INDIAN INSTITUTE OF TECHNOLOGY ROORKEE

EE-581 :Intelligent Control Techniques

Instructor

Barjeev Tyagi
Intelligent Control Techniques

• In classical & modern control, systems are rigidly modeled.

• The software does what is told.

• Intelligence comes from the designer.

• In intelligent control, systems are not rigidly modeled.

• The intelligence comes from the software.

• Software does what it wants to.

2
Intelligent Control Techniques

• Intelligent control achieves automation via the emulation of

biological intelligence.

• It either seeks to replace a human who performs a control task

(e.g., a chemical process operator)

or

• It borrows ideas from how biological systems solve problems and

applies them to the solution of control problems (e.g., the use of
neural networks for control).

3
Intelligent Control Techniques

Intelligent control is a class of control techniques that use various

artificial intelligence computing approaches like

• Fuzzy Logic

• Neural Networks

• Genetic Algorithms

• Machine Learning

• Evolutionary Computation
4
Introduction to Fuzzy Logic

5
Introduction to Fuzzy Logic

• Introduced by Dr. Lotfi Zadeh of UC/Berkley as a means to model

the uncertainty of natural language.

• A superset of conventional (Boolean) logic.

• Handles partial truth – truth values between „completely true‟ and

„completely false‟.

• Comparable to the human decision making process.

6
Introduction to Fuzzy Logic
• Fuzzy logic is a superset of conventional (Boolean) logic that has
been extended to handle the concept of partial truth –the truth
values between "completely true" and "completely false".

• A type of logic that recognizes more than simple true and false
values. With fuzzy logic, propositions can be represented with
degrees of truthfulness and falsehood. For example, the statement,
today is sunny, might be 100% true if there are no clouds, 80% true
if there are a few clouds and 0% true if it rains all day.

• A form of knowledge representation suitable for notions that

cannot be defined precisely, but which depend upon their context.
It enables computerized devices to reason more like humans.

7
The Life Cycle of a Process Control System

8
Knowledge Base System (KBS)

Direct Expert Control System (DECS)

•KBS is used in closed loop, thus replacing the

conventional control element.
•KBS replicates the process operator‟s manual
control strategy by employing a rule base.
•Rule base determines the control output signal,
give information about process output variables,
e.g. error, change of error etc.

9
Knowledge Base System (KBS)
Supervisory Expert Control System
(SECS)

•KBS is used out of the control loop acting as a

supervisor of a conventional control element.
•KBS complimenting rather than replacing
conventional controller.
•Conventional controller implement the current
control law while the KBS determines when and
how this control law should be changed.

10
Classical/Crisp Sets

11
Fuzzy Sets

12
Fuzzy Sets
A fuzzy set in a universe of discourse U is characterized by a
membership function  A ( X ) that takes values in the interval [0, 1].
• Therefore, a fuzzy set is a generalization of a classical set by
allowing the membership functions to take any value between 0 and 1
• A fuzzy set A in U may be represented as a set of ordered pairs of a
generic element X and its membership value, i.e,

A  {X ,  A ( X ) / X  U }
• When U is continuous, A is commonly written as

A   A( X ) / X
U

13
Fuzzy Sets
• When U is discrete A is commonly written as
A   A( X ) / X
U

1. List Method: List the members of the set

Universe Fuzzy Set
1,2,3,4….. {1/0.1, 2/0.2, 3/0.6, 4/1, 5/0.6….}
2. Rule Method: Choose number close to zero
0 if X  1
 X  1 if  X  X  1

A( X )  
1  X if 0  X  1
0 if 1  X

14
Different Shapes of Fuzzy Sets

15
Types of Fuzzy Sets
Non-Convex Fuzzy Sets
In which membership grade alternatively increases and decreases
on the domain

Convex Fuzzy Sets

In which membership grade do not alternatively increases and
decreases on the domain

16
Types of Fuzzy Sets

• -Function
1

0.5

 

• S-Function
• L-Function
• -Function
• -Function
• G-Function

17
Types of Fuzzy Sets

• Fuzzy sets in whole domain

18
Types of Fuzzy Sets

• Empty Fuzzy Set: A   if  A ( x)  0 x  U

• Normal Fuzzy Set: include at least one singleton equals to unity in
U i.e.  A ( x0 )  1 . If there is no singleton equal to unity, then the
fuzzy set is called subnormal.
• Equal Fuzzy Set: Two fuzzy sets are equal (A=B) iff

x  U ;  A ( x)   B ( x)
• The support of a fuzzy set is defined by S ( A)  x  U ;  A ( x)  0

• The nucleus of a fuzzy set A is defined by nucleus( A)  x  U ;  A ( x)  1

If there is only one point with membership grade equal to 1

19
Types of Fuzzy Sets

• Height of Fuzzy Set: is equal to the largest membership degree  A

hgt( A)  sup  A ( x)
xU

Fuzzy set A is called normal if hgt(A)=1, and subnormal if

hgt(A)<1

• Subset: A is a subset of B  A  B  iff

x U ;  A ( x)   B ( x)

20
Types of Fuzzy Sets

• Subset

21
Operations on fuzzy sets

Union
The union of fuzzy subsets, A and B, of the set X, is denoted as the
fuzzy subset C of X.
C = A  B such that for each X
C(x)= Max[A(x), B(x)] = A(x) B(x)
Let X={a, b, c, d}. Assume A and B are two fuzzy subsets of X
where A= {1/a, .7/b,.3/c,0/d} and B={.2/a,.9/b,.4/c.1/d}. Then
C={1/a,.9/b,.4/c,1/d}

22
Operations on fuzzy sets

Intersection
The intersection of the fuzzy subsets A and B is denoted as the
fuzzy subset D of X

D = A  B for each x X

 D(x) = Min [( A(x),  B(x)] = A(x)^ B(x)

23
Operations on fuzzy sets

Complement

Assume A is a fuzzy subset of X, the complement or negation

of A, denoted A is defined as

A  {1 / a, 0.3 / b, 0.2 / c, 0.8 / d , 0 / e}

A  {0 / a, 0.7 / b, 0.8 / c, 0.2 / d , 1 / e}

24
Fuzzy Arithmetic

Extension Principle
It provides a general method for combining non- fuzzy and fuzzy
concepts of all kind.

Let A1, A2,…….. An, be fuzzy sets defined respectively on

U1,U2,…….. Un, and let f be a non fuzzy function
f: U1 U2… Un V.

The aim is to extend f such that it operates on A1, A2,…….. An, and
returns a fuzzy set F on V.

25
Fuzzy Arithmetic

Definition:
The extension of f, operating on A1, A2,…….. An results in the
following membership function
 F (v )  sup min(  A1 (u1 )........... An (u n )),
u1 .......... u n
f ( u1 ...... u n ) v

when f-1(v) exists. Otherwise A(v)=0.

Another way to explain this results is
F min(  A1 (u1 )......... .. An (u n ) / f (u1 ........ u n )
U1 U 2 .........Un

In the binary case and on a discrete or compact domain

 F (v )  max min(  A1 (u1 )........... An (u n )),
u1 ..........u n
f ( u1 ......u n ) v

26
Operations on fuzzy sets

Cartesian Product

The Cartesian product of two sets is the set of all possible ordered
pairs whose first component is a member of the first set and
whose second component is a member of the second set.

A=(1,2,3),B=(3,4,5)

{1,3},{1,4},{1,5},{2,3},{2,4},{2,5},{3,3},{3,4},{3,5}}

27
Fuzzy Relations

Generalizes classical relation into one that allows partial membership

– Describes a relationship that holds between two or more objects
• Example: a fuzzy relation “Friend” describe the degree of friendship between
two person
• (in contrast to either being friend or not being friend in classical relation!)

• A fuzzy relation R˜ is a mapping from the Cartesian space X x Y

to the interval [0,1], where the strength of the mapping is
expressed by the membership function of the relation m (x,y)

• The “strength” of the relation between ordered pairs of the two

universes is measured with a membership function expressing
various “degree” of strength [0,1]

28
Fuzzy Relations

Definition:
A fuzzy relationship over the pair X,Y is defined as a fuzzy subset
of the Cartesian product

Example:

If A={a, b, c} and B={1, 2}, then

R={.1/(a,1),.6/(a,2), 1/(b,1),.3/(b,2), 0/(c,1),.9/(c,2)}

is a fuzzy relationship over the space A X B

More generally, if X1,X2..Xn are collection of sets, an n-ary fuzzy

relationship is a fuzzy subset over their Cartesian product X1 x X2 x .Xn

29
Fuzzy Relations

Crisp relation

Relation as rules

30
Fuzzy Relations

Fuzzy relation

31
Operations on Fuzzy Relations

Let R and S be the binary relation defined on X Y.

The Intersection of R and S is defined by

( x, y )  X  Y ;  R S ( x, y )  min(  R ( x, y ),  S ( x, y ))

The Union of R and S is defined by

( x, y )  X  Y ;  RS ( x, y )  max(  R ( x, y ),  S ( x, y ))

32
Operations on Fuzzy Relations

Let R be defined on X Y.

The Projection of R on X (or Y) defined by

proj R on X (orY )  max(  R ( x, y) / x (or y))

Let F be a fuzzy set defined on Y, the Cylindrical Extension of F

on X Y is the set of all tuples (cross product) (x,y) X Y with
membership degree equals to F(y)

ce( F )    F ( y ) /( x, y )
X Y

33
Operations on Fuzzy Relations

Composition

Let A be the fuzzy set defined on X and R be a fuzzy relation

defined on X Y. Then the composition of A and R resulting in a
fuzzy set B defined on Y is given by

B  A  R  proj(ce( A)  R)onY )

 B ( y)  max min( A ( x),  R ( x, y))

x

This is called max min composition

34
Example:

Consider the relation on the domain of height

{170, 172.5, 175, 177.5, 180, 182.5, 185} described by the
linguistically expressed relation
R: Agnes is somewhat taller than Jane
170 172.5 175 177.5 180 182.5 185

170 .4 .1 0 0 0 0 0
172.5 .7 .4 .1 0 0 0 0
175 1 .7 .4 .1 0 0 0
177.5 .7 1 .7 .4 .1 0 0
180 .4 .7 1 .7 .4 .1 0
182.5 .1 .4 .7 1 .7 .4 .1
185 0 .1 .4 .7 1 .7 .4

35
Example:

Consider the fact that „Agnes is rather taller‟

A={0/170, .1/172.5, .4/175, .7/177.5, .9/180, 1/182.5, 1/185}
170 172.5 175 177.5 180 182.5 185

170 0 0 0 0 0 0 0
172.5 .1 .1 .1 .1 .1 .1 .1
Ce(A) = 175 .4 .4 .4 .4 .4 .4 .4

177.5 .7 .7 .7 .7 .7 .7 .7

180 .9 .9 .9 .9 .9 .9 .9

182.5 1 1 1 1 1 1 1

185 1 1 1 1 1 1 1

36
Example:

37
Example:

38
Operations on Fuzzy Relations

Composition
Consider RX Y and SY Z
Extend both relation to X Y Z, then projection of the intersection
of both extended relation in X Z gives the composition of R and
S.

TX Z  R  S  proj (ce( R)  ce( S )) on X  Z )

TX Z  R  S   max min(  R ( xi , y j ),  s ( y j , zk )) /( xi , zk )
X Z j

39
Fuzzy Implication

Consider a statement
if x is A then y is B

“x is A”, “y is B” – fuzzy predicates A(x), B(y)

if A(x) then B(y) can be represented as a relation

R(x,y): A(x) B(y)

where R(x,y) can be considered a fuzzy set with 2-dimentional

membership function
μR(x,y)=f(μA(x), μ B(y))
where f is fuzzy implication function.
40
Fuzzy Implication

MIN fuzzy implication

Interprets the fuzzy implication as the minimum operation
[Mamdani].

PRODUCT fuzzy implication

Interprets the fuzzy implication as the product operation
[Larsen].

41
Fuzzy Implication (Example)

42
Fuzzy Implication (Example)

43
Fuzzy Implication (Example)

44
Fuzzy Implication (Example)

45
Fuzzy Rules Formation

IF- THEN rule base form

IF antecedent THEN consequent

There are three general forms that exist for any linguistic variables

– Assignment Statement

– Conditional Statement

– Unconditional Statement

46
Fuzzy Rules Formation

Compound rule base form: Any compound rule structure may be

decomposed and reduced to a number of simple canonical rule
forms
– Multiple conjunctive antecedents

– Multiple disjunctive antecedents

– Conditional statements (with ELSE and UNLESS)

– Nested IF-THEN rules

47
Fuzzy Logic Systems Architecture

• Fuzzification Module − It transforms the system inputs, which are

crisp numbers, into fuzzy sets.

• Knowledge Base − It stores IF-THEN rules provided by experts.

• Inference Engine − It simulates the human reasoning process by

making fuzzy inference on the inputs and IF-THEN rules.
• (A nonlinear mapping that derives its output based on fuzzy
reasoning and a set of fuzzy if-then rules.)

• Defuzzification Module − It transforms the fuzzy set obtained by

the inference engine into a crisp value.

48
Fuzzy Logic Systems Architecture

49
Fuzzy Logic Systems Architecture
• The knowledge base contains information about the boundaries,
possible transformations of the domains, and the fuzzy sets with
their corresponding linguistic terms. This information represents
the data base. In addition, the knowledge base contains a rule base
consisting of linguistic control rules.

• The decision logic represents the processing unit. It determines the

corresponding output value from the measured input according to
the knowledge base.

• The defuzzification interface has the task of determining a crisp

output value – taking the information about the control variable
provided by the decision logic into account. Finally, if necessary, it
carries out a transformation of the output value into the appropriate
domain.

50
Mamdani Implication

Consider a rule
if e is PM then u is NS Domain of e and u is [-6,6]

1 2 2 1 
PM   2, 3, 1 4, 5, 6 NB NM NS ZE PS PM PB
3 3 3 3 
1
1 2 2 1 
NS    4,  3, 1  2,  1, 0
 3 3 3 3 

-6 -5 -3 0 3 5

51
Mamdani Implication

If then rule is represented by

0 0 0 0 0 0
Mamdani Implication
0 1/3 1/3 1/3 1/3 1/3
0
0 1/3 2/3 2/3 2/3 1/3
-1
 R ( x, y)  min(  ( x),  ( y)) -2
0 1/3 2/3 1 2/3 1/3

0 1/3 2/3 2/3 1/3 1/3

-3
0 1/3 1/3 1/3 1/3 1/3
-4
0 0 0 0 0 0
-5

0 1 2 3 4 5 6
E
52
Mamdani Implication

A crisp value of e =3 0/1,0/2,1/3,0/4

0 0

0 1 0
0
0 1 0

0 1 0

0 1 0

0 1 0

2 3
53
Mamdani Implication

A crisp value of e =3 (0/1,0/2,1/3,0/4).PerformComposition

0 0

0 0 1/3 0

0 2/3 0

0 2/3 0

0 2/3 0

0 1/3 0

2 3
54
Mamdani Implication

The projection of this relation on the U axis is the output fuzzy

set
U(u)=0/-6+ 0/-5+ 1/3/-4+ 2/3/-3+ 2/3/-2+ 2/3/-1+ 1/3/0+ 0/1+.. +0/6

This can be done as follows

• For the crisp input e=3, determine the degree of membership  PM (3)  2 / 3
• Form  NS as follows

 NS (u ) if NS (u )   PM (3)  2 / 3
u U : CNS (u )  
 PM (3)  2 / 3 otherwise
or
u U : CNS (u )  min(  PM (3),  NS (u ))  min(2 / 3,  NS (u ))

55
Mamdani Implication

Graphically it can be shown as

PM NS
1 1

-6 -5 -3 0 3 5 -5 -3 0 3 5 6

56
Mamdani Implication

• A Company gives loans using the following rules:

• IF Salary is Excellent OR Debts are Small THEN Risk is

Low

• IF Salary is Good AND Debts are Large THEN Risk is

Normal

• IF Salary is Poor THEN Risk is High

57
Mamdani Implication
The membership functions for the linguistic variables Salary,
Debts, and Risk are given as in ₨ K are

58
Mamdani Implication

Jim has applied for a loan: his salary is ₨ 55K and his
debts amount to ₨ 50K.
Poor Good Excellent

10 50 60 90 100 120 Salary

Small Large

10 15 50 60 Debts
Low Medium High

20 40 60 80 Risk

59
Mamdani Implication
risk medium
salary good
debt large

50 100 15 60 20 80

55 50

risk high
salary poor

10 60 60

55

60
Mamdani Implication

20 60 80 100

61
 DEFUZZIFICATION
 Defuzzification is a mapping process from a space of fuzzy
control actions defined over an output universe of discourse into a
space of crisp (nonfuzzy) control actions.

 Defuzzification is a process of converting output fuzzy variable

into a unique number.

 Defuzzification process has the capability to reduce a fuzzy set

into a crisp single-valued quantity or into a crisp set; to convert a
fuzzy matrix into a crisp matrix; or to convert a fuzzy number into
a crisp number.

62
 METHODS OF DEFUZZIFICATION
Defuzzification is the process of conversion of a fuzzy quantity
into a precise quantity.
Defuzzification methods

 Centroid method,
 Weighted average method,
 Max-membership principle,
 Mean-max membership,
 Center of sums,
 Center of largest area,
 First of maxima, last of maxima.

63
METHODS OF DEFUZZIFICATION
CENTROID METHOD

This method is also known as center-of-mass, center-of-area, or

center-of-gravity method. It is the most commonly used
defuzzification method. The defuzzified output x* is defined as

where the symbol  denotes an algebraic integration.

64
METHODS OF DEFUZZIFICATION

65
 METHODS OF DEFUZZIFICATION
Mean of Maxima (MOM):
• The defuzzified result represent the mean value of all actions,
whose membership function reach the maximum.


n

x
x  i 1 i
n

66
 METHODS OF DEFUZZIFICATION
CENTER OF SUMS:
• This method employs the algebraic sum of the individual fuzzy
subsets
• The calculations are very fast
• The defuzzified value x* is given by

67
 METHODS OF DEFUZZIFICATION
CENTER OF LARGEST AREA:
• The output consists of at least two convex fuzzy subsets which are
not overlapping.
• The output is biased towards a side of one membership function.
• The defuzzified value x* is given by

where is the convex subregion that has the largest area making up

68
METHODS OF DEFUZZIFICATION
FIRST OF MAXIMA (LAST OF MAXIMA):

69
Takagi-Sugeno Method

70
Takagi-Sugeno Method

71
 Takagi-Sugeno Method
Takagi, T., & Sugeno, M. (1985). ‘Fuzzy Identification of Systems and its
Applications to Modeling and Control’. IEEE Transactions on Systems,
Man and Cybernetics. Volume No. SMC-15 (No.1) pp 116-132.
Tomohiro Takagi and Michio Sugeno recognised two important
points:

• Complex technological processes may be described in terms of interacting, yet

simpler sub processes. This is the mathematical equivalent of fitting a piece-
wise linear equation to a complex curve.

• The output variable(s) of a complex physical system, e.g. complex in the

sense it can take a number of input variables to produce one or more output
variable(s), can be related to the system's input variable in a linear manner
provided the output space can be subdivided into a number of distinct regions.

72
 Takagi-Sugeno Method
• Literature on conventional control systems has suggested that a complex non-
linear system can be described as a collection of subsystems that were combined
based on a logical (Boolean) switching system function.

• In realistic situations such disjoint (crisp) decomposition is impossible, due to

the inherent lack of natural region boundaries in the system, and also due to the
fragmentary nature of available knowledge about the system.

Takagi and Sugeno (1985) have argued that in order to develop a

generic and simple mathematical tool for computing fuzzy
implications one needs to look at a fuzzy partition of fuzzy input
space. In each fuzzy subspace a linear input-output relation is
formed. The output of fuzzy reasoning is given by the values
inferred by some implications that were applied to an input.

73
Takagi-Sugeno Method
Consider a domain where all fuzzy sets are associated with linear
membership functions.

Let us denote the membership function of a fuzzy set A as mA(x), x

X. All the fuzzy sets are associated with linear membership
functions. Thus, a membership function is characterised by two
parameters giving the greatest grade 1 and the least grade 0.

The truth value of a proposition “x is mA and y is mB” is expressed

as

x is μ A and y is μ B  μ A  μB

74
Takagi-Sugeno Method
Takagi and Sugeno have described a fuzzy implication R is of the
format:
R: if (x1 is mA(x1),… xk is mA(xk)) then y = g(x1, …, xk)
Where:
Variable of the consequence whose
y value is inferred

x1, …, xk Variables of the premise that appear

also in the part of the consequence

A(x1),….. Fuzzy sets with linear membership

functions representing a fuzzy
A(x1) subspace in which the implication R
can be applied for reasoning.
Logical function connects the
f propositions in the premise.
Function that implies the value of y
g when x1,…. xk satisfies the premise.

75
 Takagi-Sugeno Method
Consider a system with n implications (rules); the variable of
consequence, y, will have to be notated for each of these
implications, leading to yi variables of consequence. There are
three stages of computations in Takagi-Sugeno controllers:
• FUZZIFICATION: Fuzzify the input. For all input variables compute the
implication for each of the rules;
• INFERENCE or CONSEQUENCES: For each implication compute the
consequence for a rule which fires. Compute the output y for the rule by using
the linear relationship between the inputs and the output (y = p0 + p1x1 + … +
pkxk.).
• AGGREGATE (& DEFUZZIFICATION): The final output y is inferred from n-
implications and given as an average of all individual implications yi with
weights |y= yi |:
• y = (S |y= yi | * yi )/ S |y= yi |

76
Takagi-Sugeno Method
Consider the following fuzzy implications (or rules) R1,R2, R3 used
in the design of a Takagi-Sugeno controller:

R1  If x1 is small1 & x2 is small2 then y(1) =x1+x2

R2 If x1 is big1 then y(2) = 2x1
R3 If x2 is big2 then y(3) =3x2

where y (i) refers to the consequent variable for each rule labelled Ri
and x1 and x2 refer to the input variables that appear in premise of
the rules.

77
Takagi-Sugeno Method
The membership function for small1, small2, big1 and big2 are
given as follows x Small1 Small2 Big1 Big2
0 1 1 0 0
1 0.938 0.875 0 0
2 0.875 0.75 0 0
3 0.813 0.625 0 0.125
4 0.75 0.5 0 0.25
5 0.688 0.375 0 0.375
6 0.625 0.25 0 0.5
7 0.563 0.125 0 0.625
8 0.5 0 0 0.75
9 0.438 0 0 0.875
10 0.375 0 0 1
11 0.313 0 0.1 1
12 0.25 0 0.2 1
13 0.188 0 0.3 1
14 0.125 0 0.4 1
15 0.063 0 0.5 1
16 0 0 0.6 1
17 0 0 0.7 1
18 0 0 0.8 1
19 0 0 0.9 1
20 0 0 1 1
78
Takagi-Sugeno Method
The membership function for small1, small2, big1 and big2 are
given as follows Takagi-Sugeno Example pp 117

1
0.9
0.8
Membership Function

0.7
Small1
0.6
0.5
Small2
0.4
0.3 Big1
0.2
0.1 Big2
0
0 5 10 15 20
Input x
79
 Takagi-Sugeno Method
Let us compute the FINAL OUTPUT y for the following values:
x1 = 12 & x2 = 5
using Takagi and Sugeno‟s formula:
y = (S |y= yi | * yi )/ S |y= yi |
where |y= yi | stands for the truth value of a given proposition.

FUZZUFICATION: We have the following values of the

membership functions for the two values x1 = 12 & x2 = 5:

Small1 Small2 Big1 Big2

x1= 12 0.25 0 0.2 1
x2= 5 0.688 0.375 0 0.375
80
Takagi-Sugeno Method
INFERENCE & CONSEQUENCE: x1 = 12 & x2 = 5

Truth
Rule Premise 1 Premise 2 Consequen Value
Min (Premise 1
ce & Premise2)

R1 Small1 ( x1 ) Small2 ( x2 y(1)= x1 + x2 Min(0.25,

= 0.25 )= 0.375 =12+5 0.375)=0.2
5
R2 Big1 ( x1 ) = y(2)= 2x1 0.2
0.2 = 24

R3 Big2 ( x2 ) = y(3)= 3x2 0.375

0.375 = 15

81
 Takagi-Sugeno Method
AGGREGATION (&DEFUZZIFICATION):
x1 = 12 & x2 = 5
y = (S |y= yi | * yi )/ S |y= yi |
Using a Centre of Area computation for y we get:

 y  y (i )
* y (i )
i 1,3
y 
 y  y (i )
i 1,3
0.25 *17  0.2 * 24  0.375 *15
y   17.8
0.25  0.2  0.375

82
Takagi-Sugeno Method
In Sugeno‟s model, each rule has a crisp output, the overall input is
obtained by a weighted average – this avoids the time-consuming
process of defuzzification required in a Mamdani model. The
weighted average operator is replaced by a weighted sum to reduce
computation further.

83
 Takagi-Sugeno Method
Function approximation with a Sugeno model
The Sugeno model can be viewed as a collection of locally approximating linear
models
y

 A x  :   0,1
y  a2 x  b2
y  a1 x  b1

A  Small , Medium, Large

y  a3 x  b3

x Rules
 IF x is Small THEN y  a1 x  b1
Small Medium Large

 IF x is Medium THEN y  a2 x  b2
 IF x is Big THEN y  a3 x  b3
x

 x   0.2 ; 0.8 ; 0
84
 Further Reading
1. Y. F. Li, C. C. Lau, “Development of Fuzzy Algorithms for Servo
Systems”, IEEE Int. Conf. on Robotics and Automation,
Philadelphia, USA, April 24-29, 1988

2. Li, Y.F. and Lan, C.C. “Development of Fuzzy algorithms for

Servo systems”, IEEE Control Systems Magazine, Vol.9, No.3,
pp.65-72, 1989

3. Hao Ying, “General Takagi-Sugeno fuzzy systems with simplified

linear rule consequent are universal controllers, models and
filters” Journal of Information Sciences 108 (1998) 91 107

85
 References
1. T. J Ross, Fuzzy Logic With Engineering Application, Wiley

2. Sivanandam, Deepa, Principles of Soft Computing, Wiley

86
87