Você está na página 1de 26

Introduction to

Softcomputing

Ir. W. Herulambang

Ir. W. Herulambang -- SoftComputing

AI and Softcomputing

User

AI: predicate logic and symbol


manipulation techniques
Inference
Engine

Question

Explanation
Facility

Knowledge
Engineer
Human
Expert

ec af r et nI r es U

Response

Global
Database

KB: Fact
rules

Knowledge
Acquisition
Expert Systems
Ir. W. Herulambang -- SoftComputing

AI and Softcomputing

ANN
Learning and
adaptation

Fuzzy Set Theory


Knowledge representation
Via
Fuzzy if-then RULE

Genetic Algorithms
Systematic
Random Search
4

Ir. W. Herulambang -- SoftComputing

AI and Softcomputing

ANN
Learning and
adaptation

Fuzzy Set Theory


Knowledge representation
Via
Fuzzy if-then RULE

Genetic Algorithms
Systematic
Random Search
5

AI
Symbolic
Manipulation

Ir. W. Herulambang -- SoftComputing

AI and Softcomputing

cat
cut

Animal?

Neural character
recognition
knowledge

Ir. W. Herulambang -- SoftComputing

cat

From Conventional AI to
Computational Intelligence

Conventional AI:
Focuses on attempt to mimic human
intelligent behavior by expressing it in
language forms or symbolic rules
Manipulates symbols on the
assumption that such behavior can be
stored in symbolically structured
knowledge bases (physical symbol
system hypothesis)

Ir. W. Herulambang -- SoftComputing

Soft Computing :
Techniques and Application
Fuzzy Logic

Artificial Neural Network

Probabilistic Reasoning
(Genetic Algorithms
& Evolution Prog. )
8

Ir. W. Herulambang -- SoftComputing

Hybrid
Models

From Conventional AI to
Computational Intelligence

Perceptions

Intelligent Systems

Sensing Devices
(Vision)

Actions

Machine
Learning
Task
Generator

Inferencing
(Reasoning)

Natural
Language
Processor

Knowledge
Handler

Planning

Mechanical
Devices

Data
Handler

Knowledge
Base
Ir. W. Herulambang -- SoftComputing

Neural Networks

10

Ir. W. Herulambang -- SoftComputing

Neural Networks
f

yp(k+1)

z-1

u(k)

z-1

e(k+1)

y^p(k+1)

z-1

z-1

11

Ir. W. Herulambang -Parameter


SoftComputing
Identification - Parallel

Neural Networks

Control
Current-driven
magnetic field

Controller

Iron ball

Ball-position
sensor
12

Ir. W. Herulambang -- SoftComputing

Neural Networks

13

Ir. W. Herulambang -- SoftComputing

Neural Networks

Experimental Results

Feedback control
only

14

Feedback with ANN


Feedforward controller

Feedback with fixed gain feedforward control


Ir. W. Herulambang -- SoftComputing

Fuzzy Sets Theory

15

What is fuzzy thinking


Experts rely on common sense when they solve the
problems
How can we represent expert knowledge that uses
vague and ambiguous terms in a computer
Fuzzy logic is not logic that is fuzzy but logic that is
used to describe the fuzziness. Fuzzy logic is the
theory of fuzzy sets, set that calibrate the vagueness.
Fuzzy logic is based on the idea that all things admit
of degrees. Temperature, height, speed, distance,
beauty all come on a sliding scale.
Jim is tall guy
It is really very hot today
Ir. W. Herulambang -- SoftComputing

Fuzzy Set Theory

Communication of fuzzy idea


This box is
too heavy..

16

Therefore, we
need a lighter
one

Ir. W. Herulambang -- SoftComputing

Fuzzy Sets Theory

Boolean logic

Fuzzy logic

17

Uses sharp distinctions. It forces us to draw


a line between a members of class and non
members.
Reflects how people think. It attempt to
model our senses of words, our decision
making and our common sense -> more
human and intelligent systems

Ir. W. Herulambang -- SoftComputing

Fuzzy Sets Theory

No

Classical Set vs Fuzzy set

Name

Height
(cm)

Degree of Membership
of tall men

Crisp

Fuzzy

Boy

206

Martin

190

Dewanto

175

0.8

Joko

160

0.7

Kom

155

0.4

18

Ir. W. Herulambang -- SoftComputing

Fuzzy Sets Theory

Classical Set vs Fuzzy set

Membership value

Membership value

0
175

Height(cm)

175

Height(cm)

Universe of discourse
19

Ir. W. Herulambang -- SoftComputing

Fuzzy Sets Theory

Fuzzy Expert Systems


Kecepatan (KM)

Jarak (JM)
Posisi Pedal Rem (PPR)

20

Ir. W. Herulambang -- SoftComputing

Fuzzy Sets Theory

Membership function

Sangat Lambat

Sangat Dekat

Cukup

Cepat

Lambat

40

60

Kecepatan (km/jam)

21

Injak Penuh

Sedang

Agak Jauh

Agak
Dekat

Cepat Sekali
20

PPR

JM

KM

Injak
Sedikit

Injak Agak
Penuh
Injak Sedikit
Sekali

Jauh Sekali
80 0

Injak
Sedang

10

Jarak (m)

Ir. W. Herulambang -- SoftComputing

20

30

Posisi pedal rem (0)

40

Fuzzy Sets Theory

Fuzzy Rules

Aturan 1:
Bila kecepatan mobil cepat sekali dan jaraknya sangat dekat maka pedal
rem diinjak penuh
Aturan 2:
Bila kecepatan mobil cukup dan jaraknya agak dekat maka pedal rem
diinjak sedang
Aturan 3:
Bila kecepatan mobil cukup dan jaraknya sangat dekat maka pedal rem
diinjak agak penuh

22

Ir. W. Herulambang -- SoftComputing

Fuzzy Sets Theory


Fuzzy Expert Systems

Aturan 1:
Cepat Sekali

20

40
60
80
Kecepatan (km/jam)

23

Sangat Dekat

Injak Penuh

3
4
Jarak (m)

10

Ir. W. Herulambang -- SoftComputing

20

30

Posisi pedal rem (0)

40

Fuzzy Sets Theory


Fuzzy Expert Systems

Aturan 2:
Cukup

20

40
60
80
Kecepatan (km/jam)

24

Agak Dekat

Injak Sedang

3
4
Jarak (m)

10

Ir. W. Herulambang -- SoftComputing

20

30

40

Posisi pedal rem (0)

Fuzzy Sets Theory

Fuzzy Expert Systems

Aturan 3:
Cukup

20

40
60
80
Kecepatan (km/jam)

25

Sangat Dekat

Injak Agak
Penuh

3
4
Jarak (m)

10

Ir. W. Herulambang -- SoftComputing

20

30

40

Posisi pedal rem (0)

Fuzzy Sets Theory

Fuzzy Expert Systems


MOM : PPR = 200
MOM

COA

10x0,2+20x0,4
COA : PPR =

10

20

30

40

0,2+0,4
= 16,670

Posisi pedal rem (0)

26

Ir. W. Herulambang -- SoftComputing

Você também pode gostar