Indonesi
an
Journa
l
of El
ect
ri
cal Engineer
ing
an
d
Comp
ut
er
Scie
nce
Vo
l.
13
,
No.
1
,
Jan
uar
y
201
9
,
pp.
2
7
9
~
2
85
IS
S
N: 25
02
-
4752, DO
I: 10
.11
591/ijeecs
.v1
3
.i
1
.pp
2
7
9
-
2
85
279
Journ
al h
om
e
page
:
http:
//
ia
es
core.c
om/j
ourn
als/i
ndex.
ph
p/ij
eecs
Improvi
ng steeri
ng conve
rgence i
n auton
om
ous veh
icle
steerin
g contr
ol
Amir
Ash
r
af
Moham
ad,
F
adhl
an H
af
iz
helm
i Kam
aru Z
ama
n
, F
az
li
na
Ah
m
at
Rusl
an
Facul
t
y
of Electr
ic
a
l
Eng
ineeri
ng
,
Univer
si
ti T
ekn
ologi
Mar
a, 404
50
Shah
Alam,
Sela
ngor
,
Ma
lay
si
a
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
A
ug
2
0
, 201
8
Re
vised Oct
28
, 2
018
Accepte
d Nov
1
2
, 201
8
Stee
ring
cont
ro
l
is
a
cri
t
ical
design
e
le
m
e
nt
in
aut
onom
ous
vehi
cle
deve
lopment
sin
ce
it
wi
ll
d
et
erm
ine
wheth
er
the
vehi
c
le
c
an
nav
i
gat
e
saf
e
l
y
or
not.
For
the
prototy
pe
o
f
Ui
TM
Autonom
ous
Vehic
l
e
0
(Ui
TM
AV
0),
Vexta
m
otor
is
u
sed
to
con
trol
th
e
ste
eri
ng
wher
e
as
Puls
e
W
idt
h
Modulat
ion
(PW
M)
signal
i
s
responsible
to
drive
the
m
oto
r.
Ho
weve
r,
b
y
using
PWM
signal
it
is
diffic
ult
to
conve
rg
e
t
o
the
desire
d
ste
eri
ng
angle
and
furthe
rm
ore
ti
m
e
t
ake
n
for
st
ee
ring
angle
to
c
onver
ge
is
m
uch
longe
r
.
Thus,
P
roporti
ona
l
Inte
gra
l
Deri
vative
(PID
)
has
b
ee
n
int
rodu
ce
d
in
thi
s
aut
onom
ous
vehi
cl
e
stee
r
ing
cont
ro
ller
to
improve
th
e
conv
erg
en
ce
o
f
the
stee
r
ing.
Mea
nwhile
a
m
ic
roc
ontroller
was
used
to
cont
rol
the
Vex
ta
Motor
dire
ction
and
per
form
the
calc
ul
at
ion
o
f
the
desire
d
ste
eri
ng
angle.
Sim
ula
ti
on
r
esult
s
show
ed
PID
cont
roller
show
e
d
bet
ter
ti
m
e
t
ak
e
n
and
pre
ic
ison
of
succ
essful
c
onver
gence
of
the
desire
d
stee
ring
ang
l
e
comp
are
d
t
o
the
PW
M
cont
roller
.
Anal
y
sis
resul
t
s
show
ed
tha
t
PID
cont
roll
e
r
signifi
ca
n
tly
red
uce
th
e
over
shooting
of
stee
ring
angl
e
a
nd
signifi
c
ant
l
y
improve
the
t
ime
ta
k
en
for
conve
r
gen
ce b
y
up
to
37
sec
onds
faste
r
th
an
PW
M c
ontroller
in U
iT
M AV
0.
Ke
yw
or
ds:
Ardu
i
no m
ega
PI
D
contr
oller
P
W
M
Vex
ta
m
oto
r
Copyright
©
201
9
Instit
ut
e
o
f Ad
vanc
ed
Engi
n
ee
r
ing
and
S
cienc
e
.
Al
l
rights re
serv
ed.
Corres
pond
in
g
Aut
h
or
:
Fadhla
n Ha
fizhelm
i
Kam
aru
Zam
an
,
Faculty
of Elec
tric
al
Engineer
ing
,
Un
i
ver
sit
i Te
knol
og
i M
a
ra,
40450 S
hah A
l
a
m
, S
el
ango
r,
Ma
la
ysi
a
.
Em
a
il
:
fad
hlan
@salam
.u
itm
.e
du.m
y
1.
INTROD
U
CTION
Transp
or
t
or
t
ran
s
portat
io
n
i
s
the
m
ov
em
e
nt
of
hum
ans,
anim
al
s
and
goods
from
on
e
loc
at
io
n
to
an
othe
r
.
Tra
ns
po
rt
is
a
c
riti
cal
nee
d
f
or
pe
op
le
i
n
rural
a
r
ea,
es
pecial
ly
car.
T
his
is
be
cause,
in
r
ur
al
areas,
a
car
is
the
m
a
in
trans
portat
io
n
to
go
to
w
or
k.
The
refo
re,
hi
gh
dem
and
from
local
s
will
increase
the
nu
m
ber
of cars
in
t
he
c
ountry a
nd th
u
s w
il
l i
ncr
ea
se
the num
ber
of
acci
den
ts
occ
urs.
Nowa
days,
m
a
ny
co
un
trie
s
c
om
pe
te
against
each
ot
her
in
dev
el
op
i
ng
t
he
ir
te
chnolo
gies
especial
ly
in
ve
hicle
s
s
uc
h
as
an
im
ple
m
ent
at
ion
of
t
he
dr
i
ver
le
ss
ve
hicle
te
ch
no
l
ogy
[
1].
T
he
a
ut
onom
ou
s
ve
hicle
has
al
rea
dy
bee
n
s
ta
rted
an
d
bec
om
es
m
or
e
adv
ance
d
si
nce
f
ifty
ye
ars
ago
[2
]
.
It
has
bee
n
sta
rted
by
Ca
rn
e
gie
Me
ll
on
Un
i
versi
ty
(CMU)
in
the
1980s
in
th
ei
r
pr
ese
ntati
on
of
the
a
utonom
ou
s
veh
ic
le
t
hat
can
dr
i
ve
without
a
dr
ive
r
[
3].
The
n,
the
a
utono
m
ou
s
veh
ic
l
e
te
ch
nolo
gy
con
ti
nue
s
by
Goo
gle
an
d
com
bin
e
with
the
bes
t
eng
i
neer
s
f
r
om
D
ARP
A
that
was
le
d
by
Sta
nford
U
niv
e
rsity
P
rofess
or
S
ebasti
an
Th
run
to
im
ple
m
ent
fu
ll
y
interact
ion
of
the
sel
f
-
dri
ving
cars
[
4].
Mo
r
eov
e
r,
i
n
20
12s
Goo
gle
aut
onom
ou
s
ve
hicle
pro
j
ect
has
r
e
ached
their
go
al
w
hich
is t
heir
sel
f
-
dri
ving
ve
hicle
can dr
i
ve
m
or
e
than 3
00,
000
km
with no
acc
ident
[1
]
.
The
Au
to
nom
ou
s
Ve
hicle
S
yst
e
m
is
the
new
te
c
hnolog
ie
s
that
has
be
en
im
ple
m
ented
with
t
he
con
ce
pt
of
dr
iv
el
ess
car
.
T
he
veh
ic
le
s
a
ble
t
o
m
ake
decisi
on
on
it
s
own
suc
h
as
wh
et
her
to
t
urn
le
ft
or
righ
t
a
t
the
intersect
io
n.
T
his
is
ac
hieved
with
the
help
of
a
co
m
pu
te
r
that
ac
t
as
the
br
ai
n
that
con
tr
olli
ng
this
dr
i
veless
veh
ic
le
.
I
n
a
dd
it
io
n,
V
e
xta
m
oto
r
was
us
e
d
t
o
m
ov
e
the
ve
hicle
ste
erin
g.
T
he
m
os
t
crit
ic
al
el
e
m
ent
in
the
a
uton
om
ou
s
veh
ic
le
i
s
the
ste
eri
ng
[
5].
T
he
c
onve
ntion
al
ste
e
rin
g
has
been
re
m
ov
ed
su
c
h
a
s
ste
erin
g
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vo
l.
13
, N
o.
1
,
Ja
nu
a
ry 20
19
:
2
7
9
–
2
8
5
280
and
t
he
ste
erin
g
sh
a
ft.
T
ho
se
com
po
ne
nts
ha
ve
bee
n
re
plac
ed
by
the
el
ect
ric
m
oto
r
to
c
ontr
ol
the
direct
ion
of
the
ve
hicle
.
T
her
e
was
a
l
ot
of
ben
e
fit
du
e
to
rem
ov
ing
the
m
echan
ic
al
par
ts
of
t
he
ste
erin
g
s
uc
h
as
the
handlin
g
pe
rfo
rm
ance
can
be
i
m
pr
ov
e
d,
ve
hicle
’s
weig
ht
was
re
du
ce
d,
i
m
pact
fo
rce
to
the
dr
ive
r
in
f
rontal
acci
den
ts ca
n r
edu
ce
and als
o t
her
e
will
b
e a
la
rg
e s
pace i
n ca
bin
[
6]
,
[
7].
The
r
otary
en
cod
e
r
wa
s
us
e
d
to
gi
ve
the
sign
al
to
the
m
ic
ro
con
t
ro
ll
er
in
orde
r
to
con
tr
ol
the
directi
on
of
th
e
veh
ic
le
ste
eri
ng.
The
cal
ib
ra
ti
on
s
of
the
ti
r
es
need
t
o
be
done
fi
rst
to
m
ake
sure
the
ti
r
e
is
at
the cent
er
b
e
fore an
y m
ov
em
ent o
cc
urs.
T
he
m
axi
m
u
m
an
gl
e o
f
the
steeri
ng
f
or
UiTM
AV0
is
25° d
e
gr
ee le
ft
and
25°
de
gr
ee
r
ig
ht.
T
he
24
V
batte
ry is
ne
eded to
op
e
rate
the V
e
xta m
oto
r
. The A
rduin
o
MEG
A was
us
e
d
to
read
the
r
otary
encoder
a
nd
s
end
t
o
the
MA
TLAB
usi
ng
th
e
US
B
serial
da
ta
.
The
MAT
LAB
i
s
the
co
m
pu
te
r
so
ft
war
e
that
will
be
us
e
d
to
cal
culat
ing
th
e
data
colle
ct
ed
f
ro
m
the
ro
t
ary
encode
r
an
d
IM
U
to
pe
rfor
m
ing
the m
app
in
g.
P
W
M
si
gn
al
was
us
e
d
to
c
on
t
ro
ll
in
g
the
directi
on
of
t
he
ste
erin
g.
T
he
re
we
re
a
fe
w
prob
le
m
s
encou
ntere
d
w
it
h
us
ing
P
W
M
wh
ic
h
is
the
pr
eci
sio
n
of
the
directi
on
of
ste
erin
g
of
UiTM
Au
to
nom
ou
s
Veh
ic
le
.
This
sign
al
was
not
preci
se
because
the
c
onve
r
ge
nce
to
desire
d
ou
t
pu
t
val
ue
was
not
ac
hievab
le
durin
g
the
re
quire
d
per
i
od
of
tim
e.
Be
side
s,
the
tim
e
resp
on
ds
to
obta
in
the
desire
d
ou
t
pu
t
val
ue
of
the
ste
ering
unti
l
s
ta
ble
was
ve
ry
slow
b
y usin
g
t
he
P
WM
sig
na
l.
The
vibrat
io
ns
al
s
o
oc
cu
r
wh
il
e
m
ai
ntaining
t
he
syst
e
m
in
orde
r
to
get
the
ou
tpu
t
value.
P
WM
sig
nal
only
send
s
a
high
a
nd
lo
w
ou
t
pu
t
si
gnal
to
t
he
Vex
ta
m
oto
r
dr
ive
r
to
co
nt
r
ol
the
directi
on
with
ou
t
hav
i
ng
a
ny
cal
culat
ion
m
et
ho
d.
Pr
e
vi
ou
sly
,
P
I
D
ha
s
bee
n
i
m
ple
m
ented
as
pa
rt
of
t
he
Zie
gler
-
Ni
cho
ls
base
d
PI
D
co
ntr
ol
syst
e
m
that
con
t
ro
ls
rob
ot
[9
]
,
and
Tel
eo
per
at
ion
Ma
nipulat
or
s
Syst
em
[1
0].
Re
centl
y,
Qi
and
Zha
ng
al
s
o
disc
us
s
a
bout
a
stud
y
on
a
da
ptive
PI
D
contr
ol al
gorithm
b
ased
on RB
F
Neural
N
et
w
ork
[11
]
.
Th
e
w
ork
pr
e
s
ented
in
t
his
pa
per
is
car
ried
ou
t
t
o
a
naly
ze
and
im
pr
ov
e
t
he
c
onve
rg
e
nc
e
of
ste
eri
ng
ang
le
in
t
he
au
tonom
ou
s
co
nt
ro
l
of
UiTM
A
utono
m
ou
s
Ve
hicle
.
The
te
st
area
f
or
this
ca
r
is
arou
nd
Fa
culty
of
Ele
ct
rical
E
ng
i
neer
i
ng,
U
niv
e
rsity
Tek
nolo
gi
MAR
A,
Sh
a
h
Alam
,
Sela
ngor
.
The
sp
ee
d
of
the
veh
ic
le
wh
il
e
pe
rfor
m
i
ng
this
researc
h
will
be
m
a
intai
ned
at
a
cons
ta
nt
sp
eed
of
10KM/H
.
The
m
app
in
g
dista
nce
f
or
this
veh
ic
le
is
on
ly
100
m
et
e
rs.
T
he
obj
ect
i
ve
of
this
rese
arch
is
to
re
duce
ov
e
rsho
ot
in
the
ste
erin
g
con
t
ro
l
wh
il
e
us
in
g
th
e
P
W
M
si
gn
al
.
Ne
xt
is
to
im
ple
m
ent
the
PID
co
ntro
ll
er
an
d
com
par
e
the
s
te
ering
pe
rform
ance
in
te
rm
of
tim
e
to
co
nver
ge
nce
a
nd
occ
urren
ce
of
over
s
hoot
b
et
wee
n
PI
D
an
d
P
W
M.
Pr
e
viously
he
P
ID
con
t
ro
ll
er
was
us
e
d
t
o
s
olv
e
the
pro
blem
i
n
c
on
t
ro
ll
in
g
t
he
desire
d
ou
t
pu
t
of
the
ste
erin
g
a
ng
le
with
th
e
ste
ering
act
uat
or
[8
]
.
Be
si
des
,
it
al
so
can
re
du
ce
t
he
disturbance
s
ha
ppen
ed
on
t
he
cu
r
va
ture
wh
ic
h
in
creases
li
near
ly
res
pec
t
to
ti
m
e
[1
2]. Th
is
c
onve
ntio
nal
P
ID
co
ntr
ol
al
so
ca
n
ov
e
r
com
e
pr
eci
sio
n
w
hen
the angl
e
erro
r
too
la
r
ge
[
13]
.
Ra
ther
tha
n
us
in
g
PID
c
on
t
ro
ll
er,
t
here
al
so
ha
ve
m
any
m
et
ho
ds
for
co
ntr
olli
ng
t
he
ste
ering
[14].
The
rest
of
this
pap
e
r
is
arr
a
ng
e
d
as
f
ollows:
Sect
ion
2
el
aborates
the
re
search
m
et
ho
dolo
gy
e
m
plo
ye
d,
w
hile Sect
ion
3 p
r
esents the
r
es
ul
ts
an
d
rele
van
t
analy
sis. Lastl
y, S
ect
ion 4
conclu
des
t
he pa
per.
2.
RESEA
R
CH MET
HO
D
In
this
cha
pter
will
exp
la
in
de
ta
il
abo
ut
the
m
et
ho
do
l
og
y
of
this
pro
j
ect
.
This
proj
ect
de
velo
pm
ent
consi
sts
of
both
ha
rdwa
re
an
d
the
s
of
t
war
e
el
e
m
ents.
In
order
to
do
the
a
naly
sis
of
this
auto
no
m
ou
s
ve
hicle
m
app
in
g,
t
her
e
w
e
re f
e
w
m
eth
ods t
hat
will
b
e
us
e
d
to
ward
s achie
ving the
project
ou
tc
om
es.
Figure
1
s
how
s
the
fl
ow
c
ha
rt
of
the
pro
j
ect
for
a
naly
zi
ng
i
n
im
pr
ov
i
ng
th
e
co
nv
e
rg
e
nce
of
ste
e
rin
g
ang
le
in
a
uton
om
ou
s
co
ntro
l
of
UiTM
A
ut
onom
ou
s
Ve
hi
cl
e.
This
pro
j
e
ct
basical
ly
us
es
the
ro
ta
ry
e
ncode
r
com
par
e
with
the
set
ang
le
value
from
t
he
MATL
AB
to
con
tr
ol
th
e
ste
ering
a
nd
t
he
directi
on
of
the
auto
no
m
ou
s
v
e
hicle
.
The
data
from
the
r
otary
enc
oder
was
c
ollec
te
d
to
t
he
A
rdu
ino
Me
ga
to
be
conve
rted
int
o
a
de
gree
of
a
ng
le
because
the
r
otary
encode
r
pro
duc
es
500
pu
lse
s
per
ro
ta
ti
on.
T
he
an
gle
val
ue
sent
to
the
Ardu
i
no
Me
ga
f
ro
m
the
MATL
AB
to
be
com
par
e
d
with
the
a
ngle
of
the
ro
ta
ry
enc
od
e
r.
In
this
proces
s,
the
PID
con
t
ro
ll
er is
n
e
eded to
pe
rform
the syst
e
m
.
The
ste
erin
g
w
il
l
turn
to
t
he
righ
t
if
t
he
a
ng
le
val
ue
from
the
MATL
AB
wa
s
bi
gg
e
r
t
han
t
he
a
ngle
of
the
r
otary
e
nc
oder
.
Wh
il
e
t
he
ste
ering
will
turn
to
t
he
le
ft
if
the
a
ngle
val
ue
from
the
MA
TLAB
was
sm
al
le
r
than
the
a
ng
le
of
the
r
otary
encode
r.
T
hen
it
will
retur
n
to
sta
rt
to
com
par
e
a
gain
unt
il
the
value
is
equ
al
.
Fu
rt
her
m
or
e,
t
he
ste
erin
g
will
be
m
ai
ntaine
d
an
d
no
t
m
ov
e
if
the
ang
le
value
f
ro
m
the
MATLAB
wa
s
equ
al
with the
angle
of the
ro
ta
ry en
cod
e
r.
Evaluation Warning : The document was created with Spire.PDF for Python.
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci
IS
S
N:
25
02
-
4752
Impr
ovin
g
ste
e
ring co
n
ver
gence i
n au
t
onom
ou
s
ve
hicle
ste
eri
ng control
(
Amir As
hraf M
ohama
d
)
281
S
t
a
r
t
S
t
o
r
e
t
h
e
d
a
t
a
o
f
a
n
g
l
e
a
n
d
d
i
r
e
c
t
i
o
n
s
t
e
e
r
i
n
g
f
r
o
m
M
A
T
L
A
B
i
n
t
o
A
r
d
u
i
n
o
C
o
m
p
a
r
e
t
h
e
e
n
c
o
d
e
r
a
n
g
l
e
a
n
d
t
h
e
M
A
T
L
A
B
a
n
g
l
e
S
t
o
r
e
t
h
e
d
a
t
a
f
r
o
m
t
h
e
r
o
t
a
r
y
e
n
c
o
d
e
r
i
n
t
o
a
r
d
u
i
n
o
I
f
e
q
u
a
l
t
h
e
s
t
e
e
r
i
n
g
w
i
l
l
m
a
i
n
t
a
i
n
I
f
n
o
t
e
q
u
a
l
t
h
e
s
t
e
e
r
i
n
g
w
i
l
l
t
u
r
n
y
e
s
n
o
E
N
D
P
e
r
f
o
r
m
i
n
g
t
h
e
P
I
D
C
o
n
t
r
o
l
l
e
r
Figure
1. Flo
w
ch
a
rt of the
sy
stem
2
.
1.
PI
D C
ontrolle
r
Design
In
co
ntr
olli
ng
the
ste
erin
g
usi
ng
t
he
P
WM
sign
al
,
it
pro
duces
a
lot
of
vi
br
at
io
n
a
nd
di
sturbance
on
the
ste
erin
g.
T
his
pro
blem
ca
n
be
so
l
ved
by
i
m
ple
m
enting
the
PID
c
on
t
r
ol
syst
e
m
.
It
is
because
it
can
rej
ect
the
distu
rb
a
nc
es
happe
ne
d
on
t
he
cu
r
vatu
re
w
hic
h
increases
li
near
ly
with
resp
ect
to
ti
m
e
[9
]
.
This
co
nv
e
ntio
nal
PI
D
c
on
t
rol
al
so
can
ov
e
r
com
e
with
hig
h
pr
eci
si
on
when
the
an
gle
error
to
o
la
rg
e
[10].
Figure
2
s
hows
the PI
D
syst
e
m
b
lock
dia
gr
a
m
.
Figure
2. PID
blo
c
k diag
ram
Fo
r
t
he
P
ID
c
on
t
ro
ll
er
desig
n,
s
om
e
par
am
et
ers
nee
d
to
be
ob
ta
ine
d
firs
t
fr
om
the
or
ig
inal
gr
a
ph.
In
this
pa
rt
will
sh
ow
the
cal
culat
ion
m
et
ho
d
us
ed
f
or
deter
m
ining
the
tra
ns
fe
r
f
un
ct
io
n
G(
s
)
of
the
gra
ph
that
was
obta
ine
d
f
ro
m
AV
.
This
trans
fer
f
unct
ion
G(
s
)
was
use
d
to
obta
in
t
he
pa
ram
et
ers
of
the
P
ID
c
on
tr
oller
wh
ic
h
ar
e
K
p,
Ki,
a
nd
K
d.
As
show
n
i
n
(
1
)
,
(
2
)
,
(
3
)
,
an
d
(
4
)
s
hows
t
he
cal
culat
ion
of
pa
r
a
m
et
ers
for
tra
ns
fe
r
functi
on
G(
s
).
As
s
how
n
i
n
(
5
)
is t
he
r
es
ult o
f
the t
ransfe
r
f
un
ct
io
n G
(s). F
igure
3
s
hows
t
he param
et
ers
us
e
t
o
determ
ine the tran
s
fer f
unct
io
n.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vo
l.
13
, N
o.
1
,
Ja
nu
a
ry 20
19
:
2
7
9
–
2
8
5
282
Figure
3. Para
m
et
ers
for hig
he
r order
m
od
el
s
ℎ
(
)
=
Δ
Δ
ℎ
(
)
=
4
.
4
10
=
0
.
44
(1)
=
(
−
)
=
(
−
0
.
44
)
=
1
.
32
(2)
=
(
)
−
=
1
.
0
(
1
.
32
)
−
1
.
32
=
0
.
53
(3)
=
=
1
.
32
=
0
.
25
(4)
G
(
s
)
=
K
(
τ
2
2
+
2τε
s
+
1
)
−
G
(
s
)
=
1
(
0
.
53
)
2
2
+
2
(
0
.
53
)
(
0
.
25
)
s
+
1
)
−
(
1
)
=
1
0
.
29
2
+
0
.
27
+
1
(5)
The
MA
TLAB
so
ft
war
e
was
us
e
d
to
do
t
he
si
m
ulati
on
bas
ed
on
t
he
tra
nsfer
functi
on
G
(
s)
that
was
ob
ta
ine
d
as
s
how
n
i
n
(5).
I
n
F
ig
ur
e
4
sho
ws
th
e
tra
ns
fe
r
f
unct
ion
G
(s
)
w
as
in
serted
into
t
he
M
A
TLAB
ps
e
udo
c
ode
so
ft
war
e
.
Fi
gure
5
s
hows
t
he
gr
a
ph
of
the
tra
nsfer
f
unct
ion
afte
r
t
he
e
xec
ution
of
the
ps
e
udoc
od
e
.
T
he
pa
ram
et
ers
wh
ic
h
are
K
p,
Ki
an
d
Kd
wer
e
obta
ine
d
f
ro
m
the
tra
n
sfe
r
f
unct
io
n
gr
a
ph
in F
ig
ure
5.
>>
s=tf
(
'
s
'
)
;
>>
sys
= 1/(
0.29*s^
2
+
0.27
*s
+ 1)
;
>>
sys
sys =
1
-
------
------
---
-----
0.2
9
s^
2 +
0.27 s
+ 1
Con
ti
nu
ou
s
-
ti
me
tr
an
sfe
r
fu
nc
ti
on
.
>>
ste
p(sys)
Figure
4. MAT
LAB
ps
e
udo
c
od
e
Figure
5. G
raph
of the tra
nsfe
r
f
unct
ion
Evaluation Warning : The document was created with Spire.PDF for Python.
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci
IS
S
N:
25
02
-
4752
Impr
ovin
g
ste
e
ring co
n
ver
gence i
n au
t
onom
ou
s
ve
hicle
ste
eri
ng control
(
Amir As
hraf M
ohama
d
)
283
2
.
2
.
H
ardw
ar
e Implem
en
t
at
i
on
Hardwa
re
par
t
is
the
m
ajo
r
e
lem
ent
in
desi
gn
i
ng
the
a
utono
m
ou
s
ve
hicle
.
Ha
rdwa
re
pa
rts
nee
d
al
l
el
ect
rical
co
m
pone
nt
to
en
sure
that
the
a
uto
nom
ou
s
veh
i
cl
e
functi
on
i
ng
well
.
Ha
rdware
can
be
div
i
de
d
in
t
o
two pa
rts whic
h
a
re m
echan
ic
al
an
d el
ect
rical
p
arts.
2
.
2
.
1.
Mech
ani
cal Pa
r
t
Exam
ple
of
t
he
m
echan
ic
al
par
ts
a
re
body
,
ste
erin
g
it
sel
f,
ti
res
a
nd
m
otor.
S
om
e
of
the
ori
gi
nal
ste
erings
set
ha
d
bee
n
rem
ov
ed
s
uc
h
as
conve
ntion
al
st
eerin
g
an
d
the
ste
ering
s
haf
t
.
This
is
beca
us
e
to
rep
la
ce
the
or
i
gin
al
pa
rt
with
the
el
ect
rical
m
oto
r
wh
ic
h
nam
el
y
Vex
ta
m
oto
r
that
was
us
e
d
to
co
nt
ro
l
the
directi
on
of
th
e
fron
t
w
heel.
Ther
e
f
or
e
,
a
lum
iniu
m
m
a
ter
ia
l
was
us
ed
for
the
m
ounting
bracket
for
Vex
t
a
m
oto
r.
It
is
de
sign
e
d
s
o
th
at
the
Vex
ta
m
oto
r
sh
aft
is
connecte
d
di
rectl
y
to
the
ste
ering
rac
k
sh
aft.
Nex
t,
the
al
um
inu
m
plate
has
been
us
e
d
to
m
ount
the
brack
et
fo
r
the
r
otar
y
enco
de
r.
T
he
Rot
ary
encode
r
was
m
ou
nt and the
center
of the
steerin
g
rac
k
t
o get the a
ngle
of the
steerin
g.
2
.
2
.
2.
El
ectric
al P
art
Ele
ct
rical
par
t
require
d
tw
o
m
ic
ro
co
ntro
ll
e
rs,
ro
ta
ry
e
nc
oder
,
24
V
le
ad
aci
d
batte
ry,
Vex
ta
m
oto
r
and
Vex
ta
m
oto
r
dr
ive
r.
B
oth
m
ic
ro
con
t
ro
ll
ers
us
e
d
we
re
Ardu
i
no
Me
ga
.
Ard
uino
Me
ga
was
co
nn
ect
ed
to
the
ro
ta
ry
enc
od
e
r
usi
ng
the
interr
up
t
pin
on
Ard
uino
Me
ga
to
c
ollec
t
t
he
data
f
r
om
t
he
r
otary
enc
oder
a
nd
perform
the
ca
lc
ulati
on
.
Sec
onda
ry
A
rduin
o
Me
ga
was
us
ed
to
se
nd
the
sign
al
to
the
Vex
ta
m
oto
r
dri
ve
r
(AXHD
450K)
to
con
t
ro
l
the
directi
on
of
th
e
Vex
ta
m
oto
r.
The
Ve
xta
m
oto
r
wa
s
us
e
d
to
m
ov
e
the
dir
ect
ion
of the stee
rin
g.
2
.
2
.
3.
Softw
ar
e Impli
ment
at
ion
This
par
t
discu
sses
on
softwa
re
that
was
us
e
d
in
this
resear
ch
to
pe
rfo
rm
the
a
naly
sis
of
i
m
pr
ov
i
ng
conve
rg
e
nce
of
ste
eri
ng
a
ng
l
e
in
a
uton
om
ou
s
c
ontr
ol
of
UiTM
A
ut
onom
ou
s
Ve
hicle
.
The
Ardu
i
no
I
D
E
so
ft
war
e
w
as
us
e
d
as
a
cod
e
com
piler
wh
ic
h
la
te
r
up
l
oaded
the
co
de
to
the
Ar
duin
o
Me
ga
m
ic
ro
co
ntr
oller.
Ba
sic
al
ly
,
the
C/
C+
+
progra
m
m
ing
la
ngua
ge
was
us
e
d
i
n
this
researc
h.
The
n,
t
he
Mi
cr
os
oft
E
xcel
20
10
wa
s
us
e
d
to
sto
re
the
data
colle
ct
ed
f
ro
m
the
Ar
duin
o
Me
ga
t
o
tran
sf
or
m
into
the
grap
h
a
nd
do
t
he
anal
ysi
s.
The
e
xtensi
on
so
ft
war
e
f
or
M
ic
ro
s
of
t
E
xcel
2010
wh
ic
h
is
PLX
-
D
A
Q
w
a
s
us
e
d
for
c
onn
ect
ing
the
Mi
cro
s
oft
Excel
with
th
e
Ardu
i
no
Me
ga
via
U
SB
Serial
connecti
on.
T
his
si
m
ple
so
ftwa
re
wa
s
able
to
m
on
it
or
or
m
aking
t
he
gr
aph
i
n
real
-
ti
m
e
m
on
it
or
in
g.
It
can
direct
ly
analy
ze
the
ou
t
pu
t
gr
a
ph
ei
ther
the
ou
tpu
t
is
pr
eci
se
or
no
t.
2
.
3
.
Pr
oj
ec
t
Setu
p
This
pro
j
ect
se
tup
is
ab
out
th
e
process
of
al
l
equ
i
pm
ent
or
com
po
ne
nt
w
hi
ch
ha
d
bee
n
i
m
ple
m
ented
on
t
he
pro
j
ect
befor
e
d
oing
the
analy
sis.
T
he
F
igure
6
s
ho
ws
al
l
the
hard
war
e
had
al
rea
dy
bee
n
instal
l
on
t
he
veh
ic
le
.
T
he
m
axi
m
u
m
ang
le
of
t
he
ste
eri
ng
ha
d
be
en
ta
ken
first
before
do
i
ng
t
he
a
na
ly
sis.
The
m
a
xim
u
m
ang
le
wh
e
n
tu
r
ning
to
the
le
ft
from
the
righ
t
is
60
°
w
hich
m
eans
le
ft
is
30
°
a
nd
rig
ht
is
30°.
Fi
gure
7
s
how
s
the il
lustrati
on
of the m
axi
m
um
d
egr
ee
of
t
urnin
g
ste
eri
ng a
ng
le
.
Figure
6. The
s
et
up
of the
proj
ect
Figure
7. A
ngle
of tu
rn
i
ng ste
erin
g
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vo
l.
13
, N
o.
1
,
Ja
nu
a
ry 20
19
:
2
7
9
–
2
8
5
284
3.
RESU
LT
S
A
ND AN
ALYSIS
In
this
sect
io
n
we
com
par
e
th
e
ste
ering
c
onverge
nce
pe
rfo
rm
ance
to
a
sp
eci
fic
ste
ering
ang
le
us
in
g
bo
t
h
im
ple
m
entat
ion
of
P
I
D
c
on
t
ro
ll
er
a
nd
without
P
ID
c
on
t
ro
ll
er.
T
hr
e
e
res
ults
with
t
hr
ee
dif
fer
e
nt
t
arg
et
e
d
ste
ering
a
ng
le
wh
ic
h
a
re
10
°,
15°
an
d
25
°
wer
e
c
ollec
te
d.
Fig
ure
8
,
F
igure
9
,
an
d
F
igu
r
e
10
s
how
the
com
par
ison
in
te
rm
of
ste
erin
g
co
nv
e
rgence
to
ta
rg
et
ed
ste
erin
g
an
gle
with
PI
D
c
on
t
ro
l
le
r
and
with
ou
t
PI
D
con
t
r
oller.
Figure
8,
9
an
d
10
shows
th
at
the
ov
ers
ho
ot
has
bee
n
re
du
ce
d
with
th
e
i
m
ple
m
entat
ion
of
PID
con
t
ro
ll
er.
F
or
exam
ple,
F
igu
re
8
sho
ws
tha
t
the
a
m
pli
tud
e
of
the
overs
hoot
is
14.4°
at
tim
e
2.
2s
is
wi
thout
the
PID
c
on
t
rol
le
r
com
par
ed
with
F
ig
ure
11
shows
t
hat
th
e
am
pli
tud
e
of
the
over
sho
ot
is
10.8°
at
tim
e
2.0
s
with
PID
co
ntr
oller.
The
c
onverge
nce
tim
e
achieve
d
by
PID
co
ntro
l
is
al
so
sig
nificantl
y
faster,
as
sh
own
by
Figure 8. PID s
te
ering
c
ontrol
conve
rg
es t
o
th
e d
esi
re
d
an
gle
o
f 10
°
after
7 sec
ond
s
but, P
WM co
ntr
oller
took
19
sec
onds
to
conve
rg
e
t
o
th
e
sa
m
e
ang
le
.
Fr
om
Figure
9,
the
PID
c
on
t
r
oller
co
nver
ge
nce
to
ok
19
se
conds
faster
tha
n
P
W
M
con
tr
oller
wh
il
e
in
Fig
ur
e
10,
the
PID
con
t
ro
ll
er
c
onverge
nce
took
37
sec
onds
fas
te
r
than
P
W
M c
ontroll
er.
Ther
e
was
som
e
pr
oble
m
wh
il
e
carryin
g
ou
t
this
analy
s
is
wh
ic
h
is
co
m
ing
from
the
m
echan
ic
al
par
ts.
T
her
e
w
as
a
gap
on
the
ste
ering
rac
k
that
m
akes
the
ste
ering
beco
m
e
vib
rates.
T
hi
s
sit
uation
m
ak
es
the
value of
e
nc
oder is less accu
r
at
e. Besi
des,
th
is Vex
ta
m
oto
r
is n
ot s
ui
ta
ble
for
co
ntr
olli
ng
the stee
rin
g
be
caus
e
it
do
es
not
ha
ve
fixe
d
a
ng
le
i
n
perform
ing
the
directi
on.
T
he
Vex
ta
m
otor
dri
ve
rs
hav
e
so
m
e
pr
oble
m
wh
ic
h
is
the
loa
d
to
o
high
for
t
he
V
exta
m
oto
r
t
o
s
upport.
T
he
Ve
xta
m
oto
r
dr
i
ve
r
will
stop
w
ork
i
ng
an
d
nee
d
reset
at
the
Ve
xta
m
otor
dr
ive
r.
Th
ere
are
tw
o
wa
ys
to
reset
t
he
m
oto
r
dr
i
ver
w
hich
is
tur
n
off
the
powe
r
s
upply
t
o
the V
e
xta m
oto
r
drive
r
a
nd gi
ve
the
sig
nal hi
gh
t
o
the
r
es
et
p
in
on t
he Ve
xta m
oto
r dr
i
ve
r.
The
s
peed
of
t
he
Ve
xta
m
oto
r
al
so
o
ne
of
t
he
im
po
rtant
a
sp
ect
that
nee
ds
to
be
co
ns
i
der
e
d
w
hile
perform
ing
thi
s
pro
j
ect
.
The
exp
e
rim
ental
of
s
peed
te
st
ne
ed
to
be
do
ne
seve
ral
ti
m
es
to
get
the
acc
ur
at
e
sp
ee
d
of
the
V
exta
m
oto
r
be
f
or
e
pe
rfo
rm
ing
the
analy
sis.
Othe
r
than
that
,
the
values
of
t
he
gr
a
ph
al
s
o
need
t
o
be
c
ollec
te
d
fe
w
ti
m
es f
or
acc
ur
at
e
res
ults.
Figure
8.
Com
par
is
on of stee
rin
g
c
onverge
nc
e to
10
°
ang
le
with
ou
t
us
in
g
P
I
D
c
on
t
ro
ll
er a
nd
us
in
g
P
ID
con
t
ro
ll
er
Figure
9.
Com
par
is
on of stee
rin
g
c
onverge
nc
e to
15
°
ang
le
with
ou
t
us
i
ng P
I
D
c
on
t
ro
ll
er a
nd
us
in
g
P
ID
con
t
ro
ll
er
Figure
10.
C
om
par
ison
of stee
rin
g
c
onverg
ence to
25
°
an
gl
e w
it
hout
us
i
ng P
ID co
ntr
oller a
nd u
si
ng P
I
D
con
t
ro
ll
er
Evaluation Warning : The document was created with Spire.PDF for Python.
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci
IS
S
N:
25
02
-
4752
Impr
ovin
g
ste
e
ring co
n
ver
gence i
n au
t
onom
ou
s
ve
hicle
ste
eri
ng control
(
Amir As
hraf M
ohama
d
)
285
4.
CONCL
US
I
O
N
In
c
oncl
us
i
on,
this
pa
per
has
pro
po
se
d
a
n
i
m
ple
m
ent
at
ion
of
PID
c
ontr
ol
le
r
on
UiTM
AV0
ste
e
rin
g
con
t
ro
l
.
Re
s
ults
sh
owe
d
that
th
e
ov
e
rs
hoot
ing
of
P
WM
co
ntr
oller
has
be
en
re
du
ce
d
aft
er
i
m
ple
m
ent
ation
of
the
PID
co
ntr
ol
le
r.
The
P
I
D
c
on
t
ro
ll
er
had
s
how
n
sat
isfact
or
y
pe
r
form
ance
in
al
l
assessm
ents
includi
ng
tim
e
respo
ns
e,
sta
bi
li
ty
and
overs
hoot
of
t
he
syst
e
m
.
Fr
om
experim
ents,
PI
D
con
t
ro
ll
er
ca
n
achieve
c
onve
r
gen
ce
to
desi
red
ste
e
rin
g
an
gle
up
t
o
37
sec
onds
f
ast
er
tha
n
P
W
M
con
tr
oller,
du
e
t
o
it
s
bette
r
sta
bili
ty
.
Th
us,
the
ste
ering
c
ontr
ol
f
or
UiTM
AV0
wh
ic
h
po
s
sesses
good
sta
bili
ty
and
fast
co
nv
e
r
gen
ce
ha
s
bee
n
su
ccess
fu
ll
y de
velo
ped.
ACKN
OWLE
DGE
MENTS
This
resea
rc
h
work
was
s
up
ported
by
I
ns
ti
tute
of
Re
sea
r
ch
Ma
na
gem
e
nt
an
d
I
nnovat
ion
(I
RM
I
),
Un
i
ver
sit
i
Teknolo
gi
MAR
A
(U
iTM
),
S
hah
Alam
,
S
el
angor,
Ma
la
ysi
a.
The
aut
hors
al
so
wish
to
ackno
w
le
dge a
nd tha
nks the
F
acult
y of El
ect
rical
Enginee
ring their
lo
gisti
c sup
port.
REFERE
NCE
S
[1]
K.
R.
L
i,
G.
T.
Li
n,
L.
Y.
Lee,
and
J.
C.
Juang
,
“
Applic
at
ion
of
Parti
cle
Fil
te
r
Tr
ac
king
Algor
it
h
m
in
Autonom
o
us
Vehic
l
e
Nav
iga
t
i
on,
”
pp
.
250
–
25
5,
2013
.
[2]
R.
E
.
Fen
ton
and
R.
J.
Ma
y
h
an,
“
Autom
at
ed
High
wa
y
Stud
ie
s
a
t
T
he
Ohio
State
U
nive
rsit
y
—
An
O
ver
vie
w,
”
IE
EE
Tr
ans.
Ve
h.
Tec
hnol
.
,
vol
.
40
,
no
.
1
,
pp
.
100
–
113
,
1991.
[3]
C.
Thorpe,
M.
Hebe
rt,
T.
Kan
a
de,
and
S
.
Shaf
fer
,
“
Towa
rd
A
utonomous
Driv
ing:
Th
e
CMU
Navla
b
Part
I
—
P
erc
eption,
”
IE
E
E
E
xpe
rt
.
Syst
.
t
hei
r A
pp
l
.
,
vol
.
6
,
no
.
4
,
pp
.
31
–
4
2,
1991
.
[4]
E.
Gui
zz
o
,
“
How
Google’
s Sel
f
-
Driving
Car
W
orks,”
on
Li
gn
e.
p.
5
.
p,
2011.
[5]
J.
Huang
and
H.
Ta
n
,
“
Design
of
an
Autom
at
i
c
Stee
ring
Con
troller
for
Bus
Rev
enue
Servi
ce
B
a
sed
on
Drive
rs
’
Stee
ring
Mec
h
an
ism
,
”
pp.
3930
–
3935,
2014
.
[6]
S.
A.
Saruc
hi
,
H
.
Za
m
zur
i,
S.
A.
Maz
la
n
,
M.
Ha
t
ta
,
M.
Ariff,
and
M.
Afandi
,
“
Ac
ti
ve
Front
St
ee
ri
ng
for
Stee
r
-
by
-
W
ire
Vehi
cl
e
vi
a
Com
posite
No
nli
ne
ar
Feedb
ack Cont
rol
,
”
2015
.
[7]
S.
M.
H.
Fahami
,
H.
Za
m
zur
i
,
S.
A.
Maz
l
an,
and
M.
A.
Za
kar
ia,
“
Modeli
ng
and
si
m
ula
ti
on
of
vehicle
stee
r
b
y
wir
e
s
y
stem,”
in
SH
US
ER
2012
-
2
012
IEE
E
Symposium
on
Hum
a
nit
ie
s
,
Sci
en
ce
a
nd
Engi
nee
ring
Re
search
,
2012,
pp.
765
–
770
.
[8]
J.
K.
Subos
it
s
and
J.
C.
Gerd
es,
“
A
Sy
nth
et
i
c
Input
Approa
ch
to
Slip
Angle
Based
Ste
ering
Control
for
Autonom
ous Vehi
cles,” pp. 2297
–
2302,
2017
.
[9]
Kim
Seng
Chia
,
“
Zi
egl
er
-
Nichols
Based
Proportiona
l
-
In
te
gra
l
-
Deri
va
ti
ve
Con
tro
ll
er
for
a
L
ine
Tra
ck
ing
Robot
”
,
Indone
sian
Jour
nal
of
E
le
c
tri
c
al
Engi
ne
eri
ng
and
Com
pute
r
Sci
en
ce
,
Vol.
9,
No.
1
,
Janua
r
y
2018
,
pp.
221
-
226
[10]
Gao
W
ei
,
Zhou
Miaolei,
“
Fuzz
y
PID
Contro
l
Method
for
In
ternet
-
base
d
T
el
e
oper
ation
Mani
pula
tors
S
y
s
te
m
”
,
TE
LKOM
NIK
A
,
Vol
.
11,
No.11,
Novem
ber
2013,
pp.
6381
-
6389
.
[11]
Qi
Xuem
ei
,
Zh
ang
Jingdong,
“
Stud
y
on
Adapt
ive
PID
Contro
l
Algorit
hm
B
a
sed
on
RBF
Neura
l
Network”,
TE
LKOM
NIK
A
Indone
sian
Jour
nal
of
E
le
c
tri
c
al
Engi
ne
eri
ng,
Vo
l.
13
,
No
.
2
,
Feb
rua
r
y
2015
,
pp
.
287
-
291
[12]
R.
Marino
,
S.
Sc
al
z
i,
G.
Orlan
do
,
and
M.
Ne
tt
o
,
“
A
Neste
d
PID
Stee
ring
Con
trol
f
or
La
n
e
Kee
p
ing
in
Visio
n
Base
d
Autonom
ous Vehi
cles,” pp. 2885
–
2890,
2009
.
[13]
C.
S.
Nam
uduri,
S.
Gopala
krish
nan,
B.
V
Murt
y
,
R.
Bol
io,
R
.
Fell
er
,
and
M.
Rd,
“
Robust
Control
of
Low
-
Co
st
Actua
tor
for
Aut
om
oti
ve
Act
ive
Front
Stee
r
ing
A
ppli
c
at
ion
,
”
pp.
2108
–
2114,
200
9.
[14]
C.
Urm
son
et
al
.
,
“
Autonom
ous
drivi
ng
in
Urban
envi
ronm
ent
s:
Boss
and
the
Ur
ban
Chal
l
enge,”
in
Springer
Tra
c
ts
in
Advanc
ed
Ro
boti
cs,
2009,
vol
.
56
,
pp
.
1
–
59
.
Evaluation Warning : The document was created with Spire.PDF for Python.