Measurement)
Internal disturbance
Noise n(t)

1
 Physical Feedback Control Systems
Physical systems and the external environment in which they operate cannot
be modeled precisely, may change in an unpredictable manner, and may be
subject to significant disturbances.

Unpredicted disturbance input, D(s)

R(s) E(s) U(s) Y(s)
Gc(s) Gp(s)
Desired Actual
Controller Plant, System output
input
H(s)
Measurement output
Sensor Sensor noise, N(s)

The design of control systems in the presence of significant uncertainty

requires the designer to seek a robust system.

Uncertainty in Control Systems

D(s)
R(s) E(s) U(s) Y(s)
Gc(s) Gp(s)

H(s)
N(s)
The process model will always be an inaccurate representation of the actual
physical system because of
• Parameter changes, k k
• Unmodeled dynamics Gp(s)  Gp(s)
• Unmodeled time delays
• Changes in equilibrium point (operating point)

• Sensor noise, N(s)

• Unpredicted disturbance inputs, D(s).

2
 Introduction to Robust Control Systems
The goal of robust systems design is to retain assurance of system performance in
spite of model inaccuracies and changes Gp(s)Gp (s) where │Gp(s)│>│ΔGp(s)│.
A system is robust when the system has acceptable changes in performance due to
model changes or inaccuracies.
Unpredicted disturbance
input, D(s)
R(s) E(s) U(s) Y(s)
Gc(s) Gp(s)Gp(s)
Controller System
H(s)
Sensor
Sensor noise, N(s)

A robust control system exhibits the desired performance despite the

presence of significant system (process) uncertainty.

Introduction to Robust Control Systems

The unmodeled dynamics and parameter changes may be significant or very large,
and for these systems the challenge is to create a design that retains the desired
performance.
Unpredicted disturbance
input, D(s)
R(s) E(s) U(s) Y(s)
Gc(s) Gp(s)Gp(s)
Controller System
H(s)
Sensor
Sensor noise, N(s)

Designing highly accurate systems in the presence of significant plant uncertainty

is a classical feedback design problem. The theoretical bases for the solution of
this problem date back to the works of H. S. Black and H. W. Bode in the early
1930s, when this problem was referred to as the sensitivities design problem.

3
 Introduction to Robust Control Systems
The designer seeks to obtain a system that performs adequately over a large range
of uncertain parameters.
A control system is robust when
1. it has low sensitivities (it is durable),
2. it is stable over the range of parameter variations (it is hardy), and
3. the performance continues to meet the specifications in the presence of a set of
changes in the system parameters (it is resilient).

Unpredicted disturbance
input, D(s)
R(s) E(s) U(s) Y(s)
Gc(s) Gp(s)Gp(s)
Controller Plant, System
H(s)
Sensor Sensor noise, N(s)

Introduction to Robust Control Systems

Robustness is the low sensitivity to effects that are not considered in the analysis
and design phase, for example, disturbances, measurement noise, and unmodeled
dynamics.
The system should be able to withstand these neglected effects when performing
the tasks (tracking error, speed control, etc.) for which it was designed.

Unpredicted disturbance
input, D(s)
R(s) E(s) U(s) Y(s)
Gc(s) Gp(s)Gp(s)
Controller Plant, System
H(s)
Sensor
Sensor noise, N(s)

4
 Robust Control Systems and System Sensitivity
For small-parameter perturbations, we may use, as a measure of robustness, the
differential sensitivities. D(s)
R(s) E(s) U(s) Y(s)
The system sensitivity is defined as Gc(s) Gp(s)

T ( s ) / T ( s ) H(s)
ST ( s )  N(s)
 / 
where  is the parameter and T(s) the transfer function of the system.
The root sensitivity is defined as
ri
Sri ( s ) 
 / 
When the zeros of T(s) are independent of the parameter , then system sensitivity
is n
T r 1
S ( s )   Si ( s )
i 1 s  ri
for an nth-order system.

Robust Control Systems and System Sensitivity

Example 5.1: Consider the following first order system where the variable
parameter is ;
E(s) Y(s)
1 R(s)
G p ( s)  , H (s )  1 Gp(s)
(s   )
H(s)
Y (s ) 1
The system transfer function is:  T (s ) 
R(s ) s   1
1
and the sensitivity function for  is; ST ( s )   Sr1 ( s )
s  r1
r1
Since -r1 =-[-(+1)] or r1 =(+1) and  Sr1 ( s )    
 / 
1 
ST ( s )   Sri ( s ) 
s  ri s    1

5
 Robust Control Systems and System Sensitivity
Example 5.2: Consider the following second order system where the system has
variable gain K;
K Y(s)
G p ( s)  , H ( s)  1 R(s) E(s)
s ( s  1) Gp(s)

The system transfer function is: H(s)

Y (s) K
 T ( s)  2 j
R(s ) s sK

K
and the sensitivity function for K is;
K=0 K=0.25 K=0
T s ( s  1) 0 
S (s)  S ( s)  2 -1 -0.5

K
K
s sK
s-plane
At the critical damping (s1,2 = -0.5),  = 1,
the gain is K=0.25. Root locus of the system

Robust Control Systems and System Sensitivity

Example 5.2: R(s) E(s) Y(s)
K Gp(s)
G p (s )  , H ( s)  1
s( s  1) H(s)
Y ( s) K Step Response
 T (s)  2 1.4
R( s ) s sK
1.2 K=1
The unit step response of the 1
system for selected gain K; K=1/4
Amplitude

0.8
K=1, s1,2 = -0.5 j0.5 3,  = 0.5. K=1/6
0.6
K=1/4, s1,2 = -0.5,  = 1.
0.4
K=1/6, s1 = -0.21, s2 = -0.78,  >1.
0.2

0
0 5 10 15 20 25 30
Time (sec)

The second order system with K=0.25 is critical damped (s1,2 = -0.5),  = 1.

6
 Robust Control Systems and System Sensitivity
Example 5.2: A Bode plot of the asymptotes of 20 log|C(j)| is shown in Figure
for K = 1/4 (critical damping, s1,2 = -0.5),  = 1.
K R(s) E(s) Y(s)
G p (s )  , H ( s)  1
s (s  1) Gp(s)
H(s)
The loop transfer function L(s) is 10

L(s) = Gp(s)H(s) 20 log|C(j)|

5
Actual
0 dB/dec
The complementary sensitivity 0

Magnitude(dB)
function C(s) is
-5
Asymptotes
L( s ) K -40 dB/dec
C ( s)   2 -10
1  L( s ) s  s  K
0.25 0.25 -15
 2   = 0.5 rad/s
s  s  0.25 ( s  0.5) 2 -20
-2 -1 0 1
10 10 10 10
Frequency (rad/sec)

Robust Control Systems and System Sensitivity

Example 5.2: A Bode plot of the asymptotes of 20 log|S(j)| is shown in Figure
for K = 1/4 (critical damping, s1,2 = -0.5),  = 1.
K R(s) E(s) Y(s)
G p ( s)  , H ( s)  1 Gp(s)
s( s  1)

L(s) = Gp(s)H(s) H(s)

10
The sensitivity function S(s) is
5 -20 dB/dec
1 20 log|S(j)| Actual
S ( s )  S KT ( s ) 
1  L( s ) 0
Asymptotes 0 dB/dec
Magnitude(dB)

s ( s  1)
 2 -5
s sK 20 dB/dec
s ( s  1) -10
Actual
 2
s  s  0.25
-15
s ( s  1)
  = 0.5  = 1 rad/s
( s  0 .5 ) 2 -20
-2 -1 0 1
10 10 10 10
Frequency (rad/sec)

7
 Robust Control Systems and System Sensitivity
Example 5.2: A Bode plot of the asymptotes of 20 log|C(j)| and 20 log|S(j for
K = 1/4 (critical damping, s1,2 = -0.5),  = 1.
K R(s) E(s) Y(s)
G p (s )  , H ( s)  1 Gp(s)
s (s  1)
L(s) = Gp(s)H(s) H(s)
10
L( s ) 0.25
C (s)  
1  L( s ) ( s  0.5) 2 5 -20 dB/dec
1 s ( s  1) 20 log|C(j)|
S (s )  
1  L ( s ) ( s  0 .5 ) 2 0
0 dB/dec 0 dB/dec

Magnitude(dB)
The following special relationship -5 Asymptotes Actual
between S(s) and C(s) holds 20 dB/dec
-10 -40 dB/dec
S ( s )  C ( s )  1,  S ( j )  C ( j )  1 20 log|S(j)|
-15

over the range of frequencies, ,  = 0.5  = 1 rad/s

of interest -20
10
-2
10
-1 0
10
1
10
Frequency (rad/sec)

Robust Control Systems and System Sensitivity

Example 5.2: A Bode plot of the asymptotes of 20 log|C(j)| and 20 log|S(j for
K = 1/4 (critical damping, s1,2 = -0.5),  = 1.
L(s) = Gp(s)H(s) R(s) E(s) Y(s)
Gp(s)
L( s ) 0.25 K
G p (s)  , H ( s)  1
C (s)   s( s  1) H(s)
1  L( s ) ( s  0.5) 2
1 s ( s  1)
S (s )   10

1  L ( s ) ( s  0 .5 ) 2
S ( s)  C (s)  1 5 -20 dB/dec

Note that the sensitivity S(s) is 20 log|C(j)|

0
small for lower frequencies, while 0 dB/dec 0 dB/dec
Magnitude(dB)

the complementary sensitivity -5 Asymptotes Actual

function (the closed loop transfer 20 dB/dec
function) C(s) primarily passes -10 -40 dB/dec
20 log|S(j)|
low frequencies.
-15
Of course, the sensitivity S(s) only
represents robustness for small  = 0.5  = 1 rad/s
-20
changes in gain. 10
-2
10
-1
Frequency (rad/sec)
0
10
1
10

8
 Robust Control Systems and System Sensitivity
Example 5.2:If K changes from 1/4 within the range K = 1/16 to K = 1, the
resulting range of step response is shown in Figure.
Step Response
1.4
R(s) E(s) Y(s)
Gp(s) 1.2 K=1
H(s) 1
K=1/4

Amplitude
0.8
K K=1/6
G p (s )  , H (s )  1
s (s  1) 0.6

0.4
This system, with an expected
wide range of K, may not be 0.2

considered adequately robust.. 0

0 5 10 15 20 25 30
Time (sec)

A robust system would be expected to yield essentially the same (within an

agreed-upon variation) response to a selected input.

Sensitivity of a Controlled System

Example 5.3: Consider the following PD controlled system where the blocks
transfer functions are
R(s) E(s) U(s) Y(s)
Gc ( s)  K P  K D s, G p ( s)  1 , H (s)  1 Gc(s) Gp(s)
s2
H(s)
L(s) = Gc(s)Gp(s)H(s)
Let consider the system with n=KP and KD = 2n to achieve =1. The sensitivity
function, S(s), with respect to changes in Gp(s) is
1 s2 s2 s2
S GT p ( s )   2  2 
1  L ( s ) s  K D s  K P s  2 s  1 ( s  1) 2

for n=1 red/sec.

The complementary sensitivity function C(s) is
Gc ( s )G p ( s ) K Ds  K P 2 ( s  0 .5 )
C (s )   
1  L (s ) s2  K Ds  K P ( s  1) 2

9
 Sensitivity of a Controlled System
Example 5.3: R(s) E(s) U(s) Y(s)
1 Gc(s) Gp(s)
Gc ( s)  K P  K D s, G p ( s )  , H (s)  1
s2
H(s)
L(s) = Gc(s)Gp(s)H(s)
For n=KP , KD = 2n , and n=1 red/sec, the sensitivity functions, S(s), and T(s).
1 s2 10
S GT p ( s )   2 5 20 dB/dec
1  L ( s ) ( s  1) 20 log|C(j)| 0 dB/dec
0
Gc ( s ) G p ( s ) 2( s  0.5) 0 dB/dec
C (s)   -5 Actual
1  L( s ) ( s  1) 2

Magnitude(dB)
-10
Asymptotes -20 dB/dec
The frequency n is an indicator -15

on the boundary between the -20 20 log|S(j)|

frequency region in which the -25

sensitivity is the important 40 dB/dec

-30

design criterion and the region in -35

which the stability margin is -40
 = 0.5 n = 1 rad/s
-2 -1 0 1 2
important. 10 10 10
Frequency (rad/sec)
10 10

Sensitivity of a Controlled System

Example 5.3: R(s) E(s) U(s) Y(s)
1 Gc(s) Gp(s)
Gc ( s)  K P  K D s, G p ( s )  2 , H (s)  1
s
H(s)
L(s) = Gc(s)Gp(s)H(s)
1 s2
For n=KP , KD = 2n , and n=1 red/sec, S GT p ( s )  
1  L ( s ) ( s  1) 2
10
G ( s )G p ( s ) 2( s  0.5)
C (s)  c  5 20 dB/dec
1  L( s ) ( s  1) 2 20 log|C(j)| 0 dB/dec
0
0 dB/dec
-5 Actual
Thus, if we specify n=1
Magnitude(dB)

properly to take into -10

Asymptotes -20 dB/dec
consideration the extent of -15

modeling error and the -20 20 log|S(j)|

frequency of external -25
40 dB/dec
disturbance, we can expect the -30

system to have an acceptable -35

amount of robustness.  = 0.5  = 1 rad/s

-40
-2 -1 0 1 2
10 10 10 10 10
Frequency (rad/sec)

10
 Sensitivity of a Controlled System
Example 5.4: Consider the following P controlled system where the blocks
transfer functions are
( s  1) R(s) E(s) U(s) Y(s)
Gc ( s)  K P , G p (s )  2 ,
H (s )  1 Gc(s) Gp(s)
s
( 1)
H(s)
( s  1)
L( s)  Gc ( s)G p (s ) H ( s )  K P
( s  1) 2 G ( s )G p ( s ) K p ( s  1)
The closed transfer function, T(s), is T ( s )  c  2
1  L( s ) s  ( 2  K p ) s  (1  K p )

The system is stable for a gain 0 < Kp < 1.

The steady-state error due to a negative j
unit step input R(s) =1/s is
Kp=0 Kp=1 Kp  
1  2K P -1 0 
ess  -0.5 1
1 KP s-plane
and ess = 0 when Kp = 1/2. Root locus of the system

Sensitivity of a Controlled System

Example 5.4: R(s) E(s) U(s) Y(s)
( s  1) Gc(s) Gp(s)
Gc ( s)  K P , G p (s )  , H (s )  1
( s  1) 2 H(s)
L(s )  Gc ( s)G p ( s) H ( s) Step Response
1.2

( s  1) r(t)
 KP 1
( s  1) 2
0.8
-y(t)
G c ( s )G p ( s )
T ( s) 
Amplitude

0.6
1  L (s )
K p ( s  1) 0.4

s 2  ( 2  K p ) s  (1  K p )
0.2

1  2KP
ess  0
1  KP
-0.2
5 10 15 20 25 30
Time (sec)
The time response of the system output, -y(t) is depicted in the figure for
K=0.5. Note the initial undershoot at t = 1 s.

11
 Sensitivity of a Controlled System
Example 5.4: R(s) E(s) U(s) Y(s)
( s  1) Gc(s) Gp(s)
Gc ( s)  K P , G p (s )  , H (s )  1
( s  1) 2 H(s)

Step Response
The time response of the system 2

output, -y(t) shows that this K=0.65, ess=0.86

system is sensitive to changes in 1.5
K=0.55, ess=0.22
K. The performance of this -y(t) r(t) K=0.5, ess=0
system might be considered 1

Amplitude
K=0.45, ess=0.18
barely acceptable for a change of
gain of only ±10%. Thus, this 0.5 K=0.25, ess=0.67
system would not be considered
robust. The steady-state error of 0

this system changes greatly as K

changes. -0.5
5 10 15 20 25 30
Time (sec)

12
 L !"#$ %&'% !"#$%&'&()*(+),-&."/&&

()*+,-.-%/0%1/2#-") --
0)"&'1/2 .3/ 4$)&/15$))6 &%&./7 &3)8" '" 9':-2/; <%&./7 :)#$& '"4$-1/
7#'".#'"'": # &7#$$ .2#4='": /22)2> ?!@ A "?!@ 5 #?!@> *)2 #" '"6-. "?!@ #"1
=//6'": .3/ )-.6-. #?!@ &7#$$ *)2 # 1'&.-2,#"4/ $?!@;

B"62/1'4./1(1'&.-2,#"4/('"6-.>($?!@(
"?!@ ?!@ 5?!@ #?!@
%4?!@ %&?!@
$0!:40; 78+9)(
23*+43((04 '()*+,-./!+01 39+&9+
:*&9+
6?!@
<0)!94010*+-39+&9+
.0*!34 </"&)2(")'&/>(=?!@
C3/($))6(.2#"&*/2(*-"4.')"(>?!@('&(>?!@(A(%4?!@%6?!@6?!@;

C3/(&/"&'.'D'.%(*-"4.')"(.?!@('&E( . ?! @ !
>? ! @
>? ! @
C3/(4)76$/7/".#2%(&/"&'.'D'.%(*-"4.')"(2?!@('&;>( 2 ?!@ !
>? ! @
C3/(4$)&/15$))6(&%&./7(&3)8"('"(9':-2/(#$&)(3#&(.3/(2/$#.')"&3'6E . ?!@ 2?!@ !
 ()*+,-.-%/0%1/2#-") --
B"62/1'4./1(1'&.-2,#"4/('"6-.>($?!@(
"?!@ ?!@ 5?!@ #?!@
%4?!@ %&?!@
$0!:40; 78+9)(
23*+43((04 '()*+,-./!+01 39+&9+
:*&9+
<0)!94010*+-39+&9+ 6?!@
.0*!34 </"&)2(")'&/>(=?!@

> ? !@
>?!@(A(%4?!@%6?!@6?!@; . ?! @ ! 2 ?!@ !
>? ! @ >? ! @

C3/(2/$#.')"&3'6(,/.8//"(.?!@(#"1(2?!@('& . ? !@ 2 ? ! @ !

9)2 63%&'4#$$% 2/#$'G#,$/ &%&./7&> .3/ $))6 :#'" >?!@ A %4?!@%6?!@6?!@ 7-&. ,/
&7#$$ *)2 3':3 *2/H-/"4'/&; C3'& 7/#"& .3#. .??"@ #662)#43/& #. 3':3
*2/H-/"4'/&;

34 %5 $"#$2*"./)-%/0%5$/! --% !6"7

34 5 $"#$2*"./)- /0 "4 5$/! -- 6-,-" ;7 !6"7' C3/2/ #2/ .8) ='"1& )*
6/2.-2,#.')" 7#% )44-2E
$?!@(
"?!@ ?&@ 5?!@ #?!@
%4?!@ %&?!@

6?!@
=?!@
#7 (88.".9 6/2.-2,#.')"&E%)?!@ A %&?!@ I #%&?!@

#%&?!@

%6?!@ % ) ? ! @ ! % & ? ! @ #% & ? ! @

%)?!@(
$7 :#+".5+.!*".9 6/2.-2,#.')"&E %1?!@ A %&?!@J I #%&?!@K

#%&?!@
% 1 ? ! @ ! % & ? ! @J #% & ? ! @K
%6?!@
%1?!@(

F
 34 %*88.".9 %5 $"#$2*"./)-%/0%5$/! --% !6"7
!" #11'.'D/ 6/2.-2,#.')" 43#2#4./2'G/& .3/ &/. )* 6)&&',$/ 62)4/&&/& #& *)$$)8&
?3/2/ 8/ #&&-7/ .3#. %4?!@A6?!@A @E
% ) ? ! @ ! % & ? ! @ #% & ? ! @

#%&?!@
"?!@ #?!@
%6?!@ %) ? ! @ ! % & ? ! @ #% & ? ! @
%)?!@(

83/2/ %&?!@ '& .3/ ")7'"#$ 62)4/&&> #"1 #%&?!@ '& .3/ 6/2.-2,#.')" .3#. '& ,)-"1/1
'" 7#:"'.-1/; M/ #&&-7/ .3#. %)?!@ #"1 %&?!@ 3#D/ .3/ &#7/ "-7,/2 )* 6)$/& '"
.3/ 2':3. 3#"1> !56$#"/ ?'* #"%@; C3/" .3/ &%&./7 &.#,'$'.% 8'$$ "). 43#":/ '*
N #% & ? ?" @ N$N % & ? ?" @ N *)2(#$$(";

C3'&(#&&-2/&(&.#,'$'.%()*(*88.".9 %5 $"#$2*"./)- ,-.(").(1%"#7'4(6/2*)27#"4/;

34 %;#+".5+.!*".9 %5 $"#$2*"./)-%/0%5$/! --% !6"7

:#+".5+.!*".9 5 $"#$2*"./)-E %1?!@ A %&?!@J % #%&?!@K

#%&?!@
"?!@ #?!@
%6?!@ % 1 ? ! @ ! % & ? ! @J #% & ? ! @K
%1?!@(

O/2/ #:#'" 8/ #&&-7/ .3#. %4?!@A6?!@A

C3/ 6/2.-2,#.')" '& ,)-"1/1 '" 7#:"'.-1/> #"1 '. '& #:#'" #&&-7/1 .3#. %1?!@ #"1
%6?!@ 3#D/ .3/ &#7/ "-7,/2 )* 6)$/& '" .3/ 2':3.53#"1 &56$#"/; C3/" .3/ &%&./7
&.#,'$'.% 8'$$ "). 43#":/ '*

N #% & ? ?" @ N$N N *)2(#$$(";

% & ? ?" @
!,)D/ /H-#.')" '& 4#$$/1 .3/ $/2#-" -"*2.+.", !$." $./) *)2 ;#+".5+.!*".9
5 $"#$2*"./)-<

L
 34 %;#+".5+.!*".9 %5 $"#$2*"./)-%/0%5$/! --% !6"7
:#+".5+.!*".9 5 $"#$2*"./)-E %1?!@ A %&?!@J % #%&?!@K

#%&?!@
%1 ? !@ ! % & ? ! @J #% & ? ! @K
%6?!@
%1?!@(

N #% & ? ? " @ N$N N *)2(#$$(";

% & ? ?" @

C3'& /H-#.')" '& # ./&. *)2 2),-&."/&& 8'.3 2/&6/4. .) # 7-$.'6$'4#.'D/

6/2.-2,#.')";
C3'& *)27 )* 6/2.-2,#.')" '& )*./" -&/1 ,/4#-&/ '. &#.'&*'/& .3/ '".-'.'D/
62)6/2.'/& )*E
? @ ,/'": &7#$$ #. $)8 *2/H-/"4'/&> 83/2/ .3/ ")7'"#$ 62)4/&& 7)1/$ '& -&-#$$%
8/$$ =")8"> #"1
?F@ ,/'": $#2:/ #. 3':3 *2/H-/"4'/&> 83/2/ .3/ ")7'"#$ 7)1/$ '& #$8#%& '"/Q#4.;

34 %;#+".5+.!*".9 %5 $"#$2*"./)-%/0%5$/! --% !6"7

=>*;5+ %&<?'%0)"&'1/2(.3/(*)$$)8'":(&%&./7(8'.3(%4?!@ @-A,-6?!@(A (#"1E

$?!@( #%&?!@
"?!@ ?&@ 5?!@ #?!@
%4?!@ %6?!@
%1?!@(
6?!@
SVVVV? ! V; @ SVVVV? ! V; @
% & ?! @ ! !
!? ! L@?! F V! VVVV@ !? ! L@? ! U ? TT;RS@? ! U & ? TT;RS@
C3/(43#2#4./2'&.'4(/H-#.')"()*(.3/(*//1,#4=(&%&./7('&
'? !@ ! %8 ? !@% & ? !@ 6 ? !@ ! A% & ? ! @ 6 ? !@ ! V
C3/($))6(.2#"&*/2(*-"4.')"('&

SVVVV? ! V; @
%8 ? !@% & ? !@ 6 ? !@ ! A% & ? !@ 6 ? !@ ! A
!? ! L@? ! U ?TT;RS@? ! U & ? TT;RS@

P
 34 %;#+".5+.!*".9 %5 $"#$2*"./)-%/0%5$/! --% !6"7

=>*;5+ %&<?'%0)"&'1/2(%4?!@ @-A,-6?!@(A (#"1E

$?!@( #%&?!@
"?!@ ?&@ 5?!@ #?!@
%4?!@ %6?!@
%1?!@(
6?!@
>? !@ ! %8 ? ! @% & ? ! @ 6 ? !@
AAV;UT ?"
SVVVV? ! V; @
!A AAV
!? ! L@?! U ? TT;RS@?! U & ? TT;RS@
()A AAV AAV
C3/((2)).($)4-&()*(.3/(&%&./7('& V '
BL

AAV

!B&()*0

34 %;#+".5+.!*".9 %5 $"#$2*"./)-%/0%5$/! --% !6"7

=>*;5+ %&<?'%C3/(&%&./7(8'.3(%4?!@ @-A,-6?!@(A (#"1E

SVVVV? ! V; @
>? !@ ! A
!? ! L@? ! U ? TT;RS@? ! U & ? TT;RS@
AAV;UT ?"
AAV

()A AAV AAV

BL V '

AAV

!B&()*0

U
 34 %;#+".5+.!*".9 %5 $"#$2*"./)-%/0%5$/! --% !6"7
=>*;5+ %&<@'%0)"&'1/2(.3/(*)$$)8'":(&%&./7(8'.3(%4?!@ @-A,-6?!@(A (#"1E
SVVVV? ! V; @
% & ?! @ !
!? ! L@?! F V! VVVV@
$?!@( #%&?!@
"?!@ ?&@ 5?!@ #?!@
%4?!@ %6?!@
%1?!@(
6?!@

X)8>(4)"&'1/2(.3/(/**/4.()*(#"(-"7)1/$$/1 6)$/(#.(" @UV(4);Y!;(Z"(.3'&(4#&/>($/.(

#&&-7/(.3/(7-$.'6$'4#.'D/(6/2.-2,#.')"('&(1/./27'"/1(*2)7

UV &!
#% & ? ! @ ! )2 #% & ? !@ !
! UV ! UV
& ?"
C3/(7#:"'.-1/(,)-"1('&(.3/" N #% & ? ?" @ N!N N
?" UV

34 %;#+".5+.!*".9 %5 $"#$2*"./)-%/0%5$/! --% !6"7

=>*;5+ &<@' C3/ &%&./7 8'.3 %4?!@ @ A, 6?!@ A #"1 .3/ 7-$.'6$'4#.'D/
6/2.-2,#.')"
SVVVV? ! V; @ #%&?!@
% & ?! @ !
!? ! L@?! F V! VVVV@
&! %6?!@
#% & ? !@ ! %1?!@(
! UV
C3/(7#:"'.-1/(,)-"1()* !!

%!

N N &!

% 8 ? ?" @ % & ? ? " @ 6 ? ?" @ '!

()*+,-./01/23

#! N N
A% & ? ?" @
& ?" !
#"1( N #% & ? ?" @ N!N N
?" UV "#!
N #% & ? ? " @ N
"'!

"&!
" A(UV(4);Y!
#2/(1/6'4./1('"(9':-2/(*)2(AAV;U "%!
"# " ! # $
! ! ! ! ! !
4506.0+78915)/:;073

W
test the stability of the system with w vary from 80 to 120 with step 5 ,
at k=0.2 , k=0.5 ,k=1 ?
Sol:
1) k=0.2
w

80
1 −80𝑗
1+
170000 ∗ (80𝑗 + 0.1) 80𝑗 +50
0.2 ∗
80𝑗 ∗ 80𝑗 + 3 ∗ ( 80𝑗 2 + 10 ∗ 80𝑗 + 10000) 0.8480
8.5586
85 7.0663 0.8619
90 5.2502 0.8742
95 3.2155 0.8849
100 1.9431 0.8944
105 3.8589 0.9029
110 7.2383 0.9104
115 11.2640 0.9171
120 15.8443 0.9231
The system is stable
‫‪2) k=0.5‬‬
‫‪w‬‬

‫‪80‬‬ ‫‪0.8480‬‬
‫‪1‬‬
‫‪1+‬‬
‫)‪170000 ∗ (80𝑗 + 0.1‬‬
‫∗ ‪0.5‬‬
‫)‪80𝑗 ∗ 80𝑗 + 3 ∗ ( 80𝑗 2 + 10 ∗ 80𝑗 + 10000‬‬
‫‪3.4356‬‬
‫‪85‬‬ ‫‪2.8148‬‬ ‫‪0.8619‬‬
‫‪90‬‬ ‫‪2.0456‬‬ ‫‪0.8742‬‬
‫‪95‬‬ ‫‪1.1225‬‬ ‫‪0.8849‬‬
‫‪100‬‬ ‫‪0.1798‬‬ ‫‪0.8944‬‬
‫‪105‬‬ ‫‪1.2744‬‬ ‫‪0.9029‬‬
‫‪110‬‬ ‫‪2.7255‬‬ ‫‪0.9104‬‬
‫‪115‬‬ ‫‪4.3751‬‬ ‫‪0.9171‬‬
‫‪120‬‬ ‫‪6.2284‬‬ ‫‪0.9231‬‬
‫‪The system is may not be stable‬‬
‫مالحظت‪ :‬في المحاضرة من خالل الرسم هذه الحالت )‪ (k=0.5‬تعتبر مستقرة لكن من خالل الجدول النظام غير‬
‫مستقر لذلك يتم اعتماد الجدول ويعتبر النظام غير مستقر‬

‫‪3) k=1‬‬
‫‪w‬‬

‫‪80‬‬ ‫‪1.8401‬‬ ‫‪0.8480‬‬

‫‪85‬‬ ‫‪1.5340‬‬ ‫‪0.8619‬‬
‫‪90‬‬ ‫‪1.1628‬‬ ‫‪0.8742‬‬
‫‪95‬‬ ‫‪0.7421‬‬ ‫‪0.8849‬‬
‫‪100‬‬ ‫‪0.4121‬‬ ‫‪0.8944‬‬
‫‪105‬‬ ‫‪0.7000‬‬ ‫‪0.9029‬‬
‫‪110‬‬ ‫‪1.3637‬‬ ‫‪0.9104‬‬
‫‪115‬‬ ‫‪2.1685‬‬ ‫‪0.9171‬‬
‫‪120‬‬ ‫‪3.0865‬‬ ‫‪0.9231‬‬

‫‪the system is may not be stable‬‬

 34 %;#+".5+.!*".9 %5 $"#$2*"./)-%/0%5$/! --% !6"7

A/; B/$C %' C3/ &%&./7 8'.3 %4?!@ @ AAV;U, 6?!@ A #"1 .3/ 7-$.'6$'4#.'D/
6/2.-2,#.')"
$?!@( #%&?!@
"?!@ ?&@ 5?!@ #?!@
%4?!@ %6?!@
%1?!@(
6?!@
&!
#% & ? !@ ! SVVVV? ! V; @
! UV % & ?! @ !
!? ! L@?! F V! VVVV@
C3/(7#:"'.-1/(,)-"1()*

N N & ?"
%8 ? ?" @% & ? ?" @ 6 ? ?" @ #"1( N #% & ? ?" @ N!N N
?" UV
%<.-1%(.3/(&.#,'$'.%()*(.3/(")7'"#$(#"1(6/2.-2,/1(&%&./7(8'.3([)1/(1'#:2#7;

34 %;#+".5+.!*".9 %5 $"#$2*"./)-%/0%5$/! --% !6"7

=>*;5+ &<D' C3/ &%&./7 8'.3 %4?!@ @ % E F<G, 6?!@ A #"1 .3/ 7-$.'6$'4#.'D/
6/2.-2,#.')"
% & ?!@ !
SVVVV? ! V; @ &! #%&?!@
#% & ? ! @ !
!? ! L@?! F V ! VVVV@ ! UV
%6?!@
N N
%1?!@(
% 8 ? ?" @% & ? ?" @ 6 ? ? " @
#"1(
!!

& ?" %!
N #% & ? ?" @ N!N N
?" UV &!

'!
()*+,-./01/23

N #% & ? ?" @ N$N N #! N N

A% & ? ?" @ 6 ? ?" @ *)2(#$$(";

A% & ? ?" @
!

"#!

"'! N #% & ? ?" @ N

C3/ $/2#-" -"*2.+.", !$." $./) *)2
"&!
7-$.'6$'4#.'D/ 6/2.-2,#.')"& '& #$7)&. " A(UV(4);Y!
"%!
&#.'&*'/1 !
"#
!
"
!
!
!
#
! !
$

4506.0+78915)/:;073

S
 34 %;#+".5+.!*".9 %5 $"#$2*"./)-%/0%5$/! --% !6"7

=>*;5+ &<D' C3/ &%&./7 8'.3 %4?!@ @ %E?, 6?!@ A #"1 .3/ 7-$.'6$'4#.'D/
6/2.-2,#.')"
% & ?!@ !
SVVVV? ! V; @ &! #%&?!@
#% & ? ! @ !
!? ! L@?! F V ! VVVV@ ! UV
%6?!@
N N
%1?!@(
% 8 ? ? " @% & ? ? " @ 6 ? ? " @
#"1(
%!
& ?"
N #% & ? ?" @ N!N N
?" UV &!

'!

()*+,-./01/23
#! N N
#2/(1/6'4./1('"(9':-2/(*)2(AA (83/2/('.('&( A% & ? ?" @
!
&//"(.3#.(.3/(42'./2')"()*(\H-#.')"
"#!

N #% & ? ? " @ N
N # % & ? ?" @ N$N
"'!
N *)2(#$$(";
A% & ? ?" @ 6 ? ?" @ "&!
" A(UV(4);Y!
"%!
"# " ! # $
! ! ! ! ! !
'&(").(&#.'&*'/1;(C3-&>(.3/(&%&./7(7#%(").(,/(&.#,$/; 4506.0+78915)/:;073

34 %;#+".5+.!*".9 %5 $"#$2*"./)-%/0%5$/! --% !6"7

A/; B/$C %' C3/ &%&./7 8'.3 %4?!@ @ AA , 6?!@ A #"1 .3/ 7-$.'6$'4#.'D/
6/2.-2,#.')"
$?!@( #%&?!@
"?!@ ?&@ 5?!@ #?!@
%4?!@ %6?!@
%1?!@(
6?!@
&!
#% & ? !@ ! SVVVV? ! V; @
! UV % & ?! @ !
!? ! L@?! F V! VVVV@
C3/(7#:"'.-1/(,)-"1()*

N N & ?"
%8 ? ?" @% & ? ?" @ 6 ? ?" @ #"1( N #% & ? ?" @ N!N N
?" UV

<.-1%(.3/(&.#,'$'.%()*("4 %)/;.)*+%-,-" ;%#"1 5 $"#$2 8%-,-" ; 8'.3([)1/(

1'#:2#7;(

R
 34 %;#+".5+.!*".9 %5 $"#$2*"./)-%/0%5$/! --% !6"7

=>*;5+ &<H' C3/ &%&./7 8'.3 .3/ 7-$.'6$'4#.'D/ 6/2.-2,#.')"> 6?!@ A #"1 # $#:
4)76/"&#.)2 %4?!@E
#%&?!@
%& ?! @ !
SVVVV? ! V; @ &!
! ? ! L@?! F V! VVVV@ #% & ?! @ !
! UV %6?!@
%1?!@(
V; U? ! FU@
%8 ? !@ !
? ! F;U@
I/+#"./) &<H' <.-1% .3/ $/2#-") -- #"1 !!

%!
-"*2.+.", )* .3/ 4$)&/1 $))6 &%&./7;
&!

C3/(7#:"'.-1/(,)-"1()* '!

()*+,-./01/23
N N
#! %8 ? ?" @% & ? ?" @
N N #"1( !
% 8 ? ? " @% & ? ? " @ 6 ? ? " @ "#!
& ?" "'! N #% & ? ? " @ N
N #% & ? ?" @ N!N N
?" UV "&!
" A(UV(4);Y!
"%!
#2/(1/6'4./1('"(9':-2/; !
"# "
! !
!
!
#
!
$
!
4506.0+78915)/:;073

34 %;#+".5+.!*".9 %5 $"#$2*"./)-%/0%5$/! --% !6"7

I/+#"./) &<H' C3/ &%&./7 8'.3 .3/
7-$.'6$'4#.'D/ 6/2.-2,#.')"> 6?!@ A #"1 #%&?!@
# $#: 4)76/"&#.)2 %4?!@E
%6?!@
V; U? ! FU@ %1?!@(
%8 ? !@ !
? ! F;U@
!!
SVVVV?! V; @
% & ? !@ ! %!
!? ! L@? ! F V! VVVV@ &!

&! '!
()*+,-./01/23

#% & ? ! @ ! #!
N
%8 ? ?" @% & ? ?" @
N
! UV
!

!& &//" *2)7 .3/ 9':-2/ .3/ *-"4.')" "#!

%4??"@%6?? "@6?? "@ 2/&3#6/1 '" .3/ "'! N #% & ? ? " @ N

*2/H-/"4% 2#":/ F ] " ] FU .) #$./2/1 "&!
" A(UV(4);Y!
"%!
7#:"'.-1/ !
"#
!
"
!
!
!
#
!
$
!
4506.0+78915)/:;073

T
 34 %;#+".5+.!*".9 %5 $"#$2*"./)-%/0%5$/! --% !6"7
!!
I/+#"./) &<H'
%!

&!

'!
()*+,-./01/23
N N
#! % 8 ? ?" @% & ? ?" @
!

"#!

"'! N # % & ? ?" @ N

"&!
" A(UV(4);Y!
"%!
! "# ! " ! ! ! ! # ! $
4506.0+78915)/:;073

!& &//" *2)7 .3/ 9':-2/ .3/ *-"4.')" %4??"@%6?? "@6?? "@ 2/&3#6/1 '" .3/ *2/H-/"4%
2#":/ F ] " ] FU .) #$./2/1 7#:"'.-1/
N #% & ? ?" @ N$N
% 8 ? ?" @ % & ? ?" @ 6 ? ?" @
N *)2(#$$(";
O/2/(.3/($/2#-") -- '"/H-#$'.%(.-%-*".-0. 8>(#"1(.3/(&%&./7(2/7#'"&(-"*2+ ;

V
 34 %/2J !".9 %/0%0 82*!C%!/)"$/+%-,-" ;-

$?!@(
"?!@ ?&@ 5?!@ #?!@
!?6"7 %&?!@

6?!@
=?!@

C3/ 4)".2)$ ),^/4.'D/ '& .) 1/&':" # 4)76/"&#.)2 %4 ?!@ &) .3#.E

"4 "$*)-. )" -5 !.0.!*"./)K
"4 -" *8,L-"*" -5 !.0.!*"./)K #"1
"4 0$ M# )!,L8/;*.) -5 !.0.!*"./)
#2/ #43'/D/1 #"1 "4 !/-" /0 0 82*!C ; *-#$ 8 2, "4 2*)8B.8"4 /0 "4
!/;5 )-*"/$ !?6&"7 '& &-**'4'/".$% &7#$$;

34 %/2J !".9 %/0%0 82*!C%!/)"$/+%-,-" ;-

$?!@(
"?!@ ?&@ 5?!@ #?!@
!?6"7 %&?!@

6?!@
=?!@
C3/ 4)76/"&#.)2 %4 ?!@ '& 1/&':"/1 &-43 .3#. "4 !/-" /0 0 82*!C ; *-#$ 8 2, "4
2*)8B.8"4 /0 "4 !/;5 )-*"/$ !?6&"7 '& &-**'4'/".$% &7#$$;
C3'& ,#"18'1.3 4)"&.2#'". '& "//1/1 7#'"$% ,/4#-&/ )* ")'&/ .3#. '& '"/D'.#,$/ '"
7/#&-2'": .3/ &%&./7 )-.6-.; ! $#2:/ ")'&/ #76$'*'4#.')" 4#" &#.-2#./ /'.3/2 .3/ $#../2
&.#:/& )* %4?!@ )2 .3/ /#2$% 62)4/&& &.#:/&;
9)2 '"&.#". # 62/5*'$./2> %6*?!@ 4#" ,/ #11/1 '" # .8)51/:2//5)*5*2//1)7 4)"*':-2#.')"
.) 3/$6 #43'/D/ .3/ 1/&':" :)#$&;
$?!@(
"?!@ ?&@ 5?!@ #?!@
506"7 !?6"7 %&?!@

6?!@
=?!@
 I,-" ;-%B."4%N)! $"*.)%O*$*; " $-
`#"% 4)".2)$ &%&./7& 3#D/ &/D/2#$ 6#2#7/./2& .3#. #2/ 4)"&.#".& ,-. -"4/2.#'"
8'.3'" # 2#":/;
$?!@(
"?!@ ?&@ 5?!@ #?!@
%4?!@ %6?!@

6?!@

9)2 /Q#76$/> 4)"&'1/2 # &%&./7 8'.3 .3/ *)$$)8'": 43#2#4./2'&.'4 /H-#.')"E

'? !@ ! %8 ? !@% & ? !@ 6 ? ! @ ! V

! !* )* & ! * & )* &F ! *& F ;;; )V ! V
83/2/ )" A 8'.3 =")8" 4)/**'4'/".& 8'.3'" ,)-"1&

+ : , ): , * : #"1 : A V> > F >_> *5 ;

I,-" ;-%B."4%N)! $"*.)%O*$*; " $-

$?!@(
"?!@ ?&@ 5?!@ #?!@
%4?!@ %6?!@

C3/ 43#2#4./2'&.'4 /H-#.')"; 6?!@

'? !@ ! ! * )* & ! *& )* &F ! * &F ;;; )V ! V

C3/ =")8" 4)/**'4'/".& 8'.3'" ,)-"1&
+ : , ): , * : #"1 :AV> >F>_>*5 ;

C) #&4/2.#'" .3/ &.#,'$'.% )* .3/ &%&./7> 8/ 7':3. 3#D/ .) '"D/&.':#./ #$$ 6)&&',$/
4)7,'"#.')"& )* 6#2#7/./2&;a

F
 I,-" ;-%B."4%N)! $"*.)%O*$*; " $-
$?!@(
"?!@ ?&@ 5?!@ #?!@
%4?!@ %6?!@
C3/ 43#2#4./2'&.'4 /H-#.')";
6?!@

'? !@ ! ! * )* & ! *& )* &F ! * &F ;;; )V ! V

C3/ =")8" 4)/**'4'/".& 8'.3'" ,)-"1&
+ : , ): , * : #"1 :AV> >F>_>*5 ;

9)2.-"#./$%> '. '& 6)&&',$/ .) '"D/&.':#./ # $'7'./1 "-7,/2 )* B/$-"L!*-

5/+,)/;.*+- '"&./#1 )* '"D/&.':#.'": #$$ 6)&&',$/ 4)7,'"#.')"& )* 6#2#7/./2&;
<// .3/ *)$$)8'": 2/*/2/"4/E
M; b; c2#".3#7 #"1 C d; e'"4/".> <3;04* 23*+43( ./!+01! 7*)(/!:! )*; $0!:C*,
D3E* F:(0/ G .3*!,X/8 f)2=> TTL;

I,-" ;-%B."4%N)! $"*.)%O*$*; " $-

=>*;5+ &<G' 0)"&'1/2 # .3'215)21/2 &%&./7;
$?!@(
# ?!@ %8 ? ! @% & ? ! @ "?!@ ?&@ 5?!@ #?!@
! %4?!@ %6?!@
"?!@ %8 ? !@% & ? !@ 6 ? !@
8'.3 .3/ *)$$)8'": 43#2#4./2'&.'4 /H-#.')"; 6?!@

'? ! @ ! %8 ? !@% & ? ! @ 6 ? ! @

' ? ! @ ! ! L )F ! F ) ! )V ! V
#"1 -"4/2.#'" 4)/**'4'/".& &-43 .3#.E
R , )V , WV - + V ! R> * V ! WV
F , ) , VV - + ! F> * ! VV
S , )F , FU - + F ! S> * F ! FU
<.-1% .3/ &.#,'$'.% )* .3/ 4$)&/1 $))6 &%&./7;

L
 I,-" ;-%B."4%N)! $"*.)%O*$*; " $-
I/+#"./) &<G' C3/ 43#2#4./2'&.'4 /H-#.')";
$?!@(
'?!@ ! ! * )*& ! *& )*&F ! *&F ;;; )V ! V "?!@ ?&@ 5?!@ #?!@
%4?!@ %6?!@
C3/ =")8" 4)/**'4'/".& 8'.3'" ,)-"1&
6?!@
+ : , ): , *: #"1 :AV> >F>_>*5 ;
34 B/$-"L!*- 5/+,)/;.*+-g C3/ #"#$%&'& )* /)+, 0/#$ 5/+,)/;.*+- '& &-**'4'/".>
#"1 .3/% #2/ 2/#1'$% 1/*'"/1 *)2 * "4.$8L/$8 $ -,-" ; 8'.3 # 43#2#4./2'&.'4
/H-#.')"
'?!@ ! !L )F ! F ) ! )V ! V 1.H ? !@ ! ! + F ! * ! *V
L F

. HF ? ! @ ! ! * F ! + ! + V
L F

C3/"(.3/(*)-2(6)$%")7'#$&(#2/ 0
.HL ? !@ ! ! * F ! * ! + V
L F

.
/ HP ? ! @ ! ! + F ! + ! * V
L F

h"/ )* .3/ *)-2 6)$%")7'#$& 2/62/&/".& .3/ 8)2&. 4#&/ #"1 7#% '"1'4#./ /'.3/2
-"&.#,$/ 6/2*)27#"4/ )2 #. $/#&. .3/ 8)2&. 6/2*)27#"4/ *)2 .3/ &%&./7 '" .3#. 4#&/;

I,-" ;-%B."4%N)! $"*.)%O*$*; " $-

I/+#"./) &<G' C3/ 43#2#4./2'&.'4 /H-#.')" '&E
$?!@(
'? ! @ ! ! L )F ! F ) ! )V ! V "?!@ ?&@ 5?!@ #?!@
%4?!@ %6?!@
8'.3 .3/ -"4/2.#'" 4)/**'4'/".& )*E
R , )V , WV - + V ! R> * V ! WV 6?!@
F , ) , VV - + ! F> * ! VV
S , )F , FU - + F ! S> * F ! FU

C3/"(.3/(*)-2(6)$%")7'#$&(#2/
1H ? ! @ ! ! L + F ! F * ! *V ! ! L S ! F VV! WV
.
. HF ? ! @ ! !
L
*F! F
+ ! +V ! ! L
FU! F
F! R
0
.HL ? !@ ! !
L
*F! F
* ! +V ! ! L
FU! F
VV! R
.H ? ! @ ! ! L
+ F ! + ! * V ! ! S!
F
F! WV L F
/ P
M/ .3/" 62)4//1 .) 43/4= .3/&/ *)-2 6)$%")7'#$& ,% 7/#"& )* .3/ +)-.35O-28'.G
42'./2')";

P
 I,-" ;-%B."4%N)! $"*.)%O*$*; " $-
I/+#"./) &<G' +)-.35O-28'.G 42'./2')" *)2 *)-2 6)$%")7'#$& #2/E
H ?!@ ! ! L S! F VV! WV HF ? !@ ! ! L FU! F F! R HL ? !@ ! ! L FU! F VV! R
L L
! VV !L F ! VV
F F
! S WV ! F
FU R ! FU R
! T V ! ;WR V ! TT;WR V
! WVV
!V R !V R

HP ? ! @ ! ! L
S! F
F! WV $?!@(
"?!@ ?&@ 5?!@ #?!@
L %4?!@
! F %6?!@
!F S R
! V;UR V 6?!@
!V R

+/&-$.&g "4 -,-" ; .- -"*2+ *)2 #$$ .3/ 2#":/ )* -"4/2.#'" 6#2#7/./2&;

I,-" ;-%B."4%N)! $"*.)%O*$*; " $-

=>*;5+ &<&' 0)"&'1/2 .3/ *//1,#4= &%&./7 8'.3 .3/ 62)4/&& .2#"&*/2 *-"4.')">
?-"1/2 ")7'"#$ 4)"1'.')"&@ $?!@
P;U "?!@ ?&@ 5?!@ #?!@
% & ?! @ ! %4?!@ %6?!@
! ? ! @?! F@
83/2/(%4?!@(A 6?!@A ;( 6?!@
<.-1% .3/ &.#,'$'.% )* .3/ 4$)&/1 $))6 ")7'"#$ &%&./7;
I/+#"./) &<&' C3/ ")7'"#$ 43#2#4./2'&.'4 /H-#.')" '& .3/"
# ?!@
!
%8 ? ! @% & ? ! @ '? !@ ! %8 ? !@% & ? !@ 6 ?!@ ! V
"?!@ %8 ? ! @% & ? ! @ 6 ? ! @ L F
!! L! F! P;U ! V

B&'": .3/ +)-.35O-28'.G 42'./2')"> !L F

F
! L P ;U
! V;U V
! V P ;U

U
 I,-" ;-%B."4%N)! $"*.)%O*$*; " $-
I/+#"./) &<&' C3/ ")7'"#$ 43#2#4./2'&.'4 /H-#.')" '& .3/"
$?!@(
"?!@ ?&@ 5?!@ #?!@
'?!@ ! ! L L! F F! P;U ! V %4?!@ %6?!@

B&'": .3/ +)-.35O-28'.G 42'./2')"> 6?!@

L
! F
! F L P ;U
! V;U V
! V P ;U

C3'& &%&./7 '& ")7'"#$$% &.#,$/;

1 -#+"g .3/ ")7'"#$ 6#2. )* .3/ &%&./7 '& &.#,$/ .3-& .3/ &%&./7 '& ")7'"#$$%
&.#,$/;

I,-" ;-%B."4%N)! $"*.)%O*$*; " $-

=>*;5+ %&<P'%%4?!@(A 6?!@A ; $?!@
"?!@ ?&@ 5?!@ #?!@
P;U %4?!@ %6?!@
% & ? !@ !
! ? ! @? ! F@ 6?!@
# ?!@ %8 ? ! @% & ? ! @
!
"?!@ %8 ? ! @% & ? ! @ 6 ? ! @

'? ! @ ! %8 ? !@% & ? !@ 6 ? !@ ! ! L )F ! F ) ! )V ! V

8'.3 .3/ -"4/2.#'" 4)/**'4'/".& &-43 .3#.

P , )V , U - + V ! P> * V ! U
, ) , L-+ ! > * ! L
F , )F , P - + F ! F> * F ! P

<.-1% .3/ &.#,'$'.% )* .3/ 4$)&/1 $))6 &%&./7 8'.3 #,)D/ 6#2#7/./2& -"4/2.#'".'/&;

W
 I,-" ;-%B."4%N)! $"*.)%O*$*; " $-
I/+#"./)%&<P'%%4?!@(A 6?!@A ; $?!@
"?!@ ?&@ 5?!@ #?!@
P;U %4?!@ %6?!@
% & ? !@ !
! ? ! @? ! F@ 6?!@

# ?!@ %8 ? ! @% & ? ! @
! '? ! @ ! %8 ? !@% & ? !@ 6 ? !@
"?!@ %8 ? ! @% & ? !@ 6 ? !@
! ! L L! F F! P;U ! V
C3/ &%&./7 '& ")7'"#$$% &.#,$/;
O)8/D/2> '* .3/ &%&./7 3#& -"4/2.#'" 4)/**'4'/".& &-43 .3#.
P , )V , U - + V ! P> * V ! U
'? ! @ ! %8 ? !@% & ? !@ 6 ? !@ , ) , L-+ ! > * !L
'? ! @ ! ! L )F ! F ) ! )V ! V F , )F , P - + F ! F> * F ! P
1H ? ! @ ! ! L + F ! F * ! * V ! ! L F! F L! U
.
H ?! @ ! ! L * F ! F + ! + V ! ! L P! F ! P
.3/" 8/ 7-&. /Q#7'"/ .3/ *)-2 .0 F
6)$%")7'#$& *)2 .3/ &.#,'$'.% )* .3/ .HL ? ! @ ! !
L
* F ! F * ! + V ! ! L P ! F L! P
-"4/2.#'" &%&./7g .H ? ! @ ! ! L + F ! F + ! *V ! ! L F! F ! U
/ P

I,-" ;-%B."4%N)! $"*.)%O*$*; " $-

I/+#"./)%&<P'%C3/(43#2#4./2'&.'4(/H-#.')"()*(.3/(&%&./7('&
'? ! @ ! %8 ? !@% & ? !@ 6 ? !@
'? ! @ ! ! L
)F ! F ) ! )V ! V
)2 -"4/2.#'" 4)/**'4'/".& &-43 .3#. 1H ? ! @ ! ! L F! F L! U
P , )V , U - + V ! P> * V ! U .
C3/ *)-2 6)$%")7'#$& .HF ?! @ ! ! L P! F ! P
, ) , L-+ ! > * ! L 0
7-&. ,/ /Q#7'"/1g
.HL ? ! @ ! !
L F
F , )F , P - + F ! F> * F ! P P! L! P
.H ? ! @ ! ! L F! F
! U
/ P
B&'":(.3/(+)-.35O-28'.G(42'./2')">(H ?!@(#"1(HL?!@(#2/(&.#,$/(#"1(HF?!@(
HF ? ! @ ! ! L P! F ! P
1/#"4LA#$B."Q *$$*,' ! L
!F P P
! - V V
V
! P '&(7#2:'"#$$%(&.#,$/;(

S
 I,-" ;-%B."4%N)! $"*.)%O*$*; " $-
I/+#"./)%&<P'%C3/(43#2#4./2'&.'4(/H-#.')"()*(.3/(&%&./7('&
'? ! @ ! %8 ? !@% & ? !@ 6 ? !@
'? ! @ ! ! L )F ! F ) ! )V ! V
)2 -"4/2.#'" 4)/**'4'/".& &-43 .3#. 1H ? ! @ ! ! L F! F L! U
P , )V , U - + V ! P> * V ! U .
C3/ *)-2 6)$%")7'#$& .HF ?! @ ! ! L P! F ! P
, ) , L-+ ! > * ! L 0
7-&. ,/ /Q#7'"/1g
.HL ? ! @ ! !
L F
F , )F , P - + F ! F> * F ! P P! L! P
.H ? ! @ ! ! L F! F ! U
/ P
9)2(HP?!@(>(8/(3#D/
HP ? ! @ ! ! L F ! F ! U!V 1/#"4LA#$B."Q *$$*,' ! L
!F F U
! - &L Y F V
!V U
#"1(.3/(&%&./7('&(-"&.#,$/(*)2(.3'&(8)2&.(4#&/;

I,-" ;-%B."4%N)! $"*.)%O*$*; " $-

I/+#"./)%&<P'%C3/(43#2#4./2'&.'4(/H-#.')"(*)2(.3/(8)2&.(4#&/(HP?!@(&%&./7('&
HP ? ! @ ! ! L F ! F ! U!V
1/#"4LA#$B."Q *$$*,'
!L
C3/(&%&./7('&(-"&.#,$/(*)2(.3/(8)2&.(4#&/>(83/2/( !F F U
2+F A(F(?7'"'7-7@>(+ A (?7'"'7-7@>(#"1(*V A(U( ! - &L Y F V
?7#Q'7-7@; !V U

C3'&()44-2&(83/"(.3/(62)4/&&(3#&(43#":/1(.)
$?!@(
"?!@ ?&@ 5?!@ #?!@
U %4?!@
% & ?! @ ! %6?!@
! ?! @F
6?!@
83/2/(%4?!@(A 6?!@A ;(

R
 I,-" ;-%B."4%N)! $"*.)%O*$*; " $-
I/+#"./)%&<P' $?!@
"?!@ ?&@ 5?!@ #?!@
C3/(")7'"#$(&%&./7(8#&E %4?!@ %6?!@
P;U
% & ?! @ !
!? ! @?! F@ 6?!@

83/2/(%4?!@(A 6?!@A ;(

C3/(43#2#4./2'&.'4(/H-#.')"('&((
1! ! &F;RU
' ? ! @ ! ! L L! F F! P;U ! V - 0
/! F>L ! &V;VSP % ? ;FU
i)7'"#".(6)$/

I,-" ;-%B."4%N)! $"*.)%O*$*; " $-

I/+#"./)%&<P'
C3/(43#2#4./2'&.'4(/H-#.')"(*)2(.3/(8)2&.(4#&/(HP?!@()44-2&(83/"(.3/(62)4/&&(3#&
43#":/1(.)
$?!@
U "?!@ ?&@ 5?!@ #?!@
% & ?! @ ! F
%4?!@ %6?!@
! ?! @

83/2/(%4?!@(A 6?!@A ;( 6?!@

C3/(43#2#4./2'&.'4(/H-#.')"('&((
1! ! &F; U
'? !@ ! ! L F! F F! U ! V - 0
/!F >L ! V;VSU % ? ;UF
X)./(.3#.(.3/(.3'21(6)$/(3#&(7)D/1(.)8#21(.3/ ?"5#Q'&(.)('.&($'7'.(#.(&(A(5 (#"1(
.3#.(.3/(:#'"(3#&('"42/#&/1(.)('.&($'7'.(#.(A A(U;(
h*./">(8/(#2/(#,$/(.)(/Q#7'"/(.3/(.2#"&*/2(*-"4.')"(%6?!@(#"1(62/1'4.(.3/(8)2&.5
4#&/(4)"1'.')"&;

T
 L !"#$ %7%&% !"#$%#&'()$*+$,*-.&/$0*)/1*2$34&/#5&

'( %) *+,-%./%0.1#*"%2.-"$.3%45*" 6*

F!G$
"F!G %F&G (F!G #F!G
&7F!G &)F!G

$F!G
'F!G

!"# 6#&'() *+ 1*-.&/ 7*)/1*2 &4&/#5& '& -8&#6 *) /9* /8&:&;

7 6#/#15')')( /"# &/1.7/.1# *+ /"# 7*)/1*22#1 8)6
!7 86<.&/')( /"# 7*)/1*22#1=& >8185#/#1& /* ('?# 8) @*>/'582A &4&/#5 >#1+*158)7#B
!"'& 6#&'() >1*7#&& '& )*158224 6*)# 9'/" @8**#6 9 !.6:3 " ;-.<3 9, @ *+
/"# >1*7#&&B C.1/"#15*1#D /"# >1*7#&& '& )*158224 6#&71'-#6 -4 8 2')#81 /'5#E
')?81'8)/ 7*)/').*.& 5*6#2B !"# &/1.7/.1# *+ /"# 7*)/1*22#1 '& 7"*&#) &.7" /"8/ /"#
&4&/#5=& 1#&>*)&# 78) 5##/ 7#1/8') >#1+*158)7# 71'/#1'8B
 '( %) *+,-%./%0.1#*"%2.-"$.3%45*" 6*
F!G$
"F!G %F&G (F!G #F!G
&7F!G &)F!G

$F!G
'F!G
I)# >*&&'-2# .1= !"+> +- "( 9 *+,- ./ 8 !.-"$.3 *5*" 6 '& /"8/ /"# 7*)/1*22#6
&4&/#5=& *./>./ &"*.26 #J87/24 8)6 ')&/8)/8)#*.&24 1#>1*6.7# '/& ')>./D %F!G KB
!"8/ '&D /"# &4&/#5 /18)&+#1 +.)7/'*) &"*.26 -# .)'/4;

# F! G &, F! G& ) F ! G
+ F !G " "
" F! G ! * F !G

9"#1#$*F!G$L$&7F!G&>F!G$F!GB

'( %) *+,-%./%0.1#*"%2.-"$.3%45*" 6*
F!G$
"F!G %F&G (F!G #F!G
&7F!G &)F!G

$F!G
# F -# G 'F!G
+ F -# G "
" F -# G -5*+ 6'$%7$8 ,!#.
#$%&'()*+ ,*-.

&, F -# G& ) F -# G
!"

" "
! *F -# G !"
"
23$4+ ,*+%.

9"#1#$*F-#G$L$&7F-#G&>F-#G$F-#GB 1"

!0"
! " ! /
!" !" !" !"
97+:)+&;< ,7$*=4+;.

M) */"#1 9*16&D /"# &4&/#5 &"*.26 -# >1#&#)/8-2# *) 8 N*6# (8') ?#1&.&

+1#O.#)74 6'8(185 9'/" 8 . /0 (8') *+ ')+')'/# -8)69'6/" 8)6 P#1* >"8&# &"'+/B

H
 '( %) *+,-%./%0.1#*"%2.-"$.3%45*" 6*
F!G$
"F!G %F&G (F!G #F!G
&7F!G &)F!G

$F!G
'F!G
'( .$ "+!83 9 *+$ 9 .1= !"+> ; !* 58:# /"# &4&/#5 -# >1#&#)/8-2# *) 8 N*6#
(8') ?#1&.& +1#O.#)74 6'8(185 9'/" 8 . /0 (8') *+ ')+')'/# -8)69'6/" 8)6 P#1*
>"8&# &"'+/ *1
# F -# G &, F -# G& ) F -# G 9"#1#$*F-#G$L$&7F-#G&>F-#G$F-#GB
+ F -# G " "
"F -# G ! * F -# G
M) >187/'7#D /"'& .1= !"+> '& )*/ >*&&'-2#D &')7# #?#14 &4&/#5 9'22 7*)/8')
+-9#!"+> 8)6 !8:8!+"+> /4># 7*5>*)#)/& /"8/ *".$ - $,5 ') &*5# +*15B
!"#&# #2#5#)/& 8)6 /"#'1 ')/#17*))#7/'*)& 9'/" - $,5 9+**+:8"+> !.6:.- -"*
>1*6.7# /"# &4&/#5=& 64)85'7 1#&>*)&# 7"8187/#1'&/'7&B 3.7" &4&/#5& 1#>1*6.7#
&*5# ')>./& 825*&/ #J87/24D 9"'2# */"#1 ')>./& 81# )*/ 1#>1*6.7#6 8/ 822D
&'()'+4')( /"8/ "( *5*" 6 18-9<+9"( '& 3 ** "(8- +-/+-+" B

'( %) *+,-%./%0.1#*"%2.-"$.3%45*" 6*
'( .$ "+!83 9 *+$ 9 .1= !"+> ; !* 58:# /"# &4&/#5 "8& 8 KE/0 (8') *+ ')+')'/#
-8)69'6/" 8)6 P#1* >"8&# &"'+/ *1
F!G$
# F -# G "F!G %F&G (F!G #F!G
+ F -# G " &7F!G &)F!G
" F -# G
$F!G
&, F -# G& ) F -# G
" 'F!G
! *F -# G
9"#1#$*F-#G$L$&7F-#G&>F-#G$F-#GB

HK 2*( R + F -# G R" K /0 +*1 822 #B

'& )*/ >187/'78224 >*&&'-2#B

I)7# 9# 1#7*()'P# /"8/ /"# &4&/#5 64)85'7& 78))*/ -# '()*1#6D 9# )##6 8

)#9 6#&'() *-<#7/'?#B

Q
 '( %) *+,-%./%0.1#*"%2.-"$.3%45*" 6*
'( .$ "+!83 9 *+$ 9 .1= !"+> ; !* 58:# /"# &4&/#5 "8& 8 KE/0 (8') *+ ')+')'/#
-8)69'6/" 8)6 P#1* >"8&# &"'+/ *1 F!G$
& F ! G& F ! G "F!G %F&G (F!G #F!G
# F!G &7F!G &)F!G
+ F!G " " "
, )

"F!G ! *F ! G
9"#1#$*F!G$L$&7F!G&>F!G$F!GB $F!G
'F!G
HK 2*( R + F -# G R" K /0 +*1 822 # '& )*/ >187/'78224 >*&&'-2#B

! I)# :.**+13 9 *+,- .1= !"+> '& /* 58')/8') /"# 58()'/.6# 1#&>*)&# 7.1?#
8& +28/ 8)6 8& 72*&# /* .)'/4 +*1 8& 281(# 8 -8)69'6/" 8& >*&&'-2# +*1 8 ('?#) >28)/
8)6 7*)/1*22#1 7*5-')8/'*)B -5*+ 6'$%7$8 ,!#.
!"

#$%&'()*+ ,*-.
"

N8)69'6/"

!"
! " ! /
!" !" !" !"
97+:)+&;< ,7$*=4+;.

'( %) *+,-%./%0.1#*"%2.-"$.3%45*" 6*
-5*+ 6'$%7$8 ,!#.
!"
F!G$
(F!G
#$%&'()*+ ,*-.

"F!G %F&G #F!G

&7F!G &)F!G "
N8)69'6/"
$F!G
!"
'F!G ! " ! /
!" !" !" !"
97+:)+&;< ,7$*=4+;.

! I)# :.**+13 9 *+,- .1= !"+> '& /* 58')/8') /"# 58()'/.6# 1#&>*)&# 7.1?#
8& +28/ 8)6 8& 72*&# /* .)'/4 +*1 8& 281(# 8 -8)69'6/" 8& >*&&'-2# +*1 8 ('?#) >28)/
8)6 7*)/1*22#1 7*5-')8/'*)B
! T)*/"#1 '5>*1/8)/ (*82 *+ 8 7*)/1*2 &4&/#5 6#&'() '& /"8/ /"# #++#7/ *) /"#
*./>./ *+ /"# &4&/#5 6.# /* 6'&/.1-8)7#& '& 5')'5'P#6B !".&D 9# 9'&" /*
6+-+6+?
# F -# G & ) F -# G
" + F -# G " *?#1$8$18)(#$*+$+1#O.#)74$ # B
F -# G ! *F -# G

9"#1# * F -# G " &, F -# G& ) F -# G $ F -# G

S
 '( %) *+,-%./%0.1#*"%2.-"$.3%45*" 6*
0*)&'6#1 /"# 7*)/1*2 &4&/#5 &"*9) ') C'(.1#D 9"#1# &)F!GL& F!G&HF!G '& /"#
>28)/ 8)6 F!G '& /"# 6'&/.1-8)7#B
F!G$
"F!G %F&G (F!G #F!G
&7F!G & F!G &HF!G

$F!G
X#$/"#)$/"#$/18)&+#1$+.)7/'*)&$81#
# F!G &, F ! G& ) F ! G &, F ! G& F !G&H F ! G
+" F ! G " + F ! G " " "
"F ! G ! &, F ! G& ) F ! G $ F !G ! &, F ! G& F ! G&H F ! G $ F ! G
# F!G &H F ! G
8)6 + F!G " "
F!G ! &, F !G& F !G&H F !G $ F !G

V*/# /"8/ -*/" /"# 1#+#1#)7# 8)6 6'&/.1-8)7# /18)&+#1 +.)7/'*)& "8?# /"# &85#
6#)*5')8/*1W ') */"#1 9*16&D /"#4 "8?# /"# &85# 7"8187/#1'&/'7 #O.8/'*)D )85#24D

1F! G " ! &, F ! G& F !G&H F ! G $ F! G

'( %) *+,-%./%0.1#*"%2.-"$.3%45*" 6*
F!G$
"F!G %F&G (F!G #F!G
&7F!G & F!G &HF!G

$F!G

# F!G &, F ! G& F ! G&H F !G # F!G &H F !G

+ F!G " " 8)6 "
"F ! G ! &, F ! G& F ! G&H F ! G $ F ! G F!G ! &, F !G& F !G&H F !G $ F !G

1F! G " ! &, F ! G& F !G&H F ! G $ F! G

,#7822 /"8/ /"# &#)&'/'?'/4 *+ +F!G 9'/" 1#&>#7/ /* &)F!G '&D

2 &+ ) F ! G "
! &, F!G& F!G&H F !G $ F! G

8)6$/"#$7"8187/#1'&/'7$#O.8/'*)D$1F!GD$'&$/"#$')+2.#)7')($+87/*1$*)$/"#$&#)&'/'?'/4B

U
 '( %@.**+13 %) *+$ 9%A1= !"+>
F!G$
"F!G %F&G (F!G #F!G
&7F!G & F!G &HF!G

$F!G

2 &+ ) F ! G "
! &, F!G& F!G&H F !G $ F! G
!"# &#)&'/'?'/4 *+ +F!G 9'/" 1#&>#7/ /* &)F!G &"*9& /"8/ +*1 2*9 &#)&'/'?'/4 2F!GD
9# $ B#+$ 8 (+,( >83# *+ 2**> (8') *F!GD

*F! G " &, F ! G& F ! G&H F! G $ F ! G

-./ '/ '& :)*9) /"8/ 8 "'(" (8') 7*.26 78.&# +-*"81+3+"5 *1 :..$ $ *:.-*+> - **
*+ +F!GB

'( %@.**+13 %) *+$ 9%A1= !"+>

F!G$
"F!G %F&G (F!G #F!G
&7F!G & F!G &HF!G

$F!G

# F!G &, F ! G& F ! G&H F ! G

+" F !G " " 2&+ ) F !G "
"F!G ! &, F ! G& F ! G&H F ! G $ F !G ! &, F !G& F !G&H F !G $ F! G
*F ! G " &, F ! G& F ! G&H F ! G $ F ! G
!"# :.**+13 9 *+$ 9 .1= !"+> '& /* 6#&'() &7F!G &.7" /"8/ /* "8?#
CD +F!G 9'/" 9'6# -8)69'6/" 8)6 +8'/"+.2 1#>1*6.7/'*) *+ "F!GD
ED Z81(# 2**> (8') *F!G ') *16#1 /* 5')'5'P# &#)&'/'?'/4 2F!GD
FD Z81(# 2**> (8') *F!G 8//8')#6 >1'581'24 -4 &7F!G& F!G &')7#

# F !G
F!G ! &, F !G& F ! G

Y
 '( %@.**+13 %) *+$ 9%A1= !"+>
!* "8?# /"# +*22*9')( :.**+13 9 *+$ 9 .1= !"+> ;
CD +F!G 9'/" 9'6# -8)69'6/" 8)6 +8'/"+.2 1#>1*6.7/'*) *+ "F!GB
ED Z81(# 2**> (8') *F!G ') *16#1 /* 5')'5'P# &#)&'/'?'/4 2F!GB # F!G
FD Z81(# 2**> (8') *F!G 8//8')#6 >1'581'24 -4 &7F!G& F!G &')7# F !G ! &, F ! G & F ! G
F!G$
"F!G %F&G (F!G #F!G
&7F!G & F!G &HF!G

$F!G
4 ""+-, "( 9 *+,- ./ $.1#*" *5*" 6* +- /$ B# -!5G9.68+- " $6*D 9# 5.&/
+')6 8 :$.: $ !.6: -*8".$ "!H#7 &.7" /"8/ /"# 72*&#6E2**> &#)&'/'?'/4 '& 2#&&
/"8) &*5# /*2#18)7# ?82.#B N./ &#)&'/'?'/4 5')'5'P8/'*) ')?*2?#& +')6')( 8
>1*>#1 7*5>#)&8/*1 &.7" /"8/ /"# 72*&#6E2**> &#)&'/'?'/4 #O.82& *1 '& 81-'/181'24
72*&# /* /"# 5')'582 8//8')8-2# &#)&'/'?'/4B 3'5'28124D "( ,8+- 68$,+- :$.13 6
'& /* +')6 8 >1*>#1 7*5>#)&8/*1 /* 87"'#?# &*5# >1#&71'-#6 (8') 581(')B N./ (8')
581(') 58J'5'P8/'*) ')?*2?#& +')6')( 8 >1*>#1 7*5>#)&8/*1 /* 87"'#?# /"#
58J'582 8//8')8-2# (8') 581(')B

'( %) *+$ 9%@$.: $"+ *%/.$%I$ B# -!5%0 *:.-* %

C*1$/"#$+1#O.#)74$6*58')$&>#7'+'78/'*)&D$9#$1#O.'1#$/"#$+*22*9')($>1*>#1/'#&$
+*1$/"#$N*6#$6'8(185$*+$&7F-#G&>F-#GW

'( 9 *+$ 9 :$.: $"+ *;

B C*1 1#28/'?# &/8-'2'/4D &7F-#G&>F-#G 5.&/ "8?#D +*1 8) 86#O.8/# 18)(# *+ #D
)*/ 5*1# /"8) 8 EHK /0\/3,4/3 &2*># 8/ *1 )#81 /"# 71*&&*?#1 +1#O.#)74 #(7*+D
HB 3/#864 &/8/# 877.1874 87"'#?#6 -4 /"# 2*9 +1#O.#)74 (8')D
QB T77.1874 *?#1 8 -8)69'6/" #NXD -4 58')/8')')( R&7F-#G&>F-#GR 8-*?# 8
>1#&71'-#6 2#?#2B
SB %'&/.1-8)7# 1#<#7/'*) -4 8 "'(" (8') +*1 &7F-#G *?#1 /"# &4&/#5
-8)69'6/"D #NXB

[
 '( %) *+$ 9%@$.: $"+ *%/.$%I$ B# -!5%0 *:.-* %

HE7&%$;8C?84;<!?A7@?877/?)3@A7@=4<,3?
D,7==4</?A7667E;<8F
5.6789.R*F-#GR

HC7&%2G4B63?,@7!!7H3@?I?EHK$/0\/
HF7&%:;<;=>=? F84;<?4</?)C4!3?=4@8;<!F
)3@A7@=4<,3?B7></!

#(7*+ "7B>!G<3!!? #
B7></!
HJ7&%*7E?84;<!?G7?@3/>,3?!3<!;G;H;GJ?G7?
!3<!7@?<7;!3?4</?=7/36?><,3@G4;<GJ

'( %) *+$ 9%@$.: $"+ *%/.$%I$ B# -!5%0 *:.-* %

$;8C?84;<!?A7@?877/?)3@A7@=4<,3?
5.6789.R*F-#GR

D,7==4</?A7667E;<8F

:;<;=>=? 2G4B63?,@7!!7H3@?
)3@A7@=4<,3? D84;<?4</?)C4!3?=4@8;<!F
B7></! EHK$/0\/3,
#(7*+ #
"7B>!G<3!!?
B7></!
*7E?84;<!?G7?@3/>,3?!3<!;G;H;GJ?G7?
!3<!7@?<7;!3?4</?=7/36?><,3@G4;<GJ

'( %9 *+$ 9%.1= !"+> ;

B$C*1$1#28/'?#$&/8-'2'/4D$&7F-#G&>F-#G 5.&/$"8?#D$+*1$8)$86#O.8/#$18)(#$*+$ #D$)*/$
5*1#$/"8)$8$EHK$/0\/3,4/3 &2*>#$8/$*1$)#81$/"#$71*&&*?#1$+1#O.#)74$ #(7*+D
HB$3/#864$&/8/#$877.1874$87"'#?#6$-4$/"#$2*9$+1#O.#)74$(8')D
QB$T77.1874$*?#1$8$-8)69'6/"$#NXD$-4$58')/8')')($R&7F-#G&>F-#GR 8-*?#$8$
>1#&71'-#6$2#?#2B
SB$%'&/.1-8)7#$1#<#7/'*)$-4$8$"'("$(8')$+*1$&7F-#G *?#1$/"#$&4&/#5$-8)69'6/"D #NXB

]
 '( %) *+$ 9%@$.: $"+ *%/.$%I$ B# -!5%0 *:.-* %

$;8C?84;<!?A7@?877/?)3@A7@=4<,3?
5.6789.R*F-#GR
D,7==4</?A7667E;<8F

:;<;=>=? 2G4B63?,@7!!7H3@?
)3@A7@=4<,3? D84;<?4</?)C4!3?=4@8;<!F
B7></! EHK$/0\/3,
#(7*+ #
"7B>!G<3!!?
B7></!
*7E?84;<!?G7?@3/>,3?!3<!;G;H;GJ?G7?
!3<!7@?<7;!3?4</?=7/36?><,3@G4;<GJ

_&')( /"# 1**/ &#)&'/'?'/4 7*)7#>/D 9# 78) &/8/# /"8/ $$$ 6#*" 1 6+-+6+? 9
9"'2# 8//8')')( +F!G 9'/" 9.6+-8-" $.."* /"8/ 9'22 >1*?'6# /"# 8>>1*>1'8/#
1#&>*)&# 8)6 5')'5'P# /"# #++#7/ *+ F!GB
!"# (*82 '& /* "8?# /"# (8') *+ /"# 2**>D *F!GL"!H#7& F!G&HF!G$F!G >1'581'24
8//8')#6 -4 &7F!GK

'( %) *+$ 9%@$.: $"+ *%/.$%I$ B# -!5%0 *:.-* %

KL86:3 7DMC& Z#/ &, F!G " L D & F!G " D &H F !G " D $ F !G "
!F! ! G
F!G$
"F!G %F&G (F!G #F!G -#
L?& '

&7F!G & F!G &HF!G

Ka(a
$F!G
LLK LLKBHU D$$$( L$ LLK
!"# 7"8187/#1'&/'7 #O.8/'*) '&W %
M EKBU K
1F ! G " ! &, F ! G& F ! G&H F ! G $ F ! G " ! H ! ! ! L " K
')L

8)6 /"'& &4&/#5 "8& /9* 1**/&D 8)6 9# &#2#7/ !M)64<3

8 (8') L &* /"8/
# F!G &H F ! G
" '&$5')'5'P#6B
F!G ! &, F ! G& F ! G&H F ! G $ F ! G

2&+ ) F! G " *1$21` '&$5')'5'P#6$8)6$+F!GD

! &, F !G& F !G&H F !G $ F !G
# F!G &, F ! G& F ! G&H F ! G
+ F!G " " "8&$6#&'18-2#$6*5')8)/$1**/&D$!6B$
"F!G ! &, F ! G& F ! G&H F ! G $ F ! G

^
 '( %) *+$ 9%@$.: $"+ *%/.$%I$ B# -!5%0 *:.-* %
KL86:3 7DMC& F!G$
"F!G %F&G (F!G #F!G
&7F!G & F!G &HF!G

$F!G -#

L?& '
&, F ! G " L D & F ! G " D & H F ! G " D $ F!G " Ka(a
!F ! ! G
LLK LLKBHU D$$$( L$ LLK
!"#$1**/$2*7.&$*+$/"#$&4&/#5$'& M EKBU K %

')L
*@ L *! L !M)64<3
!"#$1**/$&#)&'/'?'/4$'& 2 L@ F !G " "
*L @ *L !"@ @

8)6$7"8187/#1'&/'7$#O.8/'*)$'& 1F!G " !F! ! G ! L " K , L " + !F! ! G

*L *! L + + !F ! ! G
" +F H! ! G X#$/"#)$*-/8') 2 L@ F !G " "
*! *L ! "@ @ H! ! ! ! "@

'( %) *+$ 9%@$.: $"+ *%/.$%I$ B# -!5%0 *:.-* %

KL86:3 7DMC& F!G$
"F!G %F&G (F!G #F!G
&7F!G & F!G &HF!G
!"#$7"8187/#1'&/'7$#O.8/'*)$'&
$F!G
L -#
1F!G " ! " !F! ! G ! L " K
L?& '

!F! ! G
Ka(a
!"#$1**/$&#)&'/'?'/4$'&
LLK LLKBHU D$$$( L$ LLK
*! L + + !F! ! G %
2 F! G "
@
L " M EKBU K
')L

*L ! "@ @ H! ! ! !" @

!M)64<3
X"#)$Ka(a D$/"#$1**/&$81#$7*5>2#J$8)6$/"#$&4&/#5$
#J"'-'/$&#7*)6$*16#1$-#"8?'*1B -#

# F!G &, F ! G& F ! G&H F ! G # <H

" #)
"F!G ! &, F ! G& F ! G&H F ! G $ F ! G ! ! H(# < ! ! #<H (
H
" 7*& - #/ " #< +( H
- " 7*& + ( -
K$ %
+ (#< " +KBU
!M)64<3

K
 '( %) *+$ 9%@$.: $"+ *%/.$%I$ B# -!5%0 *:.-*
KL86:3 %7DMC&%C*1 K$a$( a$ $/"#$7"8187/#1'&/'7$#O.8/'*)$'& -#

1.@ DH " +KBU 2 -# / #)

1F!G " ! H ! ! ! L " K , 0 ( " 7*& - # / " #< +( H
./ " +KBU 2 - KBU S L + +
- " 7*& ( -
K$ %
-# + (#< " +KBU
(N
L?& '

.?N? !M)64<3
>&'(?4(+@?7+4@5&4+
!CA
Ka(a
!C/ #@
3!! " Hb " KBKH
LLK LLKBHU D$$$( L$ LLK

<,(.
$44
!
M EKBU K %
')L

"C0
J F G ) G + J F 'G
!M)64<3 "CB :) "
J F 'G
"CA "*

!"#$58J'5.5$*?#1&"**/$/*$ "C/
.)'/$&/#>$'&
"
: ) " 3 + F( \ +( H G3 " "7 / " A B 0 !" !/
@ "
"4

'( %) *+$ 9%@$.: $"+ *%/.$%I$ B# -!5%0 *:.-*

KL86:3 7DMC& C*1 K a ( a /"# &/8)6816 &#7*)6 *16#1 &4&/#5
"F!G %F!G #<H #F!G -#
!F ! ! H(# < G # F!G # <H
" H
" F ! G ! ! H(# < ! ! #<H
#)
1 F ! G " ! ! H(# < ! ! # <
H H
( " 7*& - # / " #< +( H
/"# &4&/#5 - " 7*& + ( -
K$ %
"F!G %F!G L #F!G + (#< " +KBU
# F!G L !M)64<3
!F! ! G "
"F!G ! H ! ! ! L
-#
1F !G " ! H ! ! ! L " K ! D H " +KBU 2 -#
" +KBU 2 - KBU S L +
- " /8) + S L + #) KBU S L +
!"#$58J'5.5$*?#1&"**/$/*$.)'/$ ( " 7*& - -
&/#>$'& + F ( \ + ( H G3 %
:) "3 EKBU K$
!M)64<3
 '( %) *+$ 9%@$.: $"+ *%/.$%I$ B# -!5%0 *:.-* %
F!G$
"F!G %F&G (F!G #F!G
KL86:3 7DMC& &7F!G & F!G &HF!G
!"#$7"8187/#1'&/'7$#O.8/'*)$'&
$F!G
L
1F!G " ! " !F! ! G ! L " K
!F! ! G
*! L + + !F! ! G -#

L?& '
!"#$1**/$&#)&'/'?'/4$'& 2 L@ F! G " "
*L ! "@ @ H! ! ! !"@
Ka(a
X"#)$Ka(a D$/"#$1**/&$81#$7*5>2#J$8)6 @?L$EKBU$c$-#B$ LLK LLKBHU D$$$( L$ LLK
%
*! L + + !F ! ! G F! ! G M EKBU K
2 L@ F !G " " "

')L
*L ! "@ @ H! ! ! ! "@ H! ! @ "+KBU! -#
!M)64<3
+ KBU ! -#
"
H -#
KBHU ! # H
!"#)D R 2 L@ F !G R"
S# H

'( %) *+$ 9%@$.: $"+ *%/.$%I$ B# -!5%0 *:.-*

KL86:3 %7DMC&%C*1 K$a$( a$ $/"#
58()'/.6#$*+$/"#$&#)&'/'?'/4 ! E"
)3@,3<G?7H3@!C77G
KBHU ! # H
R 2 L@ F ! G R" "C0 !3<!;G;H;GJ A"
S# H
R 2 L@ R KK: )
8)6$/"#$>#17#)/$*?#1&"**/$ "CB D"
/*$.)'/$&/#>

: ) " 3 +F( \ +( H G3 "CA /"

81# >2*//#6 ') C'(.1# +*1 L L KBHY "C/ !"

/* L L UB
M/ '& -#&/ /* 1#6.7# /"# &#)&'/'?'/4 " "
9"'2# 2'5'/')( L /* BU *1 2#&&B " ! / D
L
A E

X# /"#) 8//8') /"# 58<*1'/4 *+ /"# 8//8')8-2# 1#6.7/'*) ') &#)&'/'?'/4 9"'2#
58')/8')')( (**6 >#1+*158)7# +*1 /"# &/#> 1#&>*)&#B

H
Example 7.1: Find the value of k ( 1 to 2 with step 0.1 ) that minimize root
sensitivity and the overshot with weighting factor =2

Solution:
𝑠 2 + 2𝜁𝑤𝑛 𝑠 + 𝑤𝑛 2 = 0

K 2Mp
1 0.5774 0.3261
1.1 0.5688 0.3640
1.2 0.5620 0.3991
1.3 0.5563 0.4318
1.4 0.5517 0.4623
1.5 0.5477 0.4908
1.6 0.5443 0.5175
1.7 0.5414 0.5426
1.8 0.5388 0.5663
1.9 0.5365 0.5888
2 0.5345 0.6100
 '( %) *+,-%@$.! 9#$ %/.$%0.1#*"%2.-"$.3

M)$(#)#182D$9#$78)$.&#$/"#$6#&'()$>1*7#6.1#$8&$+*22*9&;
4" :%CD%3:#/7"$/"#$1**/$2*7.&$*+$/"#$7*5>#)&8/#6$&4&/#5$9'/"$&7F!G$7"*&#)$/*
8//8')$/"#$6#&'1#6$2*78/'*)$+*1$/"#$6*5')8)/$1**/&B
4" :%ED%d8J'5'P#$/"#$(8')$*+$&7F!G$/*$1#6.7#$/"#$#++#7/$*+$/"#$6'&/.1-8)7#B
4" :%FD%%#/#15')#$21$ 8)6$8//8')$/"#$5')'5.5$?82.#$*+$/"#$&#)&'/'?'/4
7*)&'&/#)/$9'/"$/"#$/18)&'#)/$1#&>*)&#$1#O.'1#6D$8&$6#&71'-#6$')$3/#>$ B

'( %) *+,-%@$.! 9#$ %/.$%0.1#*"%2.-"$.3

KL86:3 7DME& 0*)&'6#1 /"# +##6-87: 7*)/1*2 &4&/#5 8& &"*9) ') C'(.1# 9"#1#

-5*+?6'$%7$8
& F !G " D $ F !G "
!H E"
#$%&'()*+?,*-.

F!G$ "

"F!G %F&G (F!G #F!G

&7F!G &F!G
E"

$F!G !F1

'F!G !F1CE
23$4+?,*+%.

!0"
N*6#$6'8(185$*+$/"#$.)7*5>#)&8/#6$&4&/#5$$
!0"CE

!0!
! " ! /
!" !" !" !"
97+:)+&;<??,7$*=4+;.

Q
 '( %) *+,-%@$.! 9#$ %/.$%0.1#*"%2.-"$.3

KL86:3 7DME& Z#/ .& 7*)&'6#1 &7F!G '& /* -# 6#&'()#6 -4 +1#O.#)74 1#&>*)&#
5#/"*6&B !"#1#+*1#D /"# 7*5>#)&8/*1 '& /* -# &#2#7/#6 /* 87"'#?# 8) 8>>1*>1'8/#
>"8&# 581(') F8" 3 8*" JNOPH%&9 4 JNM7G 8)6 (8') F8" 3 8*" CM '(D H"&9 4 CM
'(7G 9"'2# 5')'5'P')( &#)&'/'?'/4 8)6 /"# #++#7/ *+ /"# 6'&/.1-8)7# 8)6 72*&#6E
2**> -8)69'6/" #N P QBKK @4/\!3,K F!G$
"F!G %F&G (F!G #F!G
&7F!G &F!G
& F !G " D $ F!G "
!H $F!G
'F!G
!".&D 9# 7"**&# /"# +*22*9')( 7*)/1*22#1 /* -# 6#&'()#6 /* 87"'#?# /"# 6#&'1#6
>#1+*158)7#
L F! \ O ! G
&, F ! G "
F! \ ) ! G

9"#1# L '& /"# (8') *+ /"# 7*)/1*22#1 8)6 O 8)6 ) 81# 1#&>#7/'?#24 /"# P#1* 8)6
>*2# *+ /"# 7*)/1*22#1B

'( %) *+,-%@$.! 9#$ %/.$%0.1#*"%2.-"$.3

KL86:3 7DME& &7F!G '& &#2#7/#6 /* 87"'#?# /* "8?# >"8&# 581(') 8/ 2#8&/
SUeDF1:6 4 SUKG 8)6 (8') F8/ 2#8&/ K /0D F&:6 4 K /0GG 9"'2# 5')'5'P')(
&#)&'/'?'/4 8)6 /"# #++#7/ *+ /"# 6'&/.1-8)7#B
F!G$
"F!G %F&G (F!G #F!G
& F !G " H
D $ F !G " &7F!G &F!G
!
$F!G
L F! \ O ! G 'F!G
&, F !G "
F! \ ) ! G

!"# (8') *+ /"# 7*)/1*22#1 L L K '& &#2#7/#6 /* 1#6.7# /"# #++#7/ *+ /"#
6'&/.1-8)7#B !* 8//8') 8 >"8&# 581(') *+ SUeD 9# &#2#7/ O L HBK 8)6 ) L HBKB

S
 Lecture 8: Design of robust PID-controlled systems
Based on ITAE Methods

Design of Robust PID Controlled Systems

D(s)
E(s) U(s) Y(s)
The PID controller has the transfer R(s) Gc(s) Gp(s)
function;
KI H(s)
Gc ( s )  K P   KD
s N(s)
The popularity of PID controllers can be attributed partly to their robust
performance in a wide range of operating conditions and partly to their functional
simplicity, which allows engineers to operate them in a simple straightforward
manner.
To implement such a controller, three parameters must be determined for the
given process: proportional gain KP, integral gain KI, and derivative gain KD .The
PID controller can be rewritten as
K K s 2  K P s  K I K D (s 2  as  b) K D ( s  z1 )(s  z 2 )
Gc (s )  K P  I  K D s  D  
s s s s
where a = KP/KD and b = KI/KD. Therefore, a PID controller introduces a transfer
function with one pole at the origin and two zeros z1 and z2 that can be located
anywhere in the left-hand s-plane.

1
 Design of Robust PID Controlled Systems
Example 8.01: Consider the following PID controlled system;
1 D(s)
Gp ( s)  , H (s)  1 R(s) E(s) U(s) Y(s)
( s  2)( s  5) Gc(s) Gp(s)
K D ( s  z1 )(s  z 2 )
Gc (s )  H(s)
s N(s)
where zeros z1 and z2 are complex.
The characteristic equation of PID controlled system is j
P ( s)  1  Gc ( s)G p ( s ) H ( s)  0 z1
KD KD
increasing
The loop transfer function is KD =0 KD =0 KD =0
K ( s  z1 )( s  z2 ) 1 r3 =-5 r2 =-2 r1 =0 
Gc ( s)G p ( s) H ( s)  D
s ( s  2)( s  5) KD
z2
As the gain KD of the controller is increased, the
complex roots approach the zeros.

Design of Robust PID Controlled Systems

Example 8.01: The closed-loop transfer function is
D(s)
R(s) E(s) U(s) Y(s)
Y (s ) Gc ( s)G p ( s ) Gc(s) Gp(s)

R ( s ) 1  Gc ( s )G p ( s) H ( s ) H(s)
N(s)
Y ( s) K D ( s  z1 )( s  z 2 )
 j
R ( s ) ( s  r1 )(s  r2 )(s  r3 )
z1 r1 KD >>1
Large
For large enough KD the dominant pole of
r3 KD =0
the closed loop system is r3. Then closed KD =0 KD =0
loop transfer function is -10 -8 -6 -4 -2 0 

r2
Y (s) KD KD z2
 
R ( s ) ( s  r3 ) ( s  K D )

2
 Design of Robust PID Controlled Systems
Example 8.01: The closed-loop transfer function is
D(s)
R(s) E(s) U(s) Y(s)
Y (s) K D ( s  z1 )(s  z 2 ) Gc(s) Gp(s)

R ( s ) ( s  r1 )(s  r2 )(s  r3 ) H(s)
N(s)
When KD is large KD >>1 the closed loop
transfer function is j

Y (s) KD z1 r1 KD >>1
 Large
R(s) (s  K D ) r3 KD =0 KD =0 KD =0

-10 -8 -6 -4 -2 0 
Robustness: The only limiting factor is the
allowable magnitude of U(s) when KD is large. r2
If KD is 100, the system has a fast response and z2
zero steady state error. Furthermore, the effect
of the disturbance is reduced significantly

Design the parameters of PID Controller

The selection of the three coefficients of PID controllers is basically a search
problem in a three-dimensional space. A PID controller can be determined by
moving in this search space on a trial-and-error basis.
KI D(s)
Gc (s )  K P   KDs R(s) E(s) U(s) Y(s)
s Gc(s) Gp(s)
K s2  KPs  KI H(s)
 D
s N(s)
The main problem in the selection of the three coefficients of PID is that these
coefficients do not readily translate into the desired performance and
robustness characteristics that the control system designer has in mind.
Generally the rules and methods introduced to select PID parameters can be
listed as follows;
1) Select PID parameter based on the experimental results when the system is not
modeled.
2) Select PID parameter when the system model Gp(s) is given.

3
Select the parameters of PID Controller when the system model is
given
Select the parameters of PID Controller when the system model Gp(s) is
given.
Several methods have been proposed to select PID parameters.
Here we will discus the design methods using
a) Root locus and
b) Performance indices.
D(s)
K R(s) E(s) U(s) Y(s)
Gc ( s)  K P  I  K D s Gc(s) Gp(s)
s
2
K s  K Ps  KI H(s)
 D N(s)
s

Select the parameters of PID Controller Based on Root Locus

a) Design the parameters of PID Controller Based on Root Locus

D(s)
K R(s) E(s) U(s) Y(s)
Gc ( s )  K P  I  K D s Gc(s) Gp(s)
s
K D ( s  z1 )( s  z 2 ) H(s)

s N(s)

The design of robust control systems using a PID controller based on the root
locus approach may be simply stated as follows:
1. Place the desired dominant closed loop poles and plot root locus of
uncompensated on the s-plane.
2. Select a location for the zeros of Gc(s) that will result in an acceptable root
locus and suitable dominant roots.
3. Test the transient response of the compensated system and iterate Step 2, if
necessary.

4
 Design the parameters of PID Controller Based on ITAE
performance index
b) Design method uses the ITAE performance index

KI D(s)
Gc ( s)  K P   KDs R(s) E(s) U(s) Y(s)
s Gc(s) Gp(s)
K s2  K P s  K I
 D H(s)
s
N(s)

Modern control theory assumes that the systems engineer can specify
quantitatively the required system performance. Then a performance index
can be calculated or measured and used to evaluate the system's performance.
A quantitative measure of the performance of a system is necessary for the
operation of modern adaptive control systems, for automatic parameter
optimization of a control system, and for the design of optimum systems.

Design the parameters of PID Controller Based on ITAE

performance index
Performance index: A performance index is a quantitative measure of the
performance of a system and is chosen so that emphasis is given to the important
system specifications.
A system is considered an optimum control system when the system parameters
are adjusted so that the index reaches an extremum, commonly a minimum value.
To be useful, a performance index must be a number that is always positive or zero.
Then the best system is defined as the system that minimizes this index.
A suitable performance index is the Integral of the Square of the Error, ISE,
which is defined as D(s)
R(s) E(s) U(s) Y(s)
T Gc(s) Gp(s)
2
I ISE   e (t )dt
0 H(s)
N(s)
The upper limit T is a finite time chosen somewhat arbitrarily so that the integral
approaches a steady-state value e(t) = r(t) - y(t). It is usually convenient to choose
T as the settling time ts.

5
 Design the parameters of PID Controller Based on ITAE
performance index
The Integral of the Square of the Error, ISE Index.
A
D(s) r(t)
R(s) E(s) U(s) Y(s) (a)
Gc(s) Gp(s) 0 t
H(s) A
T y(t)
2
N(s)
I ISE   e (t )dt (b)
0 t
0
A
The step response for a specific feedback e(t)
control system is shown in Figure (b), and the (c)
0 t
error in Figure (c). The error squared is shown
in Figure (d), and the integral of the error A2
e2(t)
squared in Figure (e). This criterion will (d)
discriminate between excessively overdamped 0 t
and excessively underdamped systems. The 2

minimum value of the integral occurs for a 

e (t ) dt

(e)
compromise value of the damping.
0 t

Design the parameters of PID Controller Based on ITAE

performance index
A
The Integral of the Square of the Error, r(t)
ISE Index. (a)
0 t
A
The performance index of y(t)
T (b)
0 t
I ISE   e 2 (t )dt A
e(t)
0 (c)
0 t
is easily adapted for practical A2
measurements because a squaring circuit is e2(t)
(d)
readily obtained. Furthermore, the squared 0 t
error is mathematically convenient for
2

analytical and computational purposes.  e (t )dt

(e)
0 t

6
 Design the parameters of PID Controller Based on ITAE
performance index
The Integral of the Absolute magnitude of the Error, IAE Index: The integral
of the absolute magnitude of the error, IAE, is another readily instrumented
performance criterion and is written as
T
I IAE   | e(t ) | dt
0
This index is particularly useful for computer simulation studies.
The Integral of Time multiplied by Absolute Error, ITAE Index: This index
has been proposed to reduce the contribution of the large initial error to the value
of the performance integral, as well as to emphasize errors occurring later in the
response, the following
T
I ITAE   t | e(t ) | dt
0
Another similar index is the Integral of Time multiplied by the Squared
Error, ITAE Index T
I ITSE   te 2 (t )dt
0

Design the parameters of PID Controller Based on ITAE

performance index

The Integral of Time multiplied by Absolute Error, ITAE Index:

D(s)
T R(s) E(s) U(s) Y(s)
Gc(s) Gp(s)
I ITAE   t | e(t ) | dt
0 H(s)
N(s)
The performance index ITAE provides the best selectivity of the performance
indices; that is, the minimum value of the integral is readily discernible as the
system parameters are varied. The general form of the performance integral is

T
I ITAE   f (e(t ), r (t ), y (t ), t )dt
0

where f is a function of the error e(t), input r(t), output y(t), and time t.

7
 Design the parameters of PID Controller Based on ITAE
performance index
The Integral of Time multiplied by Absolute Error, ITAE Index:
D(s)
R(s) E(s) U(s) Y(s)
T Gc(s) Gp(s)
I ITAE   f (e(t ), r (t ), y (t ), t )dt
H(s)
0
N(s)

The minimization of a performance index can be directly related to the

minimization of fuel consumption for aircraft and space vehicles.
Performance indices are useful for the analysis and design of control systems.
A control system is optimum when the selected performance index is
minimized. However, the optimum value of the parameters depends directly on
the definition of optimum, that is, the performance index.

Design the parameters of PID Controller Based on ITAE

performance index
The Integral of Time multiplied by Absolute Error, ITAE Index:
D(s)
T R(s) E(s) U(s) Y(s)
Gc(s) Gp(s)
I ITAE   t | e(t ) | dt
0 H(s)
N(s)
The coefficients that will minimize the ITAE performance criterion for a step
input have been determined for the general closed-loop transfer function
Y ( s) b0
 n n 1
R ( s) s  bn1s  ...  b1s  b0
This transfer function has a steady-state error equal to zero for a step input. Note
that the transfer function has n poles and no zeros.

8
 Design the parameters of PID Controller Based on ITAE
performance index
The Integral of Time multiplied by Absolute Error, ITAE Index:
T D(s)
I ITAE   t | e(t ) | dt R(s) E(s) U(s) Y(s)
Gc(s) Gp(s)
0

Y ( s) b0 H(s)
T ( s)   n n 1 N(s)
R( s ) s  bn 1s  ...  b1s  b0
The optimum coefficients for the ITAE criterion are given in Table.
Table 1: The Optimum Coefficients of T(s) Based on the ITAE Criterion for
a Step Input s  n
s 2  1.4 n s   n2
s 3  1.75n s 2  2.15n2 s  n3
s 4  2.1n s 3  3.4 n2 s 2  2.7n3 s   n4
s 5  2.8 n s 4  5 n2 s 3  5.5 n3 s 2  3.4 n4 s   n5
s 6  3.25 n s 5  6.6n2 s 4  8.6n3 s 3  7.45 n4 s 2  3.95 n5 s   n6

Design the parameters of PID Controller Based on ITAE

performance index
The Integral of Time multiplied by Absolute Error, ITAE Index:
T D(s)
I ITAE   t | e(t ) | dt R(s) E(s) U(s) Y(s)
Gc(s) Gp(s)
0

Y (s ) H(s)
 T ( s) N(s)
R( s )
b0
 n n 1
s  bn1s  ...  b1s  b0

The responses using optimum

coefficients given in Table 1
for a step input are given in
Figure for ITAE.

9
 Design the parameters of PID Controller Based on ITAE
performance index
b) Design method uses the ITAE performance index
D(s)
K R(s) E(s) U(s) Y(s)
Gc (s )  K P  I  K D s Gf(s) Gc(s) Gp(s)
s
K s2  KPs  KI H(s)
 D
s N(s)
The three PID coefficients is selected to minimize the ITAE performance index,
which produces an excellent transient response to a step or a ramp.
The design procedure consists of three steps:
1. Select the n of the closed-loop system by specifying the settling time.
2. Determine the three coefficients using the appropriate optimum equation
(Table 1) and the n of step 1 to obtain Gc(s).
3. Determine a prefilter Gf(s) so that the closed-loop system transfer function,
T(s), does not have any zeros, as required by
Y (s ) b0
 T ( s)  n n 1
R( s ) s  bn1s  ...  b1 s  b0

Design the parameters of PID Controller Based on ITAE

performance index
Example 01: Consider a temperature controller with a control system as shown in
Figure and a process D(s)
R(s) E(s) U(s) Y(s)
1 Gf(s) Gc(s) Gp(s)
G p (s ) 
( s  1) 2
H(s)
N(s)

If Gc(s) = 1, the steady-state error is 50%, and the settling time is ts = 3.2 seconds
for a step input.
The desired performance: Use a PID controller to obtain an optimum ITAE
performance for a step input and a settling time of less than 0.5 second ts < 0.5 .

KD s2  KPs  KI Y (s) Gc ( s)G p ( s) b0

Gc (s )    n n 1
s R( s ) 1  Gc ( s )G p ( s ) H ( s ) s  bn 1s  ...  b1 s  b0

10
 Design the parameters of PID Controller Based on ITAE
performance index
Example 01: D(s)
R(s) E(s) U(s) Y(s)
1 Gf(s) Gc(s) Gp(s)
G p (s ) 
( s  1) 2
K s2  KPs  KI H(s)
Gc (s )  D N(s)
s
The closed-loop transfer function without prefiltering [Gf(s) = 1] is
Y (s) Gc ( s)G p (s ) KD s 2  K P s  K I
  3
R( s ) 1  Gc ( s )G p ( s) H ( s ) s  ( 2  K D ) s 2  (1  K P ) s  K I
The optimum coefficients of the characteristic equation for ITAE are obtained
from Table 1 as
s 3  1.75n s 2  2.15n2 s  n3
We need to select n in order to meet the settling time requirement.
Since ts = 4/(n) and  is unknown but near 0.8, we set n =10 rad/sec.

Design the parameters of PID Controller Based on ITAE

performance index
Example 01: D(s)
1 R(s) E(s) U(s) Y(s)
G p (s )  Gf(s) Gc(s) Gp(s)
( s  1) 2
KDs2  KPs  KI H(s)
Gc (s )  N(s)
s Y (s) K D s2  KP s  K I
Without prefiltering [Gf(s) = 1] 
R( s ) s 3  (2  K D ) s 2  (1  K P ) s  K P
Equating the denominator of closed loop transfer function to Equation
s 3  1.75n s 2  2.15n2 s  n3
Leads: s 3  ( 2  K D ) s 2  (1  K P ) s  K P  s 3  1.75 n s 2  2.15 n2 s   n3
Then the three coefficients of PID are KP = 214, KD = 15.5, and KI = 1000.
Y (s) KDs 2  K P s  K I 15.5s 2  214 s  1000
 3 
R( s ) s  (2  K D )s 2  (1  K P ) s  K P s 3  17.5s 2  215s  1000
15.5( s  6.9  j 4.1)( s  6.9  j 4.1)

s 3  17.5s 2  215s  1000

11
 Design the parameters of PID Controller Based on ITAE
performance index
Example 01: Without prefiltering [Gf(s) = 1]
D(s)
Y (s) Gc ( s)G p (s ) R(s) E(s) U(s) Y(s)
 Gf(s) Gc(s) Gp(s)
R( s ) 1  Gc ( s )G p ( s) H ( s )
15.5s 2  214 s  1000 H(s)
 N(s)
s  17.5s 2  215 s  1000
3

15.5( s  6.9  j 4.1)( s  6.9  j 4.1)


s 3  17.5s 2  215s  1000
The response of this system to a step input has an overshoot of 32%, as recorded
in the following Table 2.
Controller Gc(s) =1 PID and Gf(s) =1
Percent overshoot 0 31.7%
Settling time (sec) 3.2 0.2
Steady-state error 50.1% 0.0%
Disturbance error 52% 0.4%

Design the parameters of PID Controller Based on ITAE

performance index
Example 01: The following prefilter Gf(s) is selected
D(s)
64.5 R(s) E(s) U(s) Y(s)
G f ( s)  2 Gf(s) Gc(s) Gp(s)
s  13.8s  64.5
H(s)
so that we achieve the desired 1TAE response with N(s)

Y (s) G f ( s )Gc ( s )G p ( s ) 1000

  3 2
R( s ) 1  Gc ( s )G p ( s ) H ( s) s  17.5s  215s  1000

Controller Gc(s) =1 PID and Gf(s) =1 PID with Gf(s)

Percent overshoot 0 31.7% 1.9%
Settling time (sec) 3.2 0.2 0.45
Steady-state error 50.1% 0.0% 0.0%
Disturbance error 52% 0.4% 0.4%

12
 Design the parameters of PID Controller Based on ITAE
performance index
Example 01: D(s)
R(s) E(s) U(s) Y(s)
64.5 Gf(s) Gc(s) Gp(s)
G f (s) 
s 2  13.8 s  64.5
H(s)
K D s 2  K P s  K I 15.5s 2  214s  1000 N(s)
Gc (s )  
s s
Y (s) G f ( s )Gc ( s )G p ( s ) 1000
  3 2
R( s ) 1  Gc ( s )G p ( s ) H ( s) s  17.5s  215s  1000
The system has a small overshoot, a Controller PID with Gf(s)
settling time of less than 0.5 sec, and zero
Percent overshoot 1.9%
steady-state error. Furthermore, for a
disturbance D(s) =1/s, the maximum value Settling time (sec) 0.45
of y(t) due to the disturbance is 0.4% of the Steady-state error 0.0%
magnitude of the disturbance. Disturbance error 0.4%

Results: This is a very favorable design.

13
 L !"#$ %9&% !"#$%&'()'*(+,#-'.&-"*&/0'1(2"0'3(&-*(0
#4#-"5#

.67896:;'1<!8;'!8=.>6

7?" @*(+0"5 () 2"#$%&$&% / A(5@"&#/-(* -?/- @*(B$2"# /#45@-(-$A -*/AC$&% () /

*")"*"&A" $&@,- D$-? E"*( #-"/24F#-/-" "**(* $# A(&#$2"*"2G
7?" *")"*"&A" $&@,-# A(&#$2"*"2 A/& $&A0,2" #-"@#H */5@#H /&2 (-?"* @"*#$#-"&-
#$%&/0#H #,A? /# #$&,#($2#G
$K"L &K"L #K"L
# K "L ! K"L !K"L
" 2 K" L " ! ! K"L % K"L 'K(L )K(L *K(L
$ K "L ! !K "L % K "L ,K"L #K"L
%K"L
. K-1 " ! LK-+ " ! LGGGK-0 " ! L +K(L *K(L
! K" L % K "L "
" K- " ! LK-M " ! LGGGK- / " ! L
I(* / #-"@ $&@,-H D" C&(D -?/- E"*( #-"/24F#-/-" -*/AC$&% "**(*# A/& +" /A?$"B"2
D$-? / -4@"F(&" J #4#-"5G
7?$# $2"/ $# )(*5/0$E"2 ?"*" +4 $&-*(2,A$&% /& '(" $()* +,- * () -?" *")"*"&A"
$&@,- $& -?" A(5@"&#/-(*G 3K"L
$K"L &K#L #K"L
!AK"L !/K"L

%K"L
Type 1
1 1 1 1 1
𝐺𝑐 𝑠 𝐺𝑝 𝑠 = = . , step input = , ramp input =
𝑠 𝑠+𝑎 𝑠 𝑠+𝑎 𝑠 𝑠2

Type 2
1 1 1 1 1 1 1
𝐺𝑐 𝑠 𝐺𝑝 𝑠 = = . = . , step input = , ramp input =
𝑠 2 𝑠+𝑎 𝑠2 𝑠+𝑎 𝑠 𝑠 𝑠+𝑎 𝑠 𝑠2

𝑠𝐼 − 𝐴 + 𝐵𝐾 = 𝑠 + 1 − 𝑗 𝑠 + 1 + 𝑗 (𝑠 + 10)

𝑠 0 0 0 1 0 0
0 𝑠 0 − 0 0 1 + 0 𝑘1 𝑘2 𝑘3 = 𝑠 + 1 − 𝑗 𝑠 + 1 − 𝑗 (𝑠 + 10)
0 0 𝑠 0 −2 −2 1
 .67896:;'1<!8;'!8=.>6

3(&#$2"* / #-/-" B/*$/+0" 5(2"0 () -?" @0/&- %$B"& +4

4 " 74 ! ,5H * " 64

D?"*" 4#$& $# -?" #-/-" B"A-(*H 5 $# -?" $&@,-H /&2 * $# -?" (,-@,-G
N" D$00 A(&#$2"* / *")"*"&A" $&@,- -( +" %"&"*/-"2 +4 / 0$&"/* #4#-"5 () -?" )(*5

4' " 7' 4' H ' " : ' 4'

D$-? ,&C&(D& $&$-$/0 A(&2$-$(&#G :& "P,$B/0"&- 5(2"0 () -?" *")"*"&A" $&@,- 'K(L $#

' K 8L " $ 8 % ' K 8 % L ! $ 8 %M ' K 8%ML ! GGG ! $ ' ! $ O '

D?"*" 'K8L $# -?" 8-? 2"*$B/-$B" () 'K(L9

.67896:;'1<!8;'!8=.>6
:#45@-(-$A'-*/AC$&%'()')%." /%$ 0 $ (! %'(/#"
12 %- .'3(%/$,4* +%$#'-('2"#$%&'()'/'A(&-*(00"*'-('"&/+0"'-?"'-*/AC$&%'()')%." /%
$ 0 $ (! %'(/#" D$-?'E"*('#-"/24F#-/-"'"**(*G.& -?$#'A/#"H'"2 %$ 0 $ (! %'(/#"%$#'
%"&"*/-"2'+4 4 " 7 4 & 4 " OH ' " : 4 & ' " 4
' ' ' ' ' ' '
/&2'-?"'-*/AC$&%'"**(*');$#'2")$&"2'/# ) " * % 'G
7/C$&%'-?"'-$5"'2"*$B/-$B"'()'"**(*') 4$"02# ) " * % ' " 64
.)'D"'2")$&"'-?"'-D('$&-"*5"2$/-"'B/*$/+0"#Q <"4 /&2 ="5
2 ) / ,O 6 )2 ) / , O )
00 -- " * '00 -- ! * ' = 567%
1 < . +O 7 (1 < . + , (
.)'-?"'#4#-"5'$&'8P,/-$(&'K L'$#'A(&-*(00/+0"H' > " T, 7, 7M , ! 78 % , SH R > R3 O

D"'A/&')$&2')%.")" %0 -4)!8%!,("$,*%%()'-?"')(*5 = " %? ) % . M < 597%

#,A? -?/- 8P,/-$(& K L $# #-/+0" /A?$"B" -?" 2"#$*"2 +"?/B$(,*G 7?$# $5@0$"# -?/- -?"
-*/AC$&% "**(* ) $# #-/+0"Q -?,#H D" D$00 ?/B" /A?$"B"2 -?" (+U"A-$B" () /#45@-(-$A
-*/AC$&% D$-? E"*( #-"/24 #-/-" "**(* ) JOG

M
 .67896:;'1<!8;'!8=.>6
:#45@-(-$A'-*/AC$&%'()')%." /%$ 0 $ (! %'(/#"
12 %!,("$,*%'(/#"H')(,&2'+4'$&-"%*/-$&%'=H'$#
(
5K( L " %? 4 )K5 L:5 % . M 4K( L
O
7?" A(**"#@(&2$&% +0(AC 2$/%*/5 $# #?(D& $& I$%,*"G N" #"" -?/- -?" A(5@"&#/-(*
$&A0,2"# /& '(" $()* +,- * K-?/- $#H /& $&-"%*/-(*L () -?" *")"*"&A" #-"@ $&@,-G

$K"L &K#L 3K"L 4 4 #K"L

? , 6
" "
7
4
3(&-*(00"* .M

=(/#" <,("$,** $ :;." +%-;()+'!.

=-"@'
(
5 K( L " % ? 4 )K5 L:5 % . M 4 K( L
4 " 74 ! ,5
*")"*"&A" O
* " 64

.67896:;'1<!8;'!8=.>6
:#45@-(-$A'-*/AC$&%'()')%." /%$ 0 $ (! %'(/#"
>?)+/* 9@A6& =(" $()* +,- * - .'3( )(* / #('" ." / '(/#"H * J 4* W
;"- ,# A(&#$2"* / @*(A"## %$B"& +4
,4 ) , O ) , 4 ) ,O) ,4 )
* 4 ' " *% M % M ' * 4 ' ! * '5 H * "6 O7* '
+ M( + (+ M ( + ( + 4M (
(
,4 )
N" D/&- -( 2"#$%& / A(&-*(00"* 5 K( L " % ? 4 )K5 L:5 % 6? M ? MM 7* '
+ 4M (
,O 4 )
(* = " % ? ) % . M < " % ? ) % 6? M ? MM 7* ' D?"*"' . M " 6? M ? MM 7
+ 4M (
)(* -?$# #4#-"5 -( -*/AC / *")"*"&A" #-"@ $&@,- D$-? E"*( #-"/24F#-/-" "**(* D$-?
2"#$*"2 A0(#"2 0((@ @"*)(*5/&A"Q " HM J F 8 @H /&2 "V J F OG
$K"L &K#L 3K"L 4 4 #K"L
? , 6
" "
7
4
.M

V
 .67896:;'1<!8;'!8=.>6
:#45@-(-$A'-*/AC$&%'()')%." /%$ 0 $ (! %'(/#"
:,*#"',( YGO W 7?" 24&/5$A () -?" #4#-"5 $&-"*$5# () $&-"*5"2$/-" B/*$/+0"#Q
,< ) ,4 )
*< ' " *4 ' /&2 ="5 0"/2#
+ M( + M(
, ) ) ,O O ) , ) ) , O)
* < ' " *O O ' * < ' ! * O' = 567%
* ' * '* ' * '
*+ < M '( *+O % M % M'( *+ < M '( *+ '(
,4 )
7?" #-/-" )""2+/AC A(&-*(00"* $#Q = " % ? ) % . M < " % ? ) % 6? M ? MM 7* ' 597%
:'A?"AC'()'A(&-*(00/+$0$-4
+ 4M (
,O O )
> " T, 7, 7M ,S " *O % M'
* ' /&2 R > R" % 3 O
*+ % M M '(

#?(D# -?" A(&-*(00/+$0$-4 5/-*$Z > ?/# ),00 */&C #( -?/- -?" #4#-"5 2"#A*$+"2 +4
8P,/-$(& K L $# A(5@0"-"04 A(&-*(00/+0" /&2 -?" A(&-*(00"* %$B"& $& KML A/& +"
2"#$%&"2G

.67896:;'1<!8;'!8=.>6
:#45@-(-$A'-*/AC$&%'()')%." /%$ 0 $ (! %'(/#"
:,*#"',( YGO W 7?" 24&/5$A () -?" #4#-"5 $&-"*$5# () $&-"*5"2$/-" B/*$/+0"#Q
,< ) ,4 )
*< ' " *4 ' /&2 ="5 0"/2#
+ M( + M(

, ) ) ,O O ) , ) ) , O)
* < ' " *O O ' * < ' ! * O' = 567%
* ' * '* ' * '
*+ < M '( *+O % M % M'( *+ < M '( *+ '(

7( /A?$"B" 2"#$*"2 A0(#"2 0((@ @"*)(*5/&A"Q 9 HM J F 8 @H /&2 9V J F O @0/A"

5"-?(2 A/& +" ,#"2 /&2 -?" A(&-*(00"* %/$&# /*"Q

? " MOH . M " 6? M ?MM 7 " 6MO O7

7?" #-/-" )""2+/AC A(&-*(00"* $#Q
,4 )
= " %MO) % 6MO O7* ' 597%
+ 4M (

X
 .67896:;'1<!8;'!8=.>6
:#45@-(-$A'-*/AC$&%'()')%." /%$ 0 $ (! %'(/#"
:,*#"',( YGO W 7?" 24&/5$A () A0(#"2 0((@ #4#-"5 /&2 #-/-" )""2+/AC A(&-*(0 /*"Q
, ) ) ,O O ) , ) ) , O) ,4 )
* < ' " *O O ' * < ' ! *O' = 567% = " %MO) % 6MO O7* ' 597%
* ' * '* ' * '
*+ < M '( *+O % M % M'( *+ < M '( *+ '( + 4M (
7?"'A0(#"2'0((@'#4#-"5'$#'
, ) ) ,O O ) , ) ) ,O )
* < ' " *O O ' * < ' % *O ' MO) ! 6MO ,4 )
* ' * '* ' * ' O7* '
*+ < M '( *+O % M % M'( *+ <M '( *+ '( + 4M (

,)) , O O ), ) )
*< ' " * O O '* < '
* ' * '* '
*+ < M '( *+% MO % MM % M'( *+ <M '(

7?"'A0(#"2'0((@'#4#-"5'$#'/#45@-(-$A/004'#-/+0"G =(')(*'/&4'$&$-$/0'-*/AC$&%'"**(*'
)KOL'D"'/*"'%,/*/&-""2'-?/-')K(L : O'/#'(;: ;9;

.67896:;'1<!8;'!8=.>6
:#45@-(-$A'-*/AC$&%'()')%." /%$ 0 $ (! %'(/#"
:,*#"',( YGO W 7?" /#45@-(-$A #-/+$0$-4 () -?" -*/AC$&% "**(* )K(L J 'K(L F 4 K(L $#
$00,#-*/-"2 $& I$%,*" )(* / #-"@ $&@,- /&2 )(* -?" $&$-$/0 A(&2$-$(&#
T)KOLH4 KOLH4MKOLSJT HOHOSG
9"#@(&#"' -('.&$-$/0'3(&2$-$(&#' ZOJT QOQOS
!'"

'(
7*/AC$&%'"**(*

'%

'$

'"

& '"
! " # $ )*+,-./,01 %

[
 .67896:;'1<!8;'!8=.>6
:#45@-(-$A'-*/AC$&%'()')%." /%$ 0 $ (! %'(/#"
3(&#$2"* -?" +0(AC 2$/%*/5 5(2"0 () I$%,*" D?"*" -?" @*(A"## $# *"@*"#"&-"2 +4

! / K" L " 6 K "A % 7L % , " ;((@'-*/&#)"*'),&A-$(&'$#

" M ! M" ! M ?
? !B K" L! / K" L "
/&2'-?"'A/#A/2"'A(&-*(00"*'$# !B K" L " " K " M ! M " ! ML
"

$K"L &K#L 3K"L 4 4 #K"L

? , 6
" "
!AK"L
!@K"L 7

.M
4

B .#*" A6& 7?" $&-"*&/0 5(2"0 @*$&A$@0" #-/-"# -?/- '0 !5!7 /5!7 !,(")'(. "5!7
-?"& #5$7 C'** "$)!8 %5$7 ).;+/","'!)**;G .& -?$# A/#" $K"LJ ]"C D?$A? $# A(&-/$&"2
$& !AK"L!@K"LH /# D" "Z@"A-G

.67896:;'1<!8;'!8=.>6
:#45@-(-$A'-*/AC$&%'()')%$)+/%'(/#"

3(&#$2"* /%/$& / #-/-" B/*$/+0" 5(2"0 () -?" @0/&- %$B"& +4

4 " 74 ! ,5H * " 64 567%

D?"*" 4#$& $# -?" #-/-" B"A-(*H 5 $# -?" $&@,-H /&2 * $# -?" (,-@,-G
7?" @*(+0"5 () 2"#$%&$&% / A(&-*(00"* -( @*(B$2" /#45@-(-$A -*/AC$&% () ) $)+/
'(/#" D$-? E"*( #-"/24F#-/-" "**(*H 'K(L D >(H ( < EH D?"*" > $# -?" */5@ 5/%&$-,2"G
.& -?$# A/#"H -?" *")"*"&A" $&@,- 5(2"0 $#

,O )
4' " 7' 4' & 4' " * ' 4'
+O O ( 597%
' " : ' 4' & ' " 6 O74'

.&'$&@,-F(,-@,-')(*5H'-?"'*")"*"&A"'5(2"0'$&'8P,/-$(&'KML'$#'%$B"&'+4 ' " OG

\
 .67896:;'1<!8;'!8=.>6
:#45@-(-$A'-*/AC$&%'()')%$)+/%'(/#"
: #-/-" B/*$/+0" 5(2"0 () -?" @0/&- $# 4 " 74 ! ,5H * " 64
_*(A""2$&% /# +")(*"H D" 2")$&" -?" -*/AC$&% "**(* ) /# )K(LJ *K(L F 'K(LG
/&2 -/C$&% -?" -$5"F2"*$B/-$B" -D$A" 4$"02# ) " * " 64G
N$-?'-?"'2")$&$-$(&#'()'-?"'-D('$&-"*5"2$/-"'B/*$/+0"#Q < " 4G = " 5 H
2 ) / ,O O )2 ) / , O )
0 - * '0 ) - ! * O ' =
0 - *
) " O O 6 '0 - * ' 567%
0 < - *O O 7' 0 < - * , '
1 . + (1 . + (
.)'-?"'#4#-"5'$&'8P,/-$(&'K L'$#'A(&-*(00/+0"H'
> " T, 7, 7M , ! 78 % ,SH R > R3 O 2)/
0 -
-?"&')%.")" %0 -4)!8%!,("$,*%()'-?"')(*5 = " %6? ?M . V 70 ) - 597%
0<-
A/& +" 2"#$%&"2 )(* -?" 2"#$*"2 A0(#"2 0((@ "$%"&B/0," -( 5/C" -?" A0(#"2 0((@
1 .
#4#-"5 +" /#45@-(-$A/004 #-/+0"G

.67896:;'1<!8;'!8=.>6
:#45@-(-$A'-*/AC$&%'()')%$)+/%'(/#"
7?" A(&-*(00/+0" #4#-"5 $# 7?"'.")" %0 -4)!8%!,("$,*%5/C"#'-?"'
2 ) / ,O O )2 ) / , O ) A0(#"2'0((@'#4#-"5'/#45@-(-$A/004'#-/+0"'$#G'
0 - * '0 - * '
0 ) - " *O O 6 '0 ) - ! * O ' = 567% 2)/
0 < - *O O 7'0 < - * , ' 0 -
1 . + (1 . + ( = " %6? ?M . V 70 ) - 597%
0<-
1 .
B .#*" A9& 7?" A(&-*(0H 5H $# )(,&2 +4 $&-"%*/-$&% 8P,/-$(& KML -D$A"G .& I$%,*"H D"
#"" -?/- "2 $ .#*"'(3 !,("$,** $ ?/# ) -,#4* '(" 3$)",$H D?$A? $# -?" '(" $()*
+,- * ,0 "2 $ 0 $ (! $)+/ '(/#"@

$K"L &K#L 3K"L 4 4 #K"L

? , 6
" " "

!@K"L 7
?M
.M
4
6G8('GHH)';!AK"L

^
 .67896:;'1<!8;'!8=.>6
:#45@-(-$A'-*/AC$&%'()')%$)+/%'(/#"
$K"L &K#L 3K"L 4 4 #K"L
? , 6
" " "

!@K"L 7
?M
.M
4
6G8('GHH)';!AK"L

B .#*" AD& 12 '(" $()* +,- * )//$,)!2 A/& +" ?" (- - -( (-?"* *")"*"&A"
$&@,-# +4 )(00(D$&% -?" #/5" %"&"*/0 @*(A"2,*" (,-0$&"2 )(* "2 ." / /&2 $)+/
$&@,-#G

B .#*" AE& .& /22$-$(&H -?" $&-"*&/0 5(2"0 2"#$%& A/& +" ,#"2 -( *"U"A- @"*#$#-"&-
2$#-,*+/&A"# +4 $&A0,2$&% 5(2"0# () -?" 2$#-,*+/&A"# $& -?" A(5@"&#/-(*G 7?$#
#,+U"A- -( -?" )(00(D$&% @/*- () -?$# 0"A-,*"G

9<ab=7'.67896:;'1<!8;'3<679<;'=c=781

N"'&(D'A(&#$2"*'/%/$&'-?"',#"'()'-?"'$&-"*&/0'5(2"0'2"#$%&'#?(D&'$&'I$%,*"'D$-?'
#@"A$/0'/--"&-$(&'-('*(+,#-'#4#-"5'@"*)(*5/&A"G'7?"''(" $()*%+,- *%/$'(!'/*

$K"L &K#L 3K"L #K"L

!5!7 !@K"L

I(1() J1'K1+H)";
&
'

7?" $&-"*&/0 5(2"0 @*$&A$@0" #-/-"# -?/- '0 !5!7 /5!7 !,(")'(. "5!7 -?"& #5$7 C'**
"$)!8 %5$7 ).;+/","'!)**; K$& -?" #-"/24 #-/-"LH /&2 -?" -*/AC$&% $# *(+,#-G
8Z/5$&$&% -?" #4#-"5 () I$%,*"H D" &(-" -?/- )(* 0(D"*F(*2"* @*(A"##"#H #-/-"
B/*$/+0" )""2+/ACH F&' D$00 &(- +" *"P,$*"2H /&2 / #,$-/+0" !5!7 A/& +" (+-/$&"2G
d(D"B"*H D$-? ?$%?"*F(*2"* #4#-"5#H -?" )""2+/AC () /00 #-/-" B/*$/+0"# 5/4 +"
*"P,$*"2G

`
 9<ab=7'.67896:;'1<!8;'3<679<;'=c=781

>?)+/* 9@A9& 3(&#$2"* / #$5@0" #4#-"5 D$-? /5!7 F 6G!C )(* D?$A? D" #""C /
*/5@ *"#@(&#" D$-? / #-"/24F#-/-" "**(* () E"*(G : 2A A(&-*(00"* $# #,))$A$"&-H /&2
D" 0"- & J O K&(
&K#L 3K"L #K"L
#-/-" B/*$/+0" )""2+/ACLG $K"L 5!7 !@K"L
!
! / K"L "
"
.A I(1() J1'K1+H)";
!B K " L " K . 2 ! L &
" '

.A . "!.
7?"& D" ?/B" !B K "L! / K "L " K . 2 ! L " 2 M A
" " "
6(-"'-?/-')(*'/'*/5@H'"5!7 F%6G!9H'D?$A?'$#'A(&-/$&"2'/#'/')/A-(*'()' !5!7 /5!7H'/&2'
-?"'A0(#"2F0((@'-*/&#)"*'),&A-$(&'$#

# K "L !B K " L! / K " L . " ! .A

" - K"L " " M 2
$K " L ! !B K " L! / K " L " ! . 2 " ! . A

9<ab=7'.67896:;'1<!8;'3<679<;'=c=781

>?)+/* 9@A9& 7?" -*/&#)"* ),&A-$(& () -?" A0(#"2 0((@ #4#-"5 D$-? / #$5@0"
#4#-"5 D$-? 2A A(&-*(00"* /&2 .JO $#
$K"L &K#L 3K"L #K"L
!5!7 !@K"L
# K "L . " ! .A
" M 2 .A !/ K"L "
$K "L " ! . 2 " ! . A !B K" L " K . 2 !
"
L "
I(1() J1'K1+H)";
&FA '
b#$&%'-?"'.7:8'#@"A$)$A/-$(&#')(*'/'*/5@H'"5!7 F%6G!9H''*"#@(&#"'%$B"&'$&'7/+0"' '
D"'*"P,$*"'-?/-
# K "L . " ! .A VGM=8 " ! =8M
" M 2 " M
$K "L " ! . 2 " ! . A " ! VGM=8 " ! =8M
1)4* %6&%7?"'<@-$5,5' 3("))$A$"&-#' ()'-K"L'a/#"2'(&'-?"'.7:8'3*$-"*$(&' )(*'/'9/5@'.&@,-
" M ! VGM= 8 " ! = 8M
" V ! G^[=8 " M ! VGM[= 8M " ! =8V
" ! MGX =8 " V ! XGYV=8M " M ! [G X= 8V" ! = 8X
X

" [ ! MG Y=8 " X ! \G[=8M " V ! \GV= 8V" M ! [GMX= 8X " ! =8[

Y
 9<ab=7'.67896:;'1<!8;'3<679<;'=c=781
>?)+/* 9@A9& 7?" -*/&#)"* ),&A-$(& () -?" A0(#"2 0((@ #4#-"5 D$-? / #$5@0"
#4#-"5 D$-? 2A A(&-*(00"* $#
$K"L &K#L 3K"L #K"L
!5!7 !@K"L
! / K"L "
.A "
!B K" L " K . 2 ! L
" I(1() J1'K1+H)";
&FA '

# K "L . " ! .A VGM=8 " ! =8M

" M 2 " M
$K "L " ! . 2 " ! . A " ! VGM=8 " ! =8M
N" #"0"A- =& -( #/-$#)4 / #@"A$)$A/-$(& )(* -?" #"--0$&% -$5"G I(* / #"--0$&% -$5"
KD$-? / Me A*$-"*$(&L () (# J ")BH D" #"0"A- =& J [ '1:]#G 7?"& D" *"P,$*" ._ J \
/&2 .. J M[G
7?" *"#@(&#" () -?$# #4#-"5 #"--0"# $& #"A(&2 /&2 -?"& -*/AC# -?" */5@ D$-? E"*(
#-"/24F#-/-" "**(*G

9<ab=7'.67896:;'1<!8;'3<679<;'=c=781
>?)+/* 9@A9& b&$- */5@ *"#@(&#"
# K"L .2" ! .A $K"L &K#L 3K"L #K"L
" !5!7 !@K"L
$K " L " M ! . 2 " ! . A
.A ! / K "L "
VGM=8 " ! = 8M !B K" L " K . 2 ! L "
" M " I(1() J1'K1+H)";
" ! VGM=8 " ! =8M &FA '
23+4 2,/456/,
!

'(
7+48*9:;,

'%

'$

'"
7?"'#"--0$&%' -$5"'$#'(#J' ")B

'" '$ '% )*+,-./,01 !

O
 9<ab=7'.67896:;'1<!8;'3<679<;'=c=781
>?)+/* 9@A9& H('" $)+/ $ ./,(. W
7?" -*/&#)"* ),&A-$(& () -?" A0(#"2 0((@ #4#-"5 D$-? / #$5@0" #4#-"5 D$-? 2A
A(&-*(00"* $#
$K"L &K#L 3K"L #K"L
!@K"L
!5!7
.A
!B K " L " K . 2 ! L
"
I(1() J1'K1+H)";
&FA
! / K "L " '
"
# K "L . " ! .A VGM=8 " ! =8M
" M 2 " M
$K "L " ! . 2 " ! . A " ! VGM=8 " ! =8M
7( -*/AC -?" */5@ D$-? E"*( #-"/24F#-/-" "**(* )(* / #"--0$&% -$5" (#J ")B D" #"0"A-
=& J [ '1:]#H ._ J \ /&2 .. J M[G
.) -?$# #4#-"5 K2"#$%&"2 )(* / */5@ $&@,-L *"A"$B"# ) ." / '(/#"H -?" *"#@(&#" ?/#
/& (B"*#?((- () [e /&2 / #"--0$&% -$5" () (# J G[ ")BG8:"G

9<ab=7'.67896:;'1<!8;'3<679<;'=c=781
>?)+/* 9@A9& b&$- #-"@ *"#@(&#"W
$K"L &K#L 3K"L #K"L
# K"L . " ! .A !5!7 !@K"L
" M 2
$K " L " ! . 2 " ! . A .A ! / K"L "
!B K" L " K . 2 ! L "
VGM=8 " ! = M
" I(1() J1'K1+H)";
" 8
&FA
" M ! VGM= 8 " ! =8M '
=9,4-2,/456/,
!'$
7+48*9:;,

'<
7?"'#"--0$&%' -$5"'$#'(#J' G[")B

'< ! )*+,-./,01 "

 9<ab=7'.67896:;'1<!8;'3<679<;'=c=781
>?)+/* 9@A9& b&$- */5@ *"#@(&#" )(* -?" #4#-"5 D$-? . %/$&G
.A $K"L &K#L 3K"L #K"L
!B K "L " K . 2 ! L
" !5!7 !@K"L
. .
! / K" L " . ! / K "L "
" !B K" L " K . 2 ! A L "
7?" A(&-*(00"* @/*/5"-"*# /*" "
&FA
I(1() J1'K1+H)";
._ J \ /&2 .. J M[G '

# K "L . K.2" ! .A L
" !
$K " L " M ! . K . 2 " ! . A L 23+4-2,/456/,
. KVGM=8 " ! = L
M '(
" 8
" M ! . KVGM=8 " ! =8M L '%
7?$# #4#-"5 $# B"*4 $,4#." ", '$ >?!
!2)(3 . '( "2 /*)(" /5!7G .) >?!'<
>? '<
& A?/&%"# %/$& #( -?/- . #?$)-# '"
)*(5 . J +4 f[OeH -?" A?/&%" 7?"'#"--0$&%'-$5"'$#'(#J' ")B
$& -?" */5@ *"#@(&#" $# '" '$ '% '( !
)*+,-./,01
$&#$%&$)$A/&-G

9<ab=7'.67896:;'1<!8;'3<679<;'=c=781
>?)+/* 9@A9& b&$- #-"@ *"#@(&#" )(* -?" #4#-"5 D$-? . %/$&G
.A $K"L &K#L 3K"L #K"L
!B K "L " K . 2 ! L
" !5!7 !@K"L
. .
! / K" L " ! / K "L "
" . "
!B K" L " K . 2 ! A L
7?" A(&-*(00"* @/*/5"-"*# " I(1() J1'K1+H)";
/*" ._ J \ /&2 .. J M[G &FA '
!'$
# K"L . K.2 " ! .A L =9,4-2,/456/,
"
$K " L " M ! . K . 2 " ! . A L !'"

. KVGM=8 " ! =8M L !

"
" ! . KVGM=8 " ! =8M L
M
'(
>?!
'% >?!'<
>? '<
7?" A?/&%" $& #-"@ *"#@(&#" $# '$
$&#$%&$)$A/&-G
'"

'< ! !'< " "'<

)*+,-./,01

M
 9<ab=7'.67896:;'1<!8;'3<679<;'=c=781
>?)+/* 9@A9& 7?" *(+,#-&"## )(* -?" #4#-"5 D$-? . %/$&G
$K"L &K#L 3K"L #K"L
!5!7 !@K"L
.A .
!B K" L " K . 2 ! L .A ! / K" L "
"
" !B K" L " K . 2 ! L
. " I(1() J1'K1+H)";
&FA
! / K"L " '
"
D?"*" -?" A(&-*(00"* @/*/5"-"*# /*" ._ J \ /&2 .. J M[G

# K"L . K. 2 " ! . A L . KVGM=8 " ! =8M L

" M " M
$K" L " ! . K . 2 " ! . A L " ! . KVGM=8 " ! =8M L

B .#*" AI& 7?$# #4#-"5 $# B"*4 $,4#." ", !2)(3 . '( "2 /*)(" /5!7G I(*
"Z/5@0"H $) /5!7D&G! A?/&%"# %/$& #( -?/- . #?$)-# )*(5 . J +4 f[OeH -?"
A?/&%" $& -?" */5@ *"#@(&#" $# $&#$%&$)$A/&-G

!8=.>6'<I'9<ab=7'.67896:;'1<!8;'!8=.>6
J .'3( ,0 $,4#." '(" $()* +,- * !,("$,* .;." +
>?)+/* %9@AD&%3(&#$2"*'-?"'#4#-"5'()'I$%,*"'D$-?'#-/-"'B/*$/+0"')""2+/AC'/&2'/
A(5@"&#/-(*'!AK"LG'7?"'@*(+0"5'$#'-('2"#$%&'A(5@"&#/-(*'!AK"L'-('-*/AC'/'#-"@
$&@,-'D$-?'E"*('#-"/24F#-/-"'"**(*G'7?"'2"#$*"2'A0(#"2'0((@'+"?/B$(,*'$#'-(
/A?$"B"'/'#"--0$&%'-$5"'K-('D$-?$&'Me'()'-?"')$&/0'B/0,"L'$&'0"##'-?/&'(#> ")BG8:
/&2'/':)1:+)1(;')"/G8");D?$0"'*"-/$&$&%'/'*(+,#-'*"#@(&#"'-?/-'-?"'-D('@(0"#'()
!@K"L'A/&'A?/&%"'+4'f[OeG'7?"&'-?"'D(*#-FA/#"'A(&2$-$(&'$#

!g / K " L " K " ! OG[L K" ! L

K " ! OG[LK" ! L

$K"L &K#L 3K"L LMK"L #K"L

!)K"L !AK"L
"! "!M L K"L'
!@K"L
7?"'@0/&-'!@K"L'$#Q' 4M
.+
4
! / K"L " .1
K " ! LK" ! ML

V
 !8=.>6'<I'9<ab=7'.67896:;'1<!8;'!8=.>6
>?)+/* 9@AD&
3(&#$2"* -?" #4#-"5 %$B"& $& -?" +0(AC 2$/%*/5G 7?" - )-4 )" $ ./,(. /# / *"#@(&#"
-?/- @*(A""2# */@$204 -( -?" 2"#$*"2 0"B"0 /&2 ?(02# /- -?/- 0"B"0 D$-? 5$&$5/0 (B"*#?((-G
7?" - )-4 )" $ ./,(. ?/# -?" )(00(D$&% A?/*/A-"*$#-$A#W
G'=-"/24F#-/-"'"**(*'J'OH $K"L &K"L 3K"L
#K"L
!)K"L !AK"L !/K"L
MG'I/#-'*"#@(&#"
K5$&$5,5'*$#"'-$5"'/&2'#"--0$&%'-$5"L %K"L
VG'OG e'h'@"*A"&-'(B"*#?((-'hMe GM
*K(L'
XG'_"*A"&-',&2"*#?((-'hMe 8 Me
OGY
3?/*/A-"*$#-$A# KVL /&2 KXL OG`
*"P,$*" -?/- -?" *"#@(&#" OG\
*"5/$& D$-?$& -?" fMe +/&2
#( -?/- -?" "&-*4 -( -?" +/&2 OGX
(AA,*# /- -?" #"--0$&% -$5"G OGM

O'''''''''''OG[(# (# G[(# M(# MG[(# V(#

G'01HK<): (K0)

!8=.>6'<I'9<ab=7'.67896:;'1<!8;'!8=.>6
:,*#"',(%9@AD& .M "M ! .2 " ! . A
d"*"H'D"'#"0"A-'/'2AM A(&-*(00"*')(*'!AK"L9;N"'-?"&'?/B" !B K"L " K "
L
/&2'!AK"L!@K"L'D$00'A(&-/$&'$K"L'J'0]"H'-?"'$&@,-'A(55/&2G
$K"L &K#L 3K"L LMK"L #K"L
!)K"L !AK"L
"! "!M L K"L'
!@K"L
! / K"L " 4M
K " ! LK " ! ML .+
4
.1

6(-" -?/- D" )""2 +/AC +(-? #-/-" B/*$/+0"# /&2 /22 -?"#" /22$-$(&/0 #$%&/0# F.+4M
/&2 F.14 /)-"* !AK"L $& (*2"* -( *"-/$& -?" $&-"%*/-(* $& !AK"LG 7?" A0(#"2 0((@
#4#-"5 $#
$K"L &K#L 3K"L LMK"L #K"L
!)K"L !AK"L
"! "!M L K"L'
4M
.+
! K "L " 4
" M ! K V ! . + L " ! MK ! . + L ! . 1 .1
! K"L

X
 !8=.>6'<I'9<ab=7'.67896:;'1<!8;'!8=.>6
:,*#"',(%9@AD& $K"L $*K"L &K#L #K"L
!)K"L !AK"L ! K"L

! K" L "
" M ! KV ! .+ L " ! MK ! .+ L ! . 1
. M "M ! . 2 " ! .A
!B K " L " K L
"
!"#$%&'2AM A(&-*(00"*'@/*/5"-"*#Q'.!C;._C;/&2'..Q
(7%<&"'2"#$%&'/@@*(/A?'$#'-('- .'3(%"2 %!,("$,*%0,$%"2'.%C,$."K!). %!,(-'"',(G

!g / K " L "
K " ! OG[LK" ! L
)7 :&(-?"* /@@*(/A?H C2'!2 C #. 2 $ H $# -( - .'3( 0,$ "2 (,+'()* /5!7 )(-
,( K2)*0 "2 - .'$ - . ""*'(3 "'+ G 7?"& D" "Z@"A- -( 5""- -?" #"--0$&% -$5"
*"P,$*"5"&- /&2 /--/$& / B"*4 )/#-H ?$%?04 *(+,#- #4#-"5G 6(-" -?/- -?" @*")$0-"*
A(5@"&#/-(* !)K"L $# ,#"2 -( /--/$& -?" 2"#$*"2 )(*5 )(* -?" A0(#"2 0((@ -*/&#)"*
-K"LG

!8=.>6'<I'9<ab=7'.67896:;'1<!8;'!8=.>6
:,*#"',(%9@AD&%!"#$%&'2AM A(&-*(00"*'@/*/5"-"*#Q'.!C;._C;/&2'.. )(*'-?"
&(5$&/0'!@K"L /&2'(&"F?/0)'-?" 2"#$*"2'#"--0$&%'-$5"G
. "M ! .2 " ! . A ! / K"L "
!B K "L " K M L K " ! LK" ! ML
"
7?"'*"#@(&#"'2"#$*"2'$#'2"/2+"/-'7/+0"'MH'#('D"',#"'/'-?$*2F(*2"*'-*/&#)"*'),&A-$(&'
/#'-K"L'$# # K " L = 8V = 8V
" V "
$' K " L " ! $= 8 " M ! B= 8M " ! =8V " V ! GY= 8 " M ! MGM=8M " ! = 8V
1)4* %9&%3("))$A$"&-#'/&2'9"#@(&#"'1"/#,*"#'()'/'!"/2+"/-'=4#-"5'8J\
=8\
- K "L " \
" ! $=8 " ! B=8 " ! A=8V " V ! @=8X " M ! ?=8[ " ! =8\
[ M X

[
 !8=.>6'<I'9<ab=7'.67896:;'1<!8;'!8=.>6
:,*#"',( 9@AD& !"#$%& 2AM A(&-*(00"* @/*/5"-"*#Q .!C ._C /&2 .. )(* -?"
&(5$&/0 !@K"L /&2 (&"F?/0) -?" 2"#$*"2 #"--0$&% -$5"G

$K"L &K#L 3K"L LMK"L #K"L

!)K"L !AK"L
"! "!M L K"L'
!@K"L
! / K"L " 4M
K " ! LK " ! ML .+
4
M
.1
.M" ! .2" ! . A
!B K "L " K L
"
7?"'*"#@(&#"'2"#$*"2''@"*)(*5/&A"')(*'-?"'A0(#"2'0((@'#4#-"5'$#'2"/2+"/-'-K"L'$#

# K"L =8V =8V

" - K"L " V " V
$K " L " ! $= 8 " ! B= 8 " ! = 8 " ! GY=8 " ! MGM= 8M " ! =8V
M M V M

/&2'-?"'#"--0$&%'-$5"'KD$-?'/'Me'A*$-"*$(&L'$#'(# J'XGOX]=&G'I(*'/'#"--0$&%'-$5"'()' ]M'

#"A(&2H'D"',#"'=&' J'`GO`''1:]")BG

!8=.>6'<I'9<ab=7'.67896:;'1<!8;'!8=.>6
:,*#"',( 9@AD& !"#$%& 2AM A(&-*(00"* @/*/5"-"*#Q .!C ._C /&2 .. )(* -?"
&(5$&/0 !@K"L /&2 (&"F?/0) -?" 2"#$*"2 #"--0$&% -$5" (# J XGOX]=& D?"*" =&J`GO`
'1:]")B
$K"L &K#L 3K"L LMK"L #K"L
!)K"L !AK"L
"! "!M L K"L'
/5!7
! / K"L " 4M
K " ! LK" ! ML .+ 4
65!7 .1
.M"M ! .2" ! .A
!B K " L " K L
" $K"L &K#L #K"L
!)K"L !AK"L ! K"L

! K "L "
" M ! KV ! . + L " ! MK ! . + L ! . 1

7?"'A0(#"2F0((@'-*/&#)"*'),&A-$(&'()''-?"'#4#-"5'$#
# K"L ! K " L! K "L =8V
" - K"L " ! N K"L B " V
$K " L ! !B K " L! K "L " ! GY=8 " ! MGM=8M " ! =8V
M

\
 !8=.>6'<I'9<ab=7'.67896:;'1<!8;'!8=.>6
:,*#"',( 9@AD& !"#$%& 2AM A(&-*(00"* @/*/5"-"*#Q .!C ._C /&2 .. )(* -?"
&(5$&/0 !@K"L /&2 (&"F?/0) -?" 2"#$*"2 #"--0$&% -$5" (# J XGOX]=& D?"*" =&J`GO`
'1:]")B &K#L #K"L
$K"L
!)K"L !AK"L ! K"L
! K" L " M
" ! KV ! .+ L " ! MK ! .+ L ! . 1
.M"M ! .2" ! .A
!B K " L " K L
"
7?"'A0(#"2F0((@'-*/&#)"*'),&A-$(&'()'-?"'#4#-"5'()'I$%,*"'$#
=8V .A
- K"L " "
" ! GY=8 " ! MGM=8M " ! =8V
V M
" V ! KV ! . M ! . + L " M ! KM ! . 2 ! . 1 ! M . + L " ! . A
.A
D?"*"'!)K"L ! N K " L " K L
.M "M ! .2" ! .A
N" 0"- .1 J OC .+ J MH ._ J M^G\C .. J [M^G[H /&2 .! J OGV[G 6(-" -?/- -K"L
A(,02 +" /A?$"B"2 D$-? (-?"* %/$&#H $&A0,2$&% .+ J OG

!8=.>6'<I'9<ab=7'.67896:;'1<!8;'!8=.>6
:,*#"',( 9@AD& 7?" #-"@ *"#@(&#" () -?$# #4#-"5 ?/# /& (B"*#?((- () MOG e
K_"/C /5@0$-,2" >@J GML /&2 / #"--0$&% -$5" () (#JOG[ #"A(&2 )(* -?" A(&-*(00"*
@/*/5"-"*#Q .1J OC .+ J MH ._ J M^G\C .. J [M^G[H /&2 .! J OGV[G
$K"L &K#L #K"L
! / K"L " !)K"L !AK"L ! K"L
K " ! LK " ! ML

! K"L " =9,4-2,/456/,

" M ! KV ! . + L " ! MK ! . + L ! . 1 !'$
!'"
. "M ! .2" ! .A
!B K " L " K M L !
"
7+48*9:;,

'(
N$-?(,-'-?"'@*")$0-"*
A(5@"&#/-(*'!)K"L'() '%

'$
.A
! N K"L " K L '"
.M "M ! .2" ! .A ="--0$&%'-$5"'(# J'OG[
'" '$ '% '( ! !'"
)*+,-./,01

^
 !8=.>6'<I'9<ab=7'.67896:;'1<!8;'!8=.>6
:,*#"',( 9@AD& 7?" #-"@ *"#@(&#" () -?$# #4#-"5 ?/# ) - )-4 )" $ ./,(. D$-?
/& (B"*#?((- () G\\e K_"/C /5@0$-,2" >@C GOML /&2 / #"--0$&% -$5" () (#JOG[
#"A(&2 )(* -?" A(&-*(00"* @/*/5"-"*#Q .1 J OC .+ J MH ._ J M^G\C .. J [M^G[H
/&2 .! J OGV[G $K"L &K#L #K"L
!)K"L !AK"L ! K"L
! / K"L "
K " ! LK" ! ML
! K"L " =9,4-2,/456/,
" M ! KV ! . + L " ! MK ! . + L ! . 1 !'$
!'"
.M"M ! .2" ! .A
!B K " L " K L !
"
7+48*9:;,

'(
N$-?'-?"'@*")$0-"*
A(5@"&#/-(*'!)K"L'() '%

'$
.A
! N K"L " K L '"
.M "M ! .2" ! .A ="--0$&%'-$5"'(# J'OG[
'" '$ '% '( ! !'" !'$ !'%
)*+,-./,01

!8=.>6'<I'9<ab=7'.67896:;'1<!8;'!8=.>6
:,*#"',( 9@AD& 3(&#$2"* 2"#$%&"2 A(&-*(00"* /&2 D$-? -?" #/5" @/*/5"-"*# )(*
-?" D(*#-FA/#" A(&2$-$(&Q 1 J OG[ /&2 + J G
$K"L &K#L 3K"L LMK"L #K"L
!)K"L !AK"L
"!1 "!+ L K"L'
!@K"L
!g / K " L " 4M
K " ! OG[LK " ! L .+
4
.M"M ! .2" ! .A .1
!B K " L " K L
"
$K"L &K#L #K"L
!g K "L "
!)K"L !AK"L !g K " L
" M ! K 1 ! + ! . + L " ! +K 1 ! . + L ! . 1
N$-?'-?"'@*")$0-"* A(5@"&#/-(*'!)K"L'()
.A
! N K"L " K L
.M "M ! .2 " ! .A

`
 !8=.>6'<I'9<ab=7'.67896:;'1<!8;'!8=.>6
:,*#"',( 9@AD& 3(&#$2"* 2"#$%&"2 A(&-*(00"* /&2 D$-? -?" #/5" @/*/5"-"*# )(*
-?" D(*#-FA/#" A(&2$-$(&G $K"L &K#L #K"L
!)K"L !AK"L !g K " L
!g K "L "
" M ! K 1 ! + ! . + L " ! +K 1 ! . + L ! . 1

.A
! N K "L " K M
L !'$
.M" ! .2" ! .A
!'" 1J H'+JM
. "M ! .2" ! .A 1JOG[H+J
!B K " L " K M L !
"
7+48*9:;,

'(

!g / K " L " '%

K " ! OG[LK" ! L
'$
="--0$&%'-$5"' ="--0$&%'-$5"'
'"
(# J'OG[ (# J'OGY[
'" '$ '% '( ! !'" !'$ !'%
)*+,-./,01

!8=.>6'<I'9<ab=7'.67896:;'1<!8;'!8=.>6
:,*#"',( 9@AD& b&$- #-"@ @"*)(*5/&A" () -?" A(&-*(00"* )(* -?" D(*#-FA/#"
A(&2$-$(&G $K"L &K#L
g
#K"L
!)K"L !AK"L ! K "L
!g K "L " M
" ! K 1 ! + ! . + L " ! + K1 ! . + L ! . 1
!'$

.A !'" 1J H'+JM
! N K "L " K M
L 1JOG[H+J
.M" ! .2" ! .A !
7+48*9:;,

'(
.M "M ! . 2" ! . A
!B K " L " K L '%
"
'$
!g / K " L " ="--0$&%'-$5"' ="--0$&%'-$5"'
'"
K " ! OG[LK" ! L (# J'OG[ (# J'OGY[
'" '$ '% '( ! !'" !'$ !'%
)*+,-./,01
B .#*" AL& N?"& -?" @(0"# () !@K"L A?/&%" +4 f[OeH -?" (B"*#?((- A?/&%"# -(
G`^eH K_"/C /5@0$-,2" >@C GOML /&2 -?" #"--0$&% -$5" +"A(5"# OGY[ #"A(&2G 7?$#
$# /& (,-#-/&2$&% 2"#$%& () / B"*4 *(+,#-H 2"/2+"/- *"#@(&#" #4#-"5G

Y
 7d8'_=8b!<Fib:67.7:7.j8'I88!a:3k'=c=781

i,/&-$-/-$B" )""2+/AC -?"(*4 KiI7L ,#"# / A(&-*(00"*H /# #?(D& $& I$%,*"H -(

/A?$"B" *(+,#- @"*)(*5/&A"G

$K"L &K#L #K"L

. !AK"L !@K"L

7?" %(/0 $# -( /A?$"B" / D$2" +/&2D$2-? )(* -?" A0(#"2 0((@ -*/&#)"* ),&A-$(&
D$-? / ?$%? 0((@ %/$& .9 74@$A/0 iI7 2"#$%& 5"-?(2# ,#" %*/@?$A/0 /&2
&,5"*$A/0 5"-?(2# $& A(&U,&A-$(& D$-? )*"P,"&A4 *"#@(&#" 5"-?(2# Ka(2"
2$/%*/5 (* -?" 6$A?(0# A?/*-LG
>"&"*/004H iI7 2"#$%& #""C# ) 2'32 *,,/ 3)'( /&2 *)$3 /2). +)$3'( #( -?/-
*(+,#- @"*)(*5/&A" $# /A?$"B"2G

7d8'_=8b!<Fib:67.7:7.j8'I88!a:3k'=c=781
d"*"'D"'@,*#,"'/'#$5@0"'5"-?(2'()'/A?$"B$&%'-?"'%(/0#'()'iI7'D$-?'/&'"F@0/&"H'
*((-'0(A,#'/@@*(/A?'-('-?"'#"0"A-$(&'()'-?"'%/$&'.;/&2'-?"'A(5@"&#/-(*'!AK"LG'

$K"L &K#L #K"L

. !AK"L !@K"L

:##,5" !@K"L $# 5$&$5,5 @?/#" #4#-"5G 7?"& -?$# /@@*(/A?H 2,++"2 @#",2(F
iI7H )(00(D# -?"#" #-"@#W
6@%_0/A"'-?"'8;@(0"#'/&2'0;E"*(#'()'!@K"L'(&'-?"'"F@0/&"')(*'-?"'8-?'(*2"*'!@K"LG'
:0#(H'/22'/&4 @(0"# ()'!AK"LG
9@%=-/*-$&%'&"/*'-?"'(*$%$&H'@0/A"'-?"'E"*(#'()'!AK"L'$55"2$/-"04'-('-?"'0")-'()'"/A?'
()'-?"'K8 F L'@(0"#'(&'-?"'0")-F?/&2'"F@0/&"G'7?$#'0"/B"#'(&"'@(0"')/*'-('-?"'0")-'()'
-?"'0")-'?/&2'#$2" ()'-?" "F@0/&"G
D@%.&A*"/#"'-?"'%/$&'.;#('-?/-'-?"'*((-#'()'-?"'A?/*/A-"*$#-$A'"P,/-$(&'K@(0"#'()'-?"
A0(#"2F0((@'-*/&#)"*'),&A-$(&L'/*"'A0(#"'-('-?"'E"*(#'()'!AK"L!@K"LG

MO
 7d8'_=8b!<Fib:67.7:7.j8'I88!a:3k'=c=781
7?$#'5"-?(2'$&-*(2,A"#'E"*(#'#('-?/-'/00'+,-'(&"'()'-?"'*((-'0(A$'"&2'(&')$&$-"'
E"*(#G
$K"L &K#L #K"L
. !AK"L !@K"L

.) -?" %/$& . $# #,))$A$"&-04 0/*%"H -?"& -?" @(0"# () -?" A0(#"2 0((@ -*/&#)"*
),&A-$(& -K"L /*" /05(#- "P,/0 -( -?" E"*(# () !AK"L!@K"LG 7?$# 0"/B"# ,( /,* ()
-K"L D$-? ) .'3('0'!)(" /)$"')* 0$)!"',( $ .'-# /&2 -?" #4#-"5 D$-? / @?/#"
5/*%$& () /@@*(Z$5/-"04 YOl K/A-,/004 /+(,- `[lLG

!8=.>6'9<ab=7'3<679<;'b=.6> ib:67m7:7mj8'
I88!a:3k'7d8<9c'KiI7L'
>?)+/* %9@AE&%3(&#$2"*'-?"'#4#-"5'()'I$%,*"'D$-?
$K"L &K#L #K"L
. !AK"L !@K"L
! / K "L "
K " ! / LK " ! /M L

D?"*"'-?"'&(5$&/0'A/#"'$#'/ J' '/&2 /M J'MH'D$-?'f[Oe'B/*$/-$(&G'7?"'D(*#-'A/#"'$#'

D$-?'/ D;E9O;18:;/M J' G'N"'D$#?'-('2"#$%&'-?"'#4#-"5')(*'E"*('#-"/24F#-/-"'"**(*')(*'
/'#-"@'$&@,-H'#('D"',#"'-?"'2AM A(&-*(00"*

. M " M ! . 2 " ! . A K " ! < LK " ! <M L

!B K "L " "
" "
N"'-?"&'$&B(C"'-?"''(" $()*%+,- *%/$'(!'/* H'D$-?'$K"L'J' ]";$&A(*@(*/-"2'D$-?$&
!AK"L!@K"LG'

M
 !8=.>6'9<ab=7'3<679<;'b=.6> ib:67m7:7mj8'
I88!a:3k'7d8<9c'KiI7L'
>?)+/* %9@AE&%7?"'&(5$&/0'@/*/5"-"*#'()'-?"'@0/&-'/*"'/ J' '/&2 /M J'MH'D$-?
f[Oe'B/*$/-$(&G'7?"'D(*#-'A/#"'$#'D$-?'/ D;E9O;18:;/M J' G
! / K "L " $K"L &K#L #K"L
K " ! / LK" ! /M L . !AK"L !@K"L

K " ! < LK " ! < M L

!B K "L "
"
b#$&%':" /%6H'D"'@0/A"'-?"'@(0"#'()'!AK"L!@K"L'(&'-?"'"F@0/&"H'/#'#?(D&'$&'I$%,*"G'
7?"*"'/*"'-?*""'@(0"#'K/-'" J'O'H'F H'FMLH'/#'#?(D&G'
@=

EM'J'F G` E JFOG`

E D
/M'J'FM / 'J'F
"F/H18)

!8=.>6'9<ab=7'3<679<;'b=.6> ib:67m7:7mj8'
I88!a:3k'7d8<9c'KiI7L'
>?)+/* %9@AE&%7?"'&(5$&/0'@/*/5"-"*#'()'-?"'@0/&-'/*"'/ J' '/&2 /M J'MH'D$-?
f[Oe'B/*$/-$(&G'7?"'D(*#-'A/#"'$#'D$-?'/ J'OG['18:;/M J' G
! / K "L " $K"L &K#L #K"L
K " ! / LK" ! /M L . !AK"L !@K"L

K " ! < LK " ! < M L

!B K "L "
"
:" / 9 A/00# )(* @0/A$&% / E"*( -( -?" 0")- () -?" @(0" /- -?" (*$%$& " J O /&2 /- -?"
@(0" /- " J F /# #?(D& $& I$%,*"G
@=

EM'J'F G` E JFOG`

E D
/M'J'FM / 'J'F
"F/H18)

MM
 !8=.>6'9<ab=7'3<679<;'b=.6> ib:67m7:7mj8'
I88!a:3k'7d8<9c'KiI7L'
>?)+/* 9@AE& 7?" &(5$&/0 @/*/5"-"*# () -?" @0/&- /*" / J /&2 /M J MH D$-?
f[Oe B/*$/-$(&G 7?" D(*#- A/#" $# D$-? / J OG[ 18: /M J G
! / K "L " " $K"L &K#L #K"L
K" ! / LK" ! /M L K " ! LK " ! ML . !AK"L !@K"L

7?" A(5@"&#/-(* $#'-?,#

K " ! OG`LK " ! G`L
!B K "L " @=
" EM'J'F G` E JFOG`

E D
/M'J'FM / 'J'F
"F/H18)
N" #"0"A- . J OOH #( -?/- -?" *((-# () -?" A?/*/A-"*$#-$A "P,/-$(&
/*" A0(#" -( -?" E"*(#G 7?" A0(#"2F0((@ -*/&#)"* ),&A-$(& $#

# K"L .!B K" L! K " L OOK " ! OG`LK" ! G`L OO

" - K"L " " E
$K " L ! .!B K" L! K " L K " ! OG^Y`LK" ! G^Y^LK" ! OOGXL K " ! OOL

!8=.>6'9<ab=7'3<679<;'b=.6> ib:67m7:7mj8'
I88!a:3k'7d8<9c'KiI7L'
>?)+/* %9@AE&%7?"'D(*#-'A/#"'$#'D$-?'/ J'OG['/&2 /M J' G
$K"L &K#L #K"L
# K"L .!B K " L! K " L OO . !AK"L !@K"L
" E
$K " L ! .!B K " L! K " L K " ! OOL

7?$#'A0(#"2F0((@'#4#-"5'@*(B$2"#'/')/#-'*"#@(&#"'/&2'@(##"##"#'/'@?/#"'5/*%$&'
()'/@@*(Z$5/-"04'`[lG'7?"'@"*)(*5/&A"'$#'#,55/*$E"2'$&'7/+0"'VG
1)4* %D&%_"*)(*5/&A" ()'_#",2(FiI7'!"#$%&
_"*A"&- <B"*#?((- ="--0$&% 7$5"
6(5$&/0'!@K"L OGO e XO'5#
N(*#-FA/#" !@K"L OGY^e XO'5#

B .#*"%AM& N?"&'-?"'D(*#-FA/#"'A(&2$-$(&#'/*"'*"/0$E"2'K/ J'OG['/&2 /M J' L'-?"'

@"*)(*5/&A"'*"5/$&#'"##"&-$/004',&A?/&%"2H'/#'#?(D&'$&'7/+0"'VG'
_#",2(FiI7'2"#$%&'*"#,0-#'$&'B"*4'*(+,#-'#4#-"5#G

MV
 Lecture 10: Introduction to Hinf Robust Control

Systems

Introduction to Hinf Robust Control Systems

The first step in the design of a control system is to obtain a mathematical model of
the control object based on the physical law. Quite often the model may be
nonlinear and possibly with distributed parameters. Such a model may be difficult
to analyze.
Assume the system is modelled by differential equation;

x  Rn

x  f ( x, u ),  u  R r
 f : R n  Rr  Rn

The system output:
 y  R m
y  g ( x, u )  
 g : R m  R r  R m

where x(t) is the system state, u(t) and y(t) are the system input output respectively,
f(x.u) and g(x.u) are smooth real valued nonlinear functions of x and u.

1
 Introduction to Hinf Robust Control Systems
The actual system model is nonlinear and given by a nonlinear differential
equation;
x  f ( x, u )
The system output: y  g ( x, u )

It is desirable to approximate it by a linear constant-coefficient system for an

equilibrium or operation points that will approximate the actual object fairly well.
x  Ax  Bu Y (s ) 1
C B  D  C ( sI  A) 1 B  D
y  Cx  Du U (s ) ( sI  A)
Note that even though the model to be used for design purposes may be a
simplified one, it is necessary that such a model must include any intrinsic
character of the actual object.
We must get a simplified model for the purpose of designing the control system
that will require a compensator of lowest order possible.

Introduction to Hinf Robust Control Systems

The simplified models of the system to design compensator with the possible
lowest order can be given in;

State space model Transfer function model

x  Ax  Bu Y (s ) 1
 G p (s )  C BD
y  Cx  Du U (s ) ( sI  A)
 C ( sI  A) 1 B  D

Thus, a model of a control object (whatever it may be) will probably include an
error in the modeling process. Note that in the frequency-response approach to
control systems design, we use phase margin PM and gain margin GM, to take
care of the modeling errors.
However, in the state-space approach, which is based on the differential
equations of the plant dynamics, no such “margins” are involved in the design
process.

2
 Introduction to Hinf Robust Control Systems
Since the actual plant G(s) differs from the model, Gp(s) used in the design, a
question arises whether the controller Gc(s) designed for the following feedback
system using a model Gp(s) will work satisfactorily with the actual plant G(s) .
Unpredicted disturbance input, D(s)
R(s) E(s) U(s) Y(s)
Gc(s) Gp(s)
Desired Actual
Controller Plant, System output
input
Measurement output H(s)
Sensor Sensor noise, N(s)
To ensure that it will do so, robust control theory has been developed since around
1980.
Robust control theory uses the assumption that the models we use in designing
control systems have modelling errors. Basically, the robust control theory assumes
that there is an uncertainty Gp(s) or error between the actual plant G(s) and its
mathematical model Gp(s) and includes such uncertainty or error in the design
process of the control system.

Introduction to Hinf Robust Control Systems

Unpredicted disturbance input, D(s)
R(s) E(s) U(s) Y(s)
Gc(s) Gp(s)
Desired Actual
Controller Plant, System output
input
H(s)
Measurement output
Sensor Sensor noise, N(s)

Systems designed based on the robust control theory will possess the following
properties:
(1) Robust stability. The control system designed is stable in the presence of
perturbation Gp(s) .
(2) Robust performance. The control system exhibits predetermined response
characteristics in the presence of perturbation Gp(s) .

This theory requires considerations based on frequency-response analysis and

time domain analysis.

3
 Introduction to Hinf Robust Control Systems

Uncertain Elements in Plant Dynamics. The term uncertainty refers to the

differences or errors between the model of the plant Gp(s) and the actual plant G(s).
Uncertain elements that may appear in practical systems may be classified as
• structured uncertainty and
• unstructured uncertainty.
An example of structured uncertainty is any parametric variation in the plant
dynamics, such as variations in poles and zeros of the plant transfer function.
Examples of unstructured uncertainty include frequency dependent uncertainty,
such as high-frequency modes that we normally neglect in modeling plant
dynamics. For example, in the modeling of a flexible-arm system, the model may
include a finite number of modes of oscillation. The modes of oscillation that are
not included in the modeling behave as uncertainty of the system. Another example
of uncertainty occurs in the linearization of a nonlinear plant. If the actual plant is
nonlinear and its model is linear, then the difference acts as unstructured
uncertainty.

Introduction to Hinf Robust Control Systems

Uncertain Elements in Plant Dynamics. Unpredicted disturbance
input, D(s)
R(s) E(s) U(s) Y(s)
Gc(s) Gp(s)+Gp(s)
Controller System G(s)
H(s)
Sensor Sensor noise, N(s)

Here we consider the case where the uncertainty is unstructured. In addition we

assume that the plant involves only one uncertainty. (Some plants may involve
multiple uncertain elements.)
In the robust control theory, we define unstructured uncertainty as Gp(s). Since
the exact description of Gp(s) is unknown, we use an estimate of Gp(s) (as to
the magnitude and phase characteristics) and use this estimate in the design of the
controller Gc(s) that stabilizes the control system.
Stability of a system with unstructured uncertainty can then be examined by use
of the small gain theorem to be given following the definition of the H norm.

4
 Introduction to Hinf Robust Control Systems: H norm
H norm: The H norm of a stable single-input–single-output system is the largest
possible amplification factor of the steady-state response to sinusoidal excitation.
D(s)
R(s) E(s) U(s) Y(s)
Gc(s) Gp(s)

H(s)
N(s)
For a scalar (s), ||(s)|| gives the maximum value of |(s)|. It is called the H
norm and is illustrated in the following Figure;

The maximum value of |(j)|

R(s) Y(s)
|(j)| in dB
(s)

Figure: Bode diagram of the ||(j)||

transfer function and the H
norm ||(j)||


Introduction to Hinf Robust Control Systems: H norm

In robust control theory we measure the magnitude of the transfer function by the
H norm. Assume that the transfer function (s) is stable and proper, that is, a
transfer function with same number of zeros as poles. A transfer function (j) is
called proper if  () is limited and definite. If  () = 0, it is called strictly proper.
The H norm of is defined by
 
   ( j )
  ( j ) means the maximum singular value of [ (j)].
 means max.
Note that the singular value of a transfer function  (j) is defined by

 i  i (  *  )

where is the i(* ) ith largest eigenvalue of  * and it is always a non-negative

real value.

5
 Introduction to Hinf Robust Control Systems: H norm
By making || (s)|| smaller, we make the effect of input w on the output z smaller.

w z
(s)

It is frequently the case that instead of using the maximum singular value || (s)||,
we use the inequality

 


and limit the magnitude of  (s) by .

To make the magnitude of || (s)|| small, we choose a small  and require that
|| (s)|| < .

Introduction to Hinf Robust Control Systems: H norm

Small-Gain Theorem. Consider the closed-loop system shown in Figure. In the
Figure (s) and M(s) are stable and proper transfer functions.
(s)
Figure: Closed-loop system.
M(s)

The small-gain theorem states that if

 (s )M (s ) 
1
then this closed-loop system is stable. That is, if the H norm of (s)M(s) is
smaller than 1, this closed-loop system is stable. This theorem is an extension of
the Nyquist stability criterion.

It is important to note that the small-gain theorem gives a sufficient condition for
stability. That is, a system may be stable even if it does not satisfy this theorem.
However, if a system satisfies the small-gain theorem, it is always stable.

6
 Introduction to Hinf Robust Control Systems
System with Unstructured Uncertainty: In some cases an unstructured
uncertainty error may be considered multiplicative such that
G ( s )  G p ( s )[1   m ( s )] m(s)
where G(s) is the true plant dynamics and
Gp(s) is the model plant dynamics. Gp(s)
G(s)
In other cases an unstructured uncertainty error may be considered additive
such that
a(s)
G ( s )  G p ( s )   a ( s) Gp(s)
G(s)
In either case we assume that the norm of m(s) or a(s) is bounded such that
 m (s)   m ,  a (s )   a
wherem or a are positive constants.

Introduction to Hinf Robust Control Systems

Example 10.01: Consider the control system given in Figure with unstruc-
tured multiplicative uncertainty.

G ( s )  G p ( s )[1   m ( s )]
m(s)

where G(s) is the true plant dynamics, U(s) Y(s)

Gp(s)
Gp(s) is the model plant dynamics, G(s)
m(s) is unstructured multiplicative
uncertainty and K(s) is the controller. K(s)

Assume that m(s) is stable and its upper bound is known.

Obtain robust stability and robust performance conditions of the closed loop
system.

7
 Introduction to Hinf Robust Control Systems
Solution 10.01: Robust Stability; Let us examine the transfer function be-
tween point A and point B. Notice that Figure (a) can be redrawn as shown in
Figure (b).The transfer function between point A and point B can be given by
B m(s) A
B ( s) K ( s )G p ( s )
 U(s) Y(s)
A( s ) 1  K ( s )G p ( s ) Gp(s)
K ( s )G p ( s ) Figure (a)
T (s )  K(s)
1  K ( s )G p ( s )
A m(s) B
Using the transfer function instead of Y(s) U(s)
K(s) Gp(s)
dash lined part of Figure (b).
we can redraw Figure (b) as Figure (c). Figure (b)

A m(s) B

T(s)
Figure (c)

Introduction to Hinf Robust Control Systems

Solution 10.01: Robust Stability; Applying the small gain theorem to the sys-
tem consisting of m(s) and T(s) as shown in Figure (c), we obtain the condition
for stability to be
m(s)
 ( s )T ( s )  1 (1) B A
m 
Figure (c) T(s)
In general, it is impossible to precisely model m(s). Therefore, let us use a scalar
transfer function Wm(j) such that
 { m ( j )} | Wm ( j ) |
where  { m ( j  )} is the largest singular value of m (j).
Consider, instead of Inequality (1), the following inequality:
Wm ( j )T ( j ) 
 1 (2)
If Inequality (2) holds true, Inequality (1) will always be satisfied. By making the
H norm of Wm(j)T(j ) to be less than 1, we obtain the controller K(s) that will
make the system stable.

8
 Introduction to Hinf Robust Control Systems
Solution 10.01: Robust Stability; Suppose that we cut the line at point A in Fig-
ure (a).Then we obtain Figure (d). z w
B m(s) A m(s) A
U(s) Y(s) U(s) B Y(s)
Gp(s) Gp(s)
Figure (a) Figure (d)
K(s) K(s)

Replacing m(s) by Wm(s)I , we obtain Figure (e). Redrawing Figure (e), we

obtain Figure (f). Figure (f) is called a generalized plant diagram.

z w
w z
P(s)
Wm(s)I y u
u Gp(s) y y
K(s)
y
K(s) Figure (f): generalized plant diagram
Figure (e)

Introduction to Hinf Robust Control Systems

Solution 10.01: Robust Stability;
The transfer function of the system in Figure (a) B m(s) A
U(s) Y(s)
K ( s )G p ( s ) Gp(s)
T (s) 
1  K ( s )G p ( s ) Figure (a)
K(s)
The condition for stability is
Wm ( j )T ( j ) 
1 (2)
z w
Then Inequality (2) can be rewritten as P(s)
y u
K ( j )G p ( j ) y
Wm ( j  ) 1 (3) K(s)
1  K ( j ) G p ( j  )
 Figure (c): generalized plant diagram

Inequality (3) is called the condition for

stability of the feedback system.

9
 Introduction to Hinf Robust Control Systems
Solution 10.01: Robust Stability; The condition for stability of the feed-
back system is
B m(s) A
K ( j )G p ( j ) U(s) Y(s)
Wm ( j  )  1 (3) Gp(s)
1  K ( j )G p ( j ) Figure (a)

K(s)
z w
Clearly, for a stable plant model Gp(s), K(s)=0 P(s)
y u
will satisfy Inequality (3). However, K(s)=0 is
not the desirable transfer function for the y
K(s)
controller.
Figure (c): generalized plant diagram

To find an acceptable transfer function for K(s), we may add another condition, for
example, that the resulting system will have robust performance such that the
system output follows the input with minimum error, or another reasonable
condition.

Introduction to Hinf Robust Control Systems

Solution 10.01: Robust Performance. Consider the system shown in the follow-
ing Figure. Suppose that we want the output y(t) to follow the input r(t) as closely
as possible, or we wish to have
R(s) E(s) U(s) Y(s)
K(s) Gp(s)
lim[r (t )  y (t )]  lim e (t )  0
t  t 

Y (s) K ( s )G p ( s )
Since the transfer function Y(s)/R(s) is  T (s ) 
R( s) 1  K ( s )G p ( s )
From the feedback system we have
E (s ) R( s )  Y ( s ) Y (s ) 1
  1 
R (s ) R (s ) R ( s ) 1  K ( s )G p ( s )
1
Define S (s ) 
1  K ( s )G p ( s )
where S(s) is the sensitivity function and T(s) is called the complementary
sensitivity function.

10
 Introduction to Hinf Robust Control Systems
Solution 10.01: Robust Performance. In this robust performance problem we
want to make the H norm of S(s) smaller than the desired transfer function Ws-1(s)
or ||S(s)|| < Ws-1(s) which can be written as
R(s) E(s) U(s) Y(s)
|| Ws ( s ) S ( s ) ||  1 K(s) Gp(s)
Combining Inequalities
Wm ( j )T ( j ) 
1 obtained for the stability with above we get
Wm ( j )T ( j ) K ( s )G p ( s )
1 Wm ( s )
W s ( j ) S ( j  )  1  K ( s )G p ( s )
1 (4)
where T(j) + S(j) = 1, or 1
Ws ( s )
1  K ( s )G p ( s )


The problem then becomes to find K(s) that will satisfy Inequality (4). Note that
depending on the chosen Wm(s) and Ws(s) there may be many K(s) that satisfy
Inequality (4), or may be no K(s) that satisfies Inequality (4). Such a robust control
problem using Inequality (4) is called a mixed-sensitivity problem.

Introduction to Hinf Robust Control Systems

Solution 10.01: Robust Performance. Figure (a) is a generalized plant dia-
gram, where two conditions (robust stability and robust performance) are
specified. A simplified version of this diagram is shown in Figure (b).
z1
w z2 z
K ( s )G p ( s ) Wm(s)I Ws(s)I
Wm ( s )
1  K ( s )G p ( s ) y
1 u Gp(s)
1 y
Ws ( s )
1  K ( s )G p ( s ) K(s)
 Figure (a)

z w
P(s)
y u
y
K(s)
Figure (b): generalized plant diagram

11
 Introduction to Hinf Robust Control Systems
Finding Transfer Function z(s)/w(s) from a Generalized Plant Diagram
Consider the generalized plant diagram shown in
Figure. In this diagram w(s) is the exogenous P(s)
disturbance and u(s) is the manipulated variable.  P11 P12 
z(s) is the controlled variable and y(s) is the z   w
observed variable.  
y  P21 P22 
u
Consider this control system consisting of the
y
generalized plant P(s) and the controller K(s). K(s)
The equation that relates the outputs z(s) and
Generalized Plant Diagram
y(s) and the inputs w(s) and u(s) of the
generalized plant P(s) is
 z (s )   P11 P12   w(s )
 y (s )   P P   u (s ) 
   21 22   
The equation that relates u(s) and y(s) is given by u ( s )  K ( s ) y ( s )
Define the transfer function that relates the controlled variable z(s) to the
exogenous disturbance w(s) as (s).Then z ( s)   ( s) w( s)

Introduction to Hinf Robust Control Systems

Finding Transfer Function z(s)/w(s) from a Generalized Plant Diagram

The transfer function relates the controlled P(s)

variable z(s) to the exogenous disturbance w(s)  P11 P12 
is (s) and can be determined as follows: z   w
Since  
y  P21 P22 
u
z ( s)  P11w(s )  P12u ( s)
y ( s)  P21w( s)  P22u ( s) y
K(s)
u ( s )  K ( s ) y ( s)
Generalized Plant Diagram
we obtain y ( s )  P21 w( s )  P22 K ( s ) y( s )
Hence [ I  P22 K ( s)] y ( s)  P21 w( s) or y( s )  [ I  P22 K ( s )]1 P21w( s )

Therefore, z ( s)  P11w(s)  P12 K (s )[ I  P22 K (s)]1 P21w( s)

 {P11  P12 K (s)[ I  P22 K ( s)]1 P21}w(s)
z (s )
Hence, z ( s)   ( s) w( s)  (s )  P11  P12 K (s )[I  P22 K (s)]1 P21
w( s)

12
 Introduction to Hinf Robust Control Systems
Example 10.02: Consider the control system given in Figure with unstruc-
tured multiplicative uncertainty.

G ( s )  G p ( s )[1   m ( s )]
m(s)
where G(s) is the true plant dynamics, U(s) Y(s)
Gp(s) is the model plant dynamics, Gp(s)
G(s)
m(s) is unstructured multiplicative
uncertainty and K(s) is the controller.
K(s)
Assume that m(s) is stable and its upper bound is known.

Determine the P(s) matrix in the Generalized Plant Diagram of the control
system.

Introduction to Hinf Robust Control Systems

Solution 10.02: We derived Inequality (1) for the control system to be ro-
bust stable.
Robust Stability Inequality is

K ( j  )G p ( j  )
Wm ( j ) 1 (1)
1  K ( j  )G p ( j  )

B m(s) A
If we define U(s) Y(s)
Gp(s)

K ( j ) G p ( j  ) Figure (a)
 1 ( j )  W m ( j ) K(s)
1  K ( j  )G p ( j  )

Then Robust Stability Inequality is  1 ( j ) 

1

13
 Introduction to Hinf Robust Control Systems
Solution 10.02: Robust Stability Inequality is  1 ( j )   1
(1) K ( j ) G p ( j  )
 1 ( j )  W m ( j ) (2) P(s)
where  K j G p j w
z  P11 P12 
Referring to the transfer function 1 ( relates
) ( the) controlled P P 
variable z(s) to the exogenous disturbance w(s) is (s) y  21 22  u
and is rewritten as y
z (s ) 1
K(s)
  (s )  P11  P12 K (s )[ I  P22 K (s)] P21
w( s) Generalized Plant Diagram
notice that if we choose the generalized plant P(s) matrix as P(s)
0 W G  z  0 W mG p  w
Then we obtain P
m p
I  G 

z (s) I  G p  y  p  u
 ( s )  P11  P12 K ( s)[ I  P22 K ( s )]1 P21
w( s ) y
1
K(s)
 Wm K ( s )G p ( s )[ I  K ( s )G p ( s )]
Generalized Plant Diagram
which is exactly the same as  1(s) in Equation (2).

Introduction to Hinf Robust Control Systems

Solution 10.02: Robust Stability Inequality is  1 ( j )   1
(1) K ( j ) G p ( j  )
 1 ( j )  W m ( j ) (2) P(s)
where  K j G p j z w
0 W mG p 
Then Generalized Plant Diagram 1 ( is ) obtained
( ) and I  G 
y  p  u
z (s)
 Wm K ( s )G p ( s )[ I  K ( s )G p ( s )]1 y
w( s ) K(s)
Generalized Plant Diagram
Robust Performance is derived in Example 10.01 that if we wished to have
the output y follow the input r as close as possible, we needed to make the H
norm of  2(s) , where
R(s) E(s) U(s) Y(s)
1 K(s) Gp(s)
 2 ( s )  Ws ( s )
1  K ( s )G p ( s )
less than 1 or
|| Ws ( s ) S ( s ) ||   1 ||  2 ( j ) ||  1 (3)

14
 Introduction to Hinf Robust Control Systems
P(s)
Solution 10.02: Robust Performance; From Generalized
Plant Diagram Note that the controlled variable z is z  P11 P12  w
P P22 
related to the exogenous disturbance w by y  21 u
z(s)  (s)w(s) y
1 K(s)
where  (s)  P11  P12 K ( s)[ I  P22 K ( s)] P21
Generalized Plant Diagram
notice that if we choose the generalized plant P(s) matrix as W s W sG p 
P
Then z ( s)
  ( s)  P11  P12 K ( s)[ I  P22 K ( s)]1 P21  I  G p 
w( s ) P(s)
Ws K ( s )G p ( s) K (s )G p ( s ) z w
 Ws   Ws [1  W s W sGp 
[ I  K ( s)G p (s )] [ I  K ( s)G p ( s)] I  Gp 
y  u
 1 
 Ws   y
 I  K ( s )G p ( s )  K(s)
Generalized Plant Diagram
which is exactly the same as  2(s) in Equation (3).

Introduction to Hinf Robust Control Systems

Solution 10.02: If both the robust stability K ( s )G p ( s )
Wm ( s )
1  K ( s )G p ( s )
and robust performance conditions are required, 1
the control system must satisfy the condition 1
Ws ( s )
given by Inequality 1  K ( s )G p ( s )


For the P(s) matrix, we combine P(s)

z1  P11 P12 
matrix for Robust Performance; z z  w
2  P21 P22 
W s W sG p  u
P y  P31 P32 
 I  G p 
y
K(s)
and P(s) matrix for Robust Stability;
0 W mG p 
P  z1 W s W sG p 
I  G p  z z   w
2  0 W mG p 
and get W s W sG p   I  G p  u
  y 
P 0 W mG p  y
 I K(s)
  G p 

15
 Introduction to Hinf Robust Control Systems
Solution 10.02: P(s) matrix W s W sG p 
P W s W sG p 
for Robust Performance;  I  G p   
P 0 W mG p  (1)
P(s) matrix for Robust 0 W mG p   I  G p 
P  
Stability; I  G p 
If we construct P(s) as given by Equation (1), then the problem of designing a
control system to satisfy both robust stability and robust performance
conditions can be formulated by using the generalized plant represented by
Equation (1). Such a problem is called a mixed-sensitivity problem. By using
the generalized plant given by Equation (1) we are able to determine the
controller K(s) that satisfies Inequality (2).
z1
w z z
2
K ( s )G p ( s ) Wm(s)I Ws(s)I
Wm ( s )
1  K ( s )G p ( s ) u Gp(s) y
1 (2)
1
Ws ( s ) y
1  K ( s )G p ( s )  K(s)
Figure (a)

Introduction to Hinf Robust Control Systems

Solution 10.02: Both robust stability and robust performance conditions are
formulated by using the generalized plant represented by Equation (1). From
generalized plant diagram or Equation (1) the controller K(s) is determined to
satisfies Inequality (2). K ( s )G ( s ) p
Wm ( s)
W s W sG p  1  K ( s )G p ( s )
  1 (2)
P 0 W mG p  (1) 1
Ws ( s)
 I  G p  1  K ( s )G p ( s )
 

The generalized plant diagram for the system considered in Figure (a) becomes
as shown in Figure.
z1 z1 W s W sG p 
z z z   w
w 2 z2  0 W mG p 
Wm(s)I Ws(s)I u
y  I  G p 
y 
u Gp(s) y
y K(s)
K(s) Figure (b): Generalized Plant Diagram
Figure (a)

16
 Hinf Robust Control Problem
H Infinity Control Problem. To design a controller K(s) of a control system to
satisfy various stability and performance specifications, we utilize the concept of
the generalized plant Figure (b). In this diagram w(s) is the exogenous disturbance
and u(s) is the manipulated variable. z(s) is the controlled variable and y(s) is the
observed variable.

z1 z
P(s) w w z2
z
Generalized Wm(s)I Ws(s)I
y plant u u y
Gp(s)
y y
K(s) K(s)
Figure (b): Generalized Plant Diagram Figure (a): Feedback system
A generalized plant is a linear model consisting of a model of the plant and
weighting functions, Ws and Wm corresponding to the specifications for the
required performance.

17
 Hinf Robust Control Problem
H Infinity Control Problem. Referring to the generalized plant shown in Figure
(b), the H infinity control problem is a problem to design a controller K(s) that will
make the norm of the transfer function from the exogenous disturbance w to the
controlled variable z less than a specified value.
z1 z
w z2
z P(s) w Wm(s)I Ws(s)I
Generalized
y plant u u Gp(s) y
y y
K(s) K(s)
Figure (b): Generalized Plant Diagram Figure (a): Feedback system

The reason to use generalized plants, rather than individual block diagrams of
control systems, is that a number of control systems with uncertain elements have
been designed using generalized plants and, consequently, established design
approaches using such plants are available.

Hinf Robust Control Problem

H Infinity Control Problem. Any weighting function, such as W(s), is an important
parameter to influence the resulting controller K(s). In fact, the goodness of the
resulting designed system depends on the choice of the weighting function or
functions used in the design process.
z1 z
w z2
z P(s) w Wm(s)I Ws(s)I
Generalized
y plant u u Gp(s) y
y y
K(s) K(s)
Figure (b): Generalized Plant Diagram Figure (a): Feedback system

Note that the controller that is the solution to the H infinity control problem is
commonly called the H infinity controller.

18
1
2
3
4
 Introduction to Adaptive Control

– Adaptive Control: Identifier-Based

– Adaptive Control: Non–Identifier-Based
– Gain Scheduling
– Why Adaptive Control
– A Brief History

Introduction

• Adapt means to "change (oneself) so that one's behavior will

conform to new or changed circumstances."

• The words adaptive systems and adaptive control have been

used as early as 1950.

• We use the following specific definition of adaptive control:

Adaptive control is the combination of a parameter estimator,
which generates parameter estimates online, with a control
law in order to control classes of plants whose parameters are
completely unknown and/or could change with time in an
unpredictable manner.

2
 Introduction

• The choice of the parameter estimator, the choice of the control

law, and the way they are combined leads to different classes of
adaptive control schemes.

• Adaptive control as defined above has also been referred to as

identifier-based adaptive control in order to distinguish it from
other approaches referred to as non-identifier-based, where
similar control problems are solved without the use of an online
parameter estimator.

• The design of autopilots for high-performance aircraft was one

of the primary motivations for active research in adaptive
control in the early 1950s.
3

Introduction

• The controller structure consists of a feedback loop and a

controller with adjustable gains, as shown in following Figure.

General adaptive control structure for aircraft control.

4
 Adaptive Control: Identifier-Based

The class of adaptive control schemes studied in this course is

characterized by the combination of an online parameter estimator,
with a control law. The way the parameter estimator, also referred
to as adaptive law, is combined with the control law gives rise to
two different approaches:

1- In the first approach, referred to as indirect adaptive control, the

plant parameters are estimated online and used to calculate
the controller parameters. In other words, at each time t, the
estimated plant is formed and treated as if it is the true plant in
calculating the controller parameters. This approach has also been
referred to as explicit adaptive control, because the controller
design is based on an explicit plant model.

Adaptive Control: Identifier-Based

2- In the second approach, referred to as direct adaptive control,

the
plant model is parameterized in terms of the desired controller
parameters, which are then estimated directly without intermediate
calculations involving plant parameter estimates. This approach has
also been referred to as implicit adaptive control because the
design is based on the estimation of an implicit plant model.
The basic structure of indirect adaptive control is shown in
following Figure. The plant model G(q*) is parameterized with
respect to some unknown parameter vector q*.

6
Adaptive Control: Identifier-Based

Indirect adaptive control structure.

Adaptive Control: Identifier-Based

Direct adaptive control structure.

8
 Adaptive Control: Identifier-Based
In general, direct adaptive control is applicable to SISO linear plants
which are minimum phase, since for this class of plants the
parameterization of the plant with respect to the controller
parameters for some controller structures is possible.
Indirect adaptive control can be applied to a wider class of plants
with different controller structures, but it suffers
from a problem known as the stabilizability problem explained as
follows:
The controller parameters are calculated at each time t based on
the estimated plant. Such calculations are possible, provided that
the estimated plant is controllable and observable or at least
stabilizable and detectable.

Adaptive Control: Identifier-Based

Since these properties cannot be guaranteed by the online

estimator in general, the calculation of the controller parameters
may not be possible at some points in time, or it may lead to
unacceptable large controller gains.
So, solutions to this stabilizability problem are possible at the
expense of additional complexity. Efforts to relax the minimum-
phase assumption in direct adaptive control and resolve the
stabilizability problem in indirect adaptive control led to adaptive
control schemes where both the controller and plant parameters
are estimated online, leading to combined direct/indirect schemes
that are usually more complex .

10
 Adaptive Control: Non-Identifier-Based

Another class of schemes that do not involve online parameter

estimators is referred to as non-identifier-based adaptive control
schemes. In this class of schemes, the online parameter estimator
is replaced with search methods for finding the controller
parameters in the space of possible parameters, or it involves
switching between different fixed controllers, assuming that at
least one is stabilizing or uses multiple fixed models for the plant
covering all possible parametric uncertainties or consists of a
combination of these methods.
We briefly describe the main features, advantages, and limitations
of these non-identifier-based adaptive control schemes. Some of
these approaches are relatively recent and research is still going
on.

11

Adaptive Control: Non-Identifier-Based

• Gain Scheduling
The gain scheduler consists of a lookup table and the appropriate
logic for detecting the operating point and choosing the
corresponding value of control gains from the lookup table. With
this approach, plant parameter variations can be compensated by
changing the controller gains as functions of the input, output, and
auxiliary measurements. The advantage of gain scheduling is that
the controller gains can be changed as quickly as the auxiliary
measurements respond to parameter changes. Frequent and rapid
changes of the controller gains, however, may lead to instability;
therefore, there is a limit to how often and how fast the controller
gains can be changed.

12
 Adaptive Control: Non-Identifier-Based
• Gain Scheduling

Gain scheduling structure.

13

Adaptive Control: Non-Identifier-Based

• Gain Scheduling
One of the disadvantages of gain scheduling is that the adjustment
mechanism of the controller gains is precomputed offline and
provides no feedback to compensate for incorrect schedules. A
careful design of the controllers at each operating point to meet
certain robustness and performance measures can accommodate
some uncertainties in the values of the plant parameters. However
large unpredictable changes in the plant parameters, may lead to
deterioration of performance or even to complete failure.
Despite its limitations, gain scheduling is a popular method for
handling parameter variations in flight control and other systems.
While gain scheduling falls into the generic definition of adaptive
control, we do not classify it as adaptive control due to the lack of
online parameter estimation which could track unpredictable
changes in the plant parameters. 14
 Adaptive Control: Non-Identifier-Based

• Multiple Models
• Search Methods, and
• Switching Schemes
A class of non-identifier-based adaptive control schemes emerged
over the years which do not explicitly rely on online parameter
estimation. These schemes are based on search methods in the
controller parameter space until the stabilizing controller is found or
the search method is restricted to a finite set of controllers, one of
which is assumed to be stabilizing. In some approaches, after a
satisfactory controller is found it can be tuned locally using online
parameter estimation for better performance.

15

Adaptive Control: Non-Identifier-Based

• Multiple Models
• Search Methods, and
• Switching Schemes
Since the plant parameters are unknown, the parameter space is
parameterized with respect to a set of plant models which is used to
design a finite set of controllers so that each plant model from the
set can be stabilized by at least one controller from the controller
set. A switching approach is then developed so that the stabilizing
controller is selected online based on the I/O data measurements.
Without going into specific details, the general structure of this
multiple model adaptive control with switching, as it is often called,
is shown in next Figure.

16
 Adaptive Control: Non-Identifier-Based

Multiple models adaptive control with switching

17

Why Adaptive Control

The choice of adaptive control as a solution to a particular control

problem involves understanding of the plant properties as well as of
the performance requirements. The following simple example
illustrates situation where adaptive control is superior to linear
control.
Consider the scalar plant

where u is the control input and x the scalar state of the plant. The
parameter a is unknown. We want to choose the input u so that the
state x is bounded and driven to zero with time. If a is a known
parameter, then the following linear control law can meet the
control objective.

18
 Why Adaptive Control

x ® 0 as t ® ¥

In the absence of an upper bound for the plant parameter no linear

controller could stabilize the plant and drive the state to zero.
As we will establish later , the adaptive control law

guarantees that all signals are bounded and Ÿ converges to zero no

matter what the value of the parameter ˆ is. This simple example
demonstrates that adaptive control is a potential approach to use in
situations where linear controllers cannot handle the parametric
uncertainty.
19

A Brief History
• Early 1950s, the design of autopilots for high-performance aircraft
motivated intense research activity in adaptive control.
• 1958, 1961, Model reference adaptive control was suggested by
Whitaker and coworkers in to solve the autopilot control problem.
• In the late 1980s to early 1990s, the use of neural networks as
universal approximators of unknown nonlinear functions led to the
use of online parameter estimators to "train" or update the
weights of the neural networks.
• Adaptive control has a rich literature full of different techniques for
design, analysis, performance, and applications. Several survey
papers and books and thesis have already been published.
• Despite the vast literature on the subject, there is still a general
feeling that adaptive control is a collection of unrelated technical
tools and tricks. 20
Parametric Models

1
 Parametric Models
The first step in the design of online parameter identification (PI)
algorithms is to lump the unknown parameters in a vector and
separate them from known signals, transfer functions, and other
known parameters in an equation that is convenient for parameter
estimation.
In the general case, this class of parameterizations is of the form

where is the vector with all the unknown parameters and

are signals available for measurement.
We refer it as the linear "static "parametric model (SPM).
2
Parametric Models
The SPM may represent a dynamic, static, linear, or nonlinear
system.

Example:

where x, u are the scalar state and input, respectively, and a, b are
the unknown constants we want to identify online using the
measurements of x, u .

3
Parametric Models

4
 Parametric Models
Another parameterization of the above scalar plant is

In the general case, the above parametric model is of the form

5
 Parametric Models

Where are signals available for measurement and

is a known stable proper transfer function, where q is either
the shift operator in discrete time (i.e., q = z) or the differential
operator (q = s) in continuous time. We refer to this model as the
linear "dynamic"parametric model (DPM).
The importance of the SPM and DPM is that the unknown parameter
vector appears linearly.
So we refer to SPM and DPM as linear in the parameters
parameterizations. 6
 Parametric Models
We can derive SPM from DPM if we use the fact that is a
constant vector and redefine to obtain

In a similar manner, we can filter each side of SPM and DPM using a
stable proper filter and still maintain the linear in the parameters
property and the form of SPM, DPM. This shows that there exist an
infinite number of different parametric models in the form of SPM,
DPM for the same parameter vector .
7
 Parametric Models
In some cases, the unknown parameters cannot be expressed in
the form of the linear in the parameters models. In such cases the
PI algorithms based on such models cannot be shown to converge
globally. A special case of nonlinear in the parameters models for
which convergence results exist is when the unknown parameters
appear in the special bilinear form
bilinear static parametric model (B-SPM)
or
bilinear dynamic parametric model (B-DPM)

8
 Parametric Models
bilinear static parametric model (B-SPM)
bilinear dynamic parametric model (B-DPM)

where are signals available for measurement

at each time t, and are the unknown parameters.
The transfer function is a known stable transfer function.
9
 Parametric Models
state-space parametric models (SSPM)

In some applications of parameter identification or adaptive control

of plants of the form

whose state x is available for measurement, the following

parametric model may be used:

where is a stable design matrix; are the unknown

matrices; and are signal vectors available for measurement.
The model may be also expressed in the form

10
 Parametric Models
state-space parametric models (SSPM)

It is clear that the SSPM can be expressed in the form of the

DPM and SPM. 11
 Parametric Models
bilinear state-space parametric models (B-SSPM).

Another class of state-space models that appear in adaptive

control is of the form

where B is also unknown but is positive definite, is negative

definite, or the sign of each of its elements is known.

The B-SSPM model can be easily expressed as a set of scalar

B-SPM or B-DPM.

12
 Parametric Models
PI Problem
For the SPM and DPM:
Given the measurements , generate , the estimate
of the unknown vector , at each time t. The PI algorithm
updates with time so that approaches or converges to .
Since we are dealing with online PI, we would also expect that if
changes, then the PI algorithm will react to such changes and
update the estimate to match the new value of .

PI
13
Parametric Models
PI Problem
For the B-SPM and B-DPM:
Given the measurements generate estimates
respectively, at each time t the same way as
in the case of SPM and DPM.

z
 (t )
zl PI  (t )


 (t )   
*

  (t )    * 
   
14
Parametric Models
PI Problem

For the SSPM::

Given the measurements generate estimates
of , (and hence the estimates
, respectively)at each time t the same way as in the case of SPM
and DPM.

x  (Aˆ  A m )T 
  PI  
u   Bˆ T
 15
 Parametric Models
PI Problem

The online PI algorithms generate estimates at each time t, by

using the past and current measurements of signals. Convergence
is achieved asymptotically as time evolves.
For this reason they are referred to as recursive PI algorithms to
be distinguished from the non-recursive ones, in which all the
measurements are collected a priori over large intervals of time and
are processed offline to generate the estimates of the unknown
parameters.
16
Parametric Models
Example 1: Mass-Spring-Dashpot System

Let us assume that M, f, k are the constant

unknown parameters that we want to
estimate online.

express in the form of SPM

17
Parametric Models
Example 1: Mass-Spring-Dashpot System
Measurements:

To avoid of derivatives , we filter both

sides with the stable filter

18
Parametric Models
Example 1: Mass-Spring-Dashpot System
Another possible parametric model is:

19
Parametric Models
Example 2: Cart with two inverted pendulums

y : 1

20
Parametric Models
Example 2: Cart with two inverted pendulums

21
Parametric Models
Example 2: Cart with two inverted pendulums

To avoid of derivatives , we filter both

sides with the stable filter ,

22
Parametric Models
Example 2: Cart with two inverted pendulums

If in the above model we know that is nonzero, redefining the

constant parameters as we obtain the
following B-SPM:

where,

23
 Parametric Models
Example 3: nonlinear system

Filtering both sides of the equation with the filter , we can

express the system in the form of the SPM

where,

24
 Parametric Models
Example 4: DPM model

If we want W(s) to be a design transfer function with a pole, say

at we write:

25
 Parametric Models
Example 4: DPM model

26
 Parametric Models
Example 5: SSPM model
x 1  u1  a11 a12  b11 b12 
where, x    , u    , A    , B  
 2
x  2
u  21 22 
a a  21 22 
b b

where,

27
 Parametric Models
Example 6: n-th order-SISO LTI system

where,

28
 Parametric Models
Example 6: n-th order-SISO LTI system
Filtering by

where,

29