TELKOM
NIKA
, Vol.9, No.2, August 20
11, pp. 201
~202
ISSN: 1693-6
930
accredited by D
G
HE (DIKTI
), Decree No: 51/Dikti/Kep/2010
201
Progress in A
r
tificial Intellig
ence Techniques: from
Brain to Em
otion
(Tole Sutikno)
Progress in Artificial Int
e
lligence
Techn
i
ques: from
Brain to Emotion
Tole Sutikno
1
, Mochammad Fac
t
a
2
, G.R. Arab M
a
r
kadeh
3
1
Departme
n
t of Electrical En
gi
neer
ing, Un
iver
sitas Ahmad D
ahl
an, Yog
y
a
k
arta, Indon
esia
2
Departme
n
t of Electrical En
gi
neer
ing, Un
iver
sitas Dip
one
go
ro, Semaran
g
, Indon
esi
a
3
Departme
n
t of Electrical En
gi
neer
ing, Un
iver
sit
y
Sh
ahrek
or
d, Shahrek
ord,
Iran
E-mail: tole@
e
e.uad.ac.i
d
Artificial Intelligen
ce (AI) te
chni
que
s, e.g
.
ex
pert s
y
s
t
em (ES), fuz
zy
logic (FL
)
, artificial
neural network
(ANN),
geneti
c
algo
rithm (G
A
)
, particl
e swa
r
m optimizati
on (PSO)
a
nd
biologi
cally i
n
spi
r
ed
(BI)
have re
ce
ntly been a
ppli
ed wi
dely in
power
elect
r
oni
cs and
motor
drives. Th
e aim of the AI
is to model natural o
r
huma
n
intelligen
ce
in a comp
uter to think smartly
like a hum
an
[1], [2].
The
next form of AI i
s
the embedded A
I
cont
roller system whi
c
h
has ability
in
l
earni
ng,
self-o
rg
ani
zin
g
, and
self-adaptin
g. Ha
d bee
n abl
e
to solve
co
mmon a
nd
compl
e
x con
t
rol
probl
em, the AI technique
in comp
utatio
nal intelli
gen
ce applied in
wide ap
plication of indust
r
i
a
l
pro
c
e
s
s cont
rol, roboti
cs,
automate
d
planni
ng a
n
d
sche
duling,
game
s
, hyp
e
rme
d
ia, ima
ge
pro
c
e
ssi
ng,
pattern
s re
cognition (ha
ndwriting,
sp
eech, and facial
), l
ogisti
cs, data mi
ning,
medici
ne an
d
healthcare, space and di
a
gno
stic techn
o
logy [1].
Each
AI meth
od h
a
s its
own uni
que
ne
ss and
c
haracte
ristics. T
he E
S
and
FL te
chniqu
e
s
tend to mimi
c the beh
aviou
r
al natu
r
e
of the hum
an
b
r
ain an
d ba
se
on the
rule
s; the NN is
mo
re
gene
ric in
n
a
ture a
nd te
nds to patte
rn directly to the biologi
cal NN. The
GAs and t
he
evolutiona
ry comp
utation
t
e
ch
niqu
es are ba
se
d o
n
p
r
inci
ple
s
of
g
enetics. Ba
si
cally, GA
sol
v
es
optimizatio
n
probl
em
s through a
se
arching p
r
o
c
e
ss
to
find the fittest as
a su
rv
ivor for the b
e
st
solutio
n
s. Am
ong
all the
sub b
r
an
ch
es
of AI, the NN and
FL
app
ear to
be
mo
st u
s
e
s
for hi
gh-
perfo
rman
ce
motor d
r
ive
s
. Ho
weve
r t
here
are
ma
ny other fe
e
d
forward an
d re
cu
rre
nt
NN
topologi
es which requi
re
systemat
i
c
exploratio
n for their a
ppl
ication
s
[3]. In advance, the
powerful intel
ligent control
and e
s
timatio
n
tech
ni
qu
es
are
dynami
c
a
lly develope
d
throu
gh hyb
r
id
AI systems
such
as n
e
u
r
o
–
fuzzy, neuro
–gen
et
ic, and
neuro–fu
zzy
–gen
etic sy
st
ems. Th
e PSO
as a po
pulati
on-b
a
sed sto
c
ha
stic o
p
timizati
on te
chni
que ha
s be
e
n
develop
ed
sin
c
e 19
95 a
nd
inspi
r
ed by
so
cial be
hav
ior of bird flocking
o
r
fish sch
oolin
g
[4]. PSO
as evolutio
n
a
ry
comp
utation t
e
ch
niqu
es
sh
are
s
ma
ny si
milaritie
s
with
GA, but PSO offer ea
sy
impleme
n
tation
with few adju
s
table gai
ns.
PSO is con
s
i
dere
d
as a f
a
st-d
evelopi
n
g
resea
r
ch to
pic and a
ppli
e
d
su
ccessfully in optimizat
io
n function, a
r
tificial neu
ral
netwo
rk
traini
ng, and fu
zzy
system control
[5]. The biol
ogical di
spo
s
itions of a
n
i
m
als a
nd mi
mics
bio m
e
cha
n
ism
s
h
a
ve inspi
r
ed t
he BI
system. Since 1990
s, the NN tech
nol
ogy has be
co
me one of most attra
c
tive topics for the
sci
entific com
m
unity, and growth
rapi
dly in diffe
rent an
d variou
s app
lication
s
[1], [2], [5].
Re
cently, re
searche
r
s hav
e develop
ed
a com
put
atio
nal model of
emotional le
a
r
ning in
mammalia
n b
r
ain. Ba
sed
o
n
the
cog
n
itively motive op
en loo
p
mo
d
e
l, brai
n emo
t
ional lea
r
nin
g
based intellig
ent cont
rolle
r (BELBIC) was intro
d
u
c
ed
for the first time in 2004 [
6
]. Basically, the
brain
emotio
n
a
l learning
(B
EL) is divided
into
two
part
s
: amygd
a
la
and o
r
bitofro
n
tal co
rtex. T
h
e
amygdala i
s
a pa
rt of the
brain th
at mu
st be
re
spo
n
s
ible
fo
r
p
r
o
c
essing emoti
ons prim
arly and
corre
s
p
ond
with the orbit
o
frontal corte
x
, thal
amus, and sensory
input co
rtex
in the network
model. T
he o
r
bitofro
n
tal
cortex receive
s
in
puts from
the
corti
c
al
area
s
and
th
e amygd
a
la
and
respon
sibl
es
for the
rea
c
tion to
chan
g
e
the
c
onting
ency
of em
otions. E
r
ror of
the
expe
cte
d
reward o
r
pu
nishm
ent a
n
d
the lo
ss
of le
arnin
g
in th
e
amygdala
will
be ma
nag
ed
by orbitof
r
on
tal
c
o
rtex [1], [2],
[5]
.
Duri
ng the p
a
st few years, the BELBIC
ha
s been
su
ccessfully employed for makin
g
deci
s
io
ns an
d controlling
in a simple li
near
system
s and non
-lin
ear sy
stem
s su
ch a
s
spe
ed
control of a
p
e
rma
nent m
a
gnet syn
c
h
r
o
nou
s moto
r (PMSM), auto
m
atic voltag
e
reg
u
lator (A
VR)
system, flight
control, posi
t
ion tracking
and swin
g da
mping control
of single inp
u
t multi output
overhe
ad t
r
a
v
eling
cra
ne,
wa
shin
g ma
chine, a
u
to
mot
i
ve su
sp
en
sio
n
control
syst
em, micro
-
he
at
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 9, No. 2, August 20
11 : 201 – 20
2
202
excha
nge
r, v
entilating
and
air conditio
n
i
ng
cont
rol [1]
,
[2], [5]. In furthe
r d
e
velo
pment, BELB
I
C
was implemented on fiel
d
program
m
abl
e gate array
(FPGA) for
controlling
a mobile crane
in
a
model free a
nd embe
dde
d manne
r. In 2008, BELBI
C method fo
r elect
r
ical drive control
was
started
an
d
a supe
rior cont
rol
ch
ara
c
teri
stic
wa
s bo
rn
with fast
resp
on
se, si
mple
impleme
n
tation, and ro
bu
stness to distu
r
ban
ce
s [7].
In [1], the co
ntrolle
r was
use
d
for first
time to
cont
rol a
n
IM d
r
i
v
e and i
n
vest
igated it
s
indep
ende
nt
of the pa
ram
e
ters variatio
ns, e
s
pe
cia
lly
rotor re
si
sta
n
ce. Al
so th
e co
ntroll
er
wa
s
impleme
n
ted
for som
e
other ele
c
tri
c
d
r
ives
su
ccessfully [8]. Ba
sed o
n
the a
bove mentio
ned
eviden
ce of the emotio
nal
cont
rol ap
proache
s in co
mputer
and
c
ontrol e
ngin
e
e
ring, it can
be
con
c
lu
ded th
at the ap
plication of e
m
ot
ion in
sy
ste
m
s coul
d
by its
sim
p
le an
d
uniq
ue co
n
t
rol
desi
gn, overcome the probl
ems of non
-li
near
sy
stem,
manufa
c
turi
n
g
imperfe
ctio
ns, acce
ptabl
y.
The
results i
ndicate the
ability of BELBI
C to
cont
rol unknown non-linear
dynami
c
system
s. The
impleme
n
tation of the em
otional
controller sho
w
s good co
ntrol perfo
rman
ce in
terms o
r
robu
stne
ss an
d a
daptability in
high
auto
l
earning fe
ature.
Flexibility is o
ne of BELBI
C’s
cha
r
a
c
teri
stics and it ha
s the cap
a
city to choo
se th
e most-favo
u
r
ed respon
se
. Therefo
r
e, the
BELBIC can
be ea
sily ado
pted for niche
mechat
roni
cs and in
du
stri
al appli
c
ation
s
.
Among
st sel
e
cted
pa
pers in thi
s
editio
n
, seve
nth of
them
are
a
p
p
roved
to
prese
n
t AI
techni
que
s in
various a
ppli
c
ation
s
. The
s
e pape
rs a
r
e
expecte
d to enco
u
ra
ge the
rese
arch of AI
impleme
n
tation to create a
better tech
no
logy for the future.
Referen
ces
[1]
Markad
eh GR
A, Dar
y
ab
eig
i
E, Lucas C, R
ahma
n
MA. Spee
d a
nd F
l
u
x
Contro
l of Ind
u
ction M
o
to
r
s
Using Em
otio
n
a
l Intel
lig
ent
Contro
ller.
IEE
E
T
r
ansactio
n
s
on
Ind
u
stry Applic
atio
ns.
201
1; 47(3):
112
6-11
35.
[2]
Z
a
rchi HA, Da
r
y
a
b
e
i
gi E, Ma
rkade
h GRA, Soltan
i J.
Emo
t
iona
l control
l
e
r
(BELBIC) ba
sed DT
C for
Encod
e
rless Synchro
n
o
u
s Reluct
ance
M
o
tor
driv
es
. Po
w
e
r E
l
ectronics, Drive S
y
stems a
n
d
T
e
chnolog
ies Confer
ence
(P
EDST
C)
.
T
ehran, Iran. 201
1; 478-
483.
[3]
Bose BK. Nonli
near Net
w
o
r
k Appl
icatio
ns in
Po
w
e
r Electro
n
i
cs and Motor
Drives-An Intro
ductio
n
an
d
Perspective.
IEEE Trans. on Ind. Electronics
. 2007; 54(
1): 14-33.
[4]
Kenn
ed
y J, Eberh
a
rt RC.
Particle Sw
arm Opti
mi
z
a
t
i
on
. In: Proceed
ings-IEEE
Internation
a
l
Confer
ence
on
Neura
l
Net
w
or
ks (ICNN).
Perth, Australia. 1
9
95; IV: 1942–
1
948.
[5]
Dorrah HT
, El-Garh
y
AM, El-
S
him
y
ME. PS
O-BELBIC Scheme F
o
r T
w
o-
Cou
p
le
d Distil
l
a
tion C
o
lum
Process.
Journ
a
l of Advanc
ed
Researc
h
. 201
1; 2: 73-83
[6]
Lucas
C, Sha
h
mirzad
i D,
a
nd Sh
eik
hol
esl
a
mi
N. Intro
d
u
c
ing BE
LBIC:
Brain Em
otio
n
a
l L
ear
nin
g
Based Intelligent Control.
Int. J. Intell. Au
tomat. Soft Com
p
ut.
2004; 10(1):
11–
22.
[7]
Rahm
an MA,
Milasi
RM,
Lu
cas C, Arr
abi
BN
, a
n
d
Ra
d
w
a
n
T
S
. Implem
entatio
n
of Emotio
n
a
l
Contro
ller
for I
n
terior
Perma
n
ent Ma
gn
et S
y
nchro
nous
Mot
o
r Driv
e.
IEEE
T
r
ans. Ind.
A
ppl
. 20
08;
44(5): 14
66
–14
76.
[8]
Dar
y
a
bei
gi E, Markad
eh GR
A, Lucas C.
Em
otional controller (BE
L
BIC)
for Electric Drives - A review
.
36
th
Annu
al C
onfere
n
ce o
n
IEEE Industria
l
Electr
onics S
o
ciet
y
(IECON
). Glendal
e, Arizon
a. 201
0;
290
1-29
07.
Evaluation Warning : The document was created with Spire.PDF for Python.