INTRODUCTION TO CONTROL SYSTEM

DESIGN

The purpose of the control system design is to

make the system performance to meet the
specifications.

Performance specifications

The time response of a control system consists of two

parts:

Transient response: the transient response goes from

the initial state to the final state.

Steady state requirements: the steady state response

means the system output behaves as the time
approaches infinity.

The performance specifications should not be more

stringent than necessary to perform the given task.

Typical test signals

The commonly used test input signals are those of

step functions, ramp function, acceleration functions,
impulse functions, sinusoidal functions and the like.

3
 Transient response

First order systems

Unit step response of first order systems

When,

6
 Unit ramp response of the first order systems

When,

Second order systems

The closed loop transfer function is:

This is called the standard form of the second order system.

The system has two closed loop poles,

The dynamic behaviour of the second order system

is determined by the position of two poles or
described in terms of two parameters .

9
Relation between the characteristic equation roots of the
second order system and the damping ratio and the nature
frequency.

Overdamped

Critically damped

10

Underdamped

Undamped

11

Divergent

12
 13

Definition of transient response specification

Delay time
Rise time
Peak time
Maximum
Overshoot
Settling time

14

Second-order system and transient-response

specifications

Zn
2
G(s)
C (s) R( s) 2 R( s)
1  G(s) s  2]Z n s  Zn
2

e ]Zn t 1 ] 2
c(t ) 1  sin(Z d t  tan 1 )
1 ] 2 ]

15
Rise time:

Peak time:

16

Maximum (percent) overshoot:

c(t p )  c(f)
Mp e  (V / Z ) S
d
e (] / 1] 2 ) S

c (f )

Settling time: 4
ts (2%criterion)
]Z n
3
ts (5%criterion)
]Z n

17

Example:

The response specification of a second order

control system for a step input are:

a) Percentage overshoot < 5%

b) Settling time (2% criterion) < 4 sec
c) Peak time < 1 sec

Sketch the permissible area for the closed loop

poles on the s-plane in order to achieve the
desired response.

18
 Design consideration related to transient
response (High Order System)

1. The response of a system is dominated by

those poles closest to the origin or jZ axis in the
s plane. Those closed loop poles, near the jZ
axis, are called dominant closed loop poles.

19

2. If a pole is located very far from the origin,

the transient corresponding to such a remote
pole are small and last a short time, therefore
it may be neglected to reduce the system
order.

20

a q aj r
b ( s  ] k Zk )  ck Zk 1  ] k2
C (s)   k
s j 1 s  pj k 1 s 2  2] k Zk s  Zk2
q r
a  a je  bk e] k Zk t cos Zk 1  ] k2 t
 p jt
c(t )
j 1 k 1
r
 ck e  ] k Zk t
sin Zk 1  ] k2 t
k 1

21
 3. If a zero is located very near the
dominant complex poles, the zero will
affect the transient response of the system.

4.A pair of closely located poles and

zeros will effectively cancel each other.

22

Steady state errors in unity feedback control systems

Any physical control system inherently suffers

steady state error in response to certain types of
inputs. A system may have no steady state error
to a step input, but the same system may exhibit
nonzero steady state error to a ramp input.

Whether a given system will exhibit steady state

error for a given type of input depends on the
type of open-loop transfer function of the system.

23

Typical steady-state error due to a step input.

24
Typical steady-state error due to a ramp-function input.

25

Typical steady-state error due to a parabolic-function input

26

27
 The steady state error in unity feedback system
is defined as E(s).

sR ( s )
ess (t ) lim e(t ) lim sE ( s ) lim
t of s o0 s o0
1  G ( s)

28

Steady state error constants

Steady state error constants indicate the ability

of the system following the typical reference
signals.

29

Static position error constant.

When a system has a unit step input,

sR ( s ) s 1 1
ess (t ) lim lim
s o0
1  G ( s ) so0 1  G ( s ) s 1  G (0)
Kp lim G ( s ) G (0)
s o0

1
ess (t )
1 Kp

30
Static velocity error constant.

When a system has a unit ramp input,

31

Static acceleration error constant.

When a system has a unit acceleration input,

sR ( s) s 1 1
ess (t ) lim lim lim
s o0 1  G ( s ) s o0 1  G ( s ) s3 s o0 s 2G ( s )
Ka lim s 2G ( s)
s o0
1
ess (t )
Ka

32

Classification of control systems

Control systems may be classified according to

their ability to follow step inputs, ramp inputs,
parabolic inputs and so on.

33
 K (Ta s  1)(Tb s  1)(Tm s  1)
G(s)
s N (T1s  1)(T2 s  1)(Tp s  1)

The classification scheme is based on the

number of integrations indicated by the open-
loop transfer function, which is N .

34

35

Design considerations related to steady

state errors

1. The error constants describe the ability

of a unity-feedback system to follow the
given type of input. They are indicative
of the steady state performance.

36
 2. It is generally desirable to increase
the error constants, while
maintaining transient response
within an acceptable range.

37

3. To improve the steady state

performance, we can increase the
type of the system by adding an
integrator or integrators to the feed
forward path. However, this
introduces an additional stability
problem.

38

Review on drawing the root locus

1. The basic characteristic of the transient

response of a closed loop system is
closely related to the location of the
closed loop poles.

39
 2. The value of the loop gain effects the
locations of the closed loop poles (roots
of the characteristic equation).
3. The Characteristic equation: 1  KG ( s ) H ( s ) 0

C (s) KG ( s )
R( s) 1  KG ( s ) H ( s )

40

4. The root locus is a locus of the

closed loop roots for all values
of a loop gain.

41

All the points on the locus meet:

1). Angle condition:

KG ( s) H ( s) r180o (2k  1) (k 0,1,2,)

2). Magnitude condition:

KG ( s ) H ( s ) 1

42
Procedures

43

44

45
Example:

Consider a unit feedback system with the feed-

forward transfer function of:

K ( s  1.5)
G ( s)
s( s  1)( s  10)

Design the value of gain K to produce 1.5%

overshoot. Also determine the settling time.

Solution: K=40, Ts=0.87

46

Design methods

Time domain (Root locus method)

Frequency domain (Frequency response

method)

Compensators

Lead compensators
Lag compensators
PID compensators

47

Compensation configuration

Series compensation
Parallel compensation
State feedback

48
 49

DESIGN IN TIME DOMAIN (ROOT LOCUS

METHOD)

The root locus method is a graphical method for

determining the locations of all closed loop poles
from knowledge of the locations of the open loop
poles and zeros as some parameter (usually the
gain) is varied from zero to infinity.

In some cases, the desired system performance

cant be achieved just by the adjustment of gain.
Therefore it is necessary to insert compensators to
reshape the root locus to meet the performance
specifications.

Series compensator

Effects of the addition of zeros

The addition of a zero to the open loop
transfer function has the effect of pulling the
root locus to the left, tending to make the
system more stable and to speed up the
settling of the response.

2
Effects of the addition of poles
The addition of a pole to the open loop
transfer function has the effect of pulling the
root locus to the right, tending lower the
systems relative stability and to slow down
the settling of the response.

PID Controllers

KI
Gc ( s ) Kp   KDs
s
PD Control

PD Control is equivalent to adding a zero

the open loop transfer function.

PI Control

Gc ( s ) K p  KI / s

PI Control increases the system type by 1.

5
Example: Consider the following system:

Motor
Vi
1
Z
G(s)
+
- s 2  5s  6

Figure 1

System with the proportional controller G(s) K

System with a PD controller. G ( s) K ( s  4)

8
 Step response for P and PD control:

System with the integral controller . G ( s ) K/s

10

System with the proportional and integral (PI) controller

.
G ( s) K (1  1 / s )

11
 Step response for P, I and PI control:

Blue: original
Green: I control
Red: PI control

12

clear; %clear all variables

clf; %clear all figures
numg=[1];
deng=[1 5 6];
sysg=tf(numg,deng);
t=[0:0.1:15];
%*********
sys1=sysg;
figure(1);
rlocus(sys1);axis([-4 0 -6 6]);
grid;
hold on
plot([0 -6],[0 -6*tan(acos(0.5))],'--',[0 -6],[0 6*tan(acos(0.5))],'--');
hold off
[kp,ploes]=rlocfind(sys1)

13

numc=[1];
denc=[1 0];
sysc=tf(numc,denc);
sys2=series(sysc,sysg);
figure(2);
rlocus(sys2);axis([-4 0 -6 6]);
grid;
hold on
plot([0 -6],[0 -6*tan(acos(0.5))],'--',[0 -6],[0 6*tan(acos(0.5))],'--');
hold off
[ki,ploes]=rlocfind(sys2)

14
 numc=[1 1];
denc=[1 0];
sysc=tf(numc,denc);
sys3=series(sysc,sysg);
figure(3);
rlocus(sys3);axis([-4 0 -6 6]);
grid;
hold on
plot([0 -6],[0 -6*tan(acos(0.5))],'--',[0 -6],[0 6*tan(acos(0.5))],'--');
hold off
[kpi,ploes]=rlocfind(sys3)

15

figure(4)
sys1_o=kp*sys1;sys1_cl=feedback(sys1_o,[1]);
sys2_o=ki*sys2;sys2_cl=feedback(sys2_o,[1]);
sys3_o=kpi*sys3;sys3_cl=feedback(sys3_o,[1]);
[y1,t]=step(sys1_cl,t);
[y2,t]=step(sys2_cl,t);
[y3,t]=step(sys3_cl,t);
plot(t,y1,'-',t,y2,'--',t,y3,':');grid;
xlabel('time [sec]'),ylabel('y(t)')
title('Gc(s):proportional (solid), integral (dashed) & PI (dotted)')

16

Lead compensation

In general, we can regard the design of

compensator of control systems as a filter
design problem, then there are a large number
of possible schemes. The lead compensator has
a phase lead frequency response.

17
 18

Its transfer function is:

1 1
s s
Eo ( s ) R4 C1 R1C1 Ts  1 T
K cD Kc
Ei ( s ) R3C2 s  1 DTs  1 s
1
R2C2 DT
s  z1
Gc ( s ) Kc
s  p1
where
R4C1
T R1C1 , DT R2C2 , K c
R3C2

19

For the lead compensator,

For the lag compensator,

20
Pole-zero configurations

21

A lead compensator is used where the

original system either is unstable for all
values of gain or is stable but has undesirable
transient response characteristics.

Therefore the reshaping of the root locus is

necessary so that the new root locus passes
through the dominant closed loop poles at the
desired locations.

22

The procedures of designing a lead compensator

are:

1. From the performance specifications,

determine the desired location for the
dominant closed loop poles.

23
 2. By drawing the root locus plot of the
uncompensated system ascertain
whether or not the gain adjustment alone
to yield the desired closed loop poles. If
not, calculate the angle deficiency I ,
which must be contributed by the lead
compensator, if the new root locus is to
pass through the desired locations for the
dominant closed loop poles.

24

3. Assume the lead compensator to

be
1
s
Gc ( s ) Kc T 0 D 1
1
s
DT

Determine the location of the pole and

zero of the lead compensator so that the
lead compensator will contribute the
necessary angle.

25

4. Determine the open loop gain of the compensated

system from the magnitude condition.
Example: Design a lead compensator so that
Zn 4rad / sec, 9 0.5

4
s2 +2s
Step Transfer Fcn Scope

s2
Answer:
4
s4
26
 Lag compensation

The lag compensator has a phase lag frequency

response.

Its transfer function is:

1 1
s s
Eo ( s ) R4 C1 R1C1 Ts  1 T
Kc E Kc
Ei ( s ) R3C2 s  1 ETs  1 s
1
R2C2 ET
s  z1
Gc ( s ) K c
s  p1
R4C1
where T R1C1 , ET R2C2 , K c
R3C2

For the lag compensator, E ! 1 or R1C1  R2C2

Lead Controller Lag Controller

3
 A lag compensator is used where the original
system has satisfactory transient response
characteristics, but unsatisfactory steady
state characteristics.

The compensation is required to increase the

open loop gain without appreciably changing
the transient response.

Therefore we need to make the magnitude

contribution of the lag compensator is unity
and the angle contribution of the lag
compensator is very small. We need to

1. Place the zero and pole very

close to each other and is
unity.

2. Place the pole and zero very

close to the origin so that the
value of E can be made large.

The compensated the system will have the

static error constant:

Therefore the static error constant is

increased by a factor of E .

6
The procedures of designing a lag
compensator are:

1. Draw the root locus plot for the

uncompensated system and locate the
dominant closed loop poles on the root
locus.
2. Evaluate the particular static error
constant specified in the problem and
determine the amount of increase in the
static error constant necessary to satisfy
the specifications.

3. Assume the lag compensator to be

1
s
Gc ( s ) K c T 1 E
1
s
ET

Determine the location of the pole and zero

of the lag compensator that produce the
necessary increase in the particular static
error constant without appreciably altering
the original root locus.

4. Draw a new root locus plot for the

compensated system. Locate the
desired dominant closed loop poles on
the root locus.

5. Adjust gain K c from the magnitude

condition so that the dominant closed loop
pole lie at the desired location.

9
Example Consider the system shown below.

Design a lag compensator so that the static velocity

error constant is about 5 without appreciably
changing the location of the dominant closed loop
poles (=0.5).

10

11

12
13

14

15
 16

17

Parallel compensation

A velocity (tachometer) feedback system is an example of a

parallel-compensated system.

Example: The tachometer is added to the following system

to reduce the overshoot less than 10%. Calculate an
appropriate tachometer feedback coefficient to achieve this.

10 1
s+2 s
Step Transfer Fcn Transfer Fcn1 Scope

Gain

(K=0.18)
18
Example: Design the value of K such that the
damping ratio of the dominant closed loop pole is
0.4.

20 1
s2 +5s+4 s+1
Step Transfer Fcn Transfer Fcn1 Scope

Gain

(k=0.449)

19

CONTROL SYSTEM ANALYSIS IN

FREQUENCY DOMAIN

One advantage of the frequency response

approach is that we can use the data
obtained from measurements on the
physical system without deriving its
mathematic model.

Bode Diagram

A Bode diagram consists of two graphs:

Plot of the logarithm of the magnitude
Plot of the phase angle.

Both are plotted against the frequency

on a logarithmic scale.

The main advantage of using the Bode

diagram is that multiplication of
magnitudes can be converted into
addition.
Frequency
response of
a second
order
system
 General procedure for plotting Bode
diagram

Write the sinusoidal transfer function as

a product of basic factors.

Identify the corner frequencies

associated with these basic factors.

Draw the asymptotic curves with proper

slops between the corner frequencies.

The exact curve can be obtained by

adding proper corrections, especially
for a second order system.

Relative stability

In designing a control system, we require that

the system be stable. Furthermore, it is
necessary that the system have adequate
relative stability.

It is common practice to represent the relative

stability in terms of phase margin and gain
margin.
Phase margin
The gain crossover frequency is the frequency
at which |G(jw)| is unity.
The phase margin is 180 degree plus the
phase angle of the open loop transfer function
at the gain crossover frequency,

J 180 0  I

For a stable system, the phase margin is

positive.

Gain Margin
The phase crossover frequency is defined to be the
frequency at which the phase angle of the open loop
transfer function equals 180 degree.
The gain margin is the reciprocal of the magnitude
|G(jw)| at the phase crossover frequency.

1
Kg
G ( jZ1 )
Z1 is the phase crossover frequency.
 In terms of decibels,

K g dB 20 log K g 20 log G ( jZ1 )

For a stable system, the gain margin is positive.

For a stable system, the gain margin indicates
how much the gain can be increased before the
system becomes unstable.
For an unstable system, the gain margin
indicates how much the gain must be decreased
to make the system stable.

Information obtainable from open loop

frequency response

The low frequency response plots indicate the

system types.

The high frequency response plots indicate the

degree difference of the numerator and
denominator (n  m) .

The middle frequency response plots indicate

relative stability.

frequency response in the standard second

order system

Resonant Peak Magnitude and Resonant

frequency

1
Zr Z n 1  2] 2 Mr
2] 1  ] 2
The magnitude of the resonant peak gives an
indication of the relative stability of the
system.
A large resonant peak magnitude indicates a
small damping ratio.
 Phase margin and the damping ratio are directly related.
J
]
100

Comparing Z r Z n 1  2] 2 and Z d Zn 1  ] 2 , the

Z r and Zd are very close for a small ] .

Therefore the value of Z r is indicative of the speed

of the transient response of the system.

Cut off frequency and bandwidth.

The frequency at which the

magnitude of the closed
loop frequency response is
3dB below its zero
frequency value is called
the cut off frequency Z .
b

The frequency range 0  Z  Zb is called the

bandwidth of the system.
The bandwidth indicates how well the system will
track an input sinusoid.

For a second order system, the bandwidth

decreased with the increase in ] .
The figure indicates that the rise time increases with the
increase in ] for a given Zn ,
for a second order system.

Therefore, the rise time and the bandwidth are

inversely proportional to each other.

Or the bandwidth is proportional to the speed

response.

Correlation between step transient response

and frequency response in general system

If the frequency response of a high order

system is dominated by a pair of complex
conjugate closed loop poles, the transient
response frequency response correlation
existing for the standard second order system
can be extended to the higher order system.
 Closed loop frequency response of unity
feedback systems

Consider a unity feedback system with the

closed loop transfer function:

C ( jZ ) G ( jZ )
Me jD
R ( jZ ) 1  G ( jZ )

The constant magnitude loci and constant

phase angle loci are used to determining the
closed loop frequency response from the
open loop frequency response.

Nichols (log-magnitude versus phase) plane is

called Nichols chart.

Example: Find M r ,Z r , Zb , J and K g

K
G(s) K 1 :
s ( s  1)(0.5s  1)
CONTROL SYSTEM DESIGN BY FREQUENCY
RESPONSE

The advantages of using Bode diagram in design:

A bode diagram of the compensator can be simply

added to the original Bode diagram.

If the open loop gain is varied, the magnitude

curve is shifted up or down without changing the
slope of the curve and the phase curve remains
the same.

Common approach of the design by frequency

response:

1. We first adjust the open loop gain so that

the requirement on the steady state
accuracy is met.

2. Plot the frequency response of the

uncompensated system.

3. If the specification on the phase margin and gain

margin are not satisfied, then a suitable
compensator that will reshape the open loop
transfer function is determined.

In many cases, compensation is essentially a

compromise between steady state accuracy and
relative stability.
Lead compensation

Characteristics of lead compensator

1
s
Ts  1 T s  z1
Gc ( s ) K cD Kc Kc
DTs  1 s
1 s  p1
DT
j ZT  1
K cD 0 D 1
jZDT  1

1D
2 1D
sin I m
1 D 1 D
2

Figure 9-3 Polar plot of a lead compensator

Bode plot of a
lead
compensator

Kc 1, D 0.1

1 1 1 1
log Z m (log  log ) , Z m
2 T DT DT
j ZT  1 1
jZDT  1 Z Z 1 D
DT
m
 The lead compensator is basically a high
pass filter.

It increases the system bandwidth and

improves in transient response.

But it may accentuate high frequency

noise effects.

Lead compensator design

The primary function of the lead compensator

is to reshape the frequency response curve to
provide sufficient phase lead angle.

Consider the system below, procedure for designing

a lead compensator:
1. Assume the following lead compensator:

1
s
T Ts  1
Gc ( s ) Kc K cD
s
1 DTs  1
DT
K K cD

Determine gain K to be added by the lead

compensator to satisfy the requirement on
the given static error constant.
Example: Consider the system shown below:
Bode plot of the gain adjusted, but uncompensated system

Bode plot of compensated system

Unit step response

 Unit ramp response

Lag compensation

Characteristics of lag compensators

1
s
Ts  1 T s  z1
Gc ( s ) KcE Kc Kc
ETs  1 s
1 s  p1
ET
jZT  1
Kc E E !1
jZET  1

Bode plot of a lag compensator

Lag compensator design

The primary function of a lag compensator is to

provide attenuation in the high frequency range
to give the system sufficient phase margin.

The procedure for designing lag

compensators may be stated as follows:

1. Assume the following lag compensator:

1
s
T Ts  1
Gc ( s ) Kc Kc E
s
1 ETs  1
DT
K Kc E

The open loop transfer function of the

compensated system is:

Ts  1 Ts  1
Gc ( s )G ( s ) K G( s) G1 ( s )
ETs  1 ETs  1
G1 ( s ) KG ( s )

Determine gain K
to satisfy the requirement on the given
static error constant.
Bode plots of compensated and uncompensated system
 Unit step response

Unit ramp response

CONTROL SYSTEM MODELLING, ANALYSIS

AND DESIGN IN STATE SPACE

The classical technique is based on converting a

systems differential equation to a transfer
function, which relates the output to the input.

The disadvantage of the classical approach is its

limited application: linear, time-invariant systems

The advantage of the classical approach is that

they rapidly provide stability and transient
response information. Thus we can immediately
see the effect of varying system parameters until
an acceptable design is met.
Why the modern control is developed:

Complexity of systems (multi inputs and multi

outputs, nonlinearity, time varying systems)

Stringent requirement on the performance of

control system

Availability of numerous state space software

packages for the personal computer

State variables: The state variables of a

dynamic system are the variables making
up the smallest set of variables that
determine the state of the dynamic system.

State space: The n-dimensional space

whose coordinate axes are n state
variables is called a state space.
State space equation:

In state space analysis we are concerned with

three types of variables that are involved in the
modelling of dynamic systems: input variables,
output variables and state variables.

It is a set of n simultaneous, first order

differential equations with n state variables.

Output equation: the algebraic equation

that expresses the output variables of a
system as linear combinations of the state
variables and the input.
Modelling electronic circuits in state space

Examples:

x1 vo
x
x2 vo
x 0 1 x 0
xx1 1 R 1  1 vi
x  LC 
L 2 LC
x
2

x1 vo
x2 i
1
x 0 0
xx1 C x1  1 v
1 R x2 i
x   L
2
L L

Note:
The state variables need not be physically
measurable or observable quantities.

The state space representation for a given

system is not unique.

Modelling DC motor in state space

Select ia and Z as the states:

x1 ia
x2 Z
Ra Kb 1
x  L 
La x1  La
0
vt
xx1 a
x B x 1 TL
2 Km  2 0 
J J J

Deriving the state space model from the transfer

functions
b0
T (s)
s n  an 1s n 1    a0
x1 y
dy
x2
dt
d2y
x3
dt 2

d n 1 y
xn
dt n 1
0
x 1 0  0 x1 0
xx1 0 0 1  0 x2 0
x
2         u
 0 0 0  1 xn 1 0
x
xn  a0  a1  a2   an 1 xn b0

b2 s 2  b1s  b0
T ( s)
s  a2 s 2  a1s  a0
3

x
x1 x2
x
x2 x3
x
xx1 0 1 0 x1 0
x 0 0 1 x2  0 r
x2
x3  a0  a1  a2 x3 1

x1
y >b0 b1 b2 @ x2
x3
Deriving the state space model from the
system block diagrams

5( s  1) 1 6
R( s) o o o o Y (s)
s5 s2 s3
x
xx1  3 6 0 x1 0
x 0  2  4 x  5 r
x2 2
x3 0 0  5 x3 5

x1
y >1 0 0@ x2
x3

CONTROL SYSTEM ANALYSIS IN STATE SPACE

Correlation between transfer function and

state space equations

T ( s ) C( sI  A) 1 B  D
Q( s)
sI  A
adj ( sI  A )
( sI  A )  1
det( sI  A )

sI  A is equal to the characteristic polynomial of

G(s).

In other words, the eigenvalues of A are

identical to the poles of T(s).

The stability of the system can be

checked by looking at eigenvalues of A.
 Example 6.11
a >0 3 1 ; 2 8 1 ;  10  5  2@
eig (a);
 0.8821 r 2.4330 j
7.7642

The command ss2tf and tf2ss can be used to

transform the model from state space to transfer
function and vice versa.

x 0 1 x1 1 1 u1
xx1 
x
 24
 4 x 2 0 1 u 2
2
y1 1 0 x1 0 0 u1
y 
2
0
1 x 2 0
0 u 2
A >0 1 ;  24  4 @
B >1 1 ; 0 1@
C >1 0 ; 0 1@
D >0 0 ; 0 0@
[ num , den ] ss 2 tf ( A , B , C , D ,1 )
num
0 1 4 U1
0 0  24
dem
1 4 5
Y1 ( s ) s  4
U 1(s) s 2
 4 s  25
Y2 (s)  24
U 1(s) s 2
 4 s  25

Solving the time invariant state equation

t
x (t ) e At x(0)  e A ( t W ) Bu (W )dW
0
Laplace transform approach to solve the time
invariant state equation

X(s) ( sI  A) 1 x(0)  ( sI  A) 1 BU(s)

e At (sI  A) 1

k k
1 2 2 1 f A t
e At I  At  A t    Akt k  
2! k! k 0 k!
 If the system is completely state controllable, then the
following matrix has a rank of n.

>B AB x x x A n1B @

Observability

The system is said to be completely observable

if every state X (t0 ) can be determined from the
observation of y (t ) over a finite time interval,
t0 d t d t1 .
 It can be shown that this requires the following
matrix has a rank of n.

C
CA
.

.

. n 1
CA

Steady state errors

E (s) R ( s )[1  T ( s )] R ( s )[1  C( sI  A ) 1 B  D]

lim sE ( s ) lim sR ( s )[1  C( sI  A ) 1 B  D]
s  !0 s  !0

CONTROL SYSTEM DESIGN IN STATE SPACE

Meeting the transient response (pole

placement design)

In the pole placement method, the closed loop

poles are located at desired locations.
It can be shown that if the system considered
is completely state controllable, then poles of
the closed loop system may be placed at any
desired location by means of state feedback
through an appropriate state feed back gain
matrix.

x
x Ax  Bu
y Cx
Solving pole placement problem with Matlab
Meeting steady state response

To have desired steady state error as well as

the desired transient response, the integral
control can be inserted in the state feedback
system.