I
A
E
S
I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
r
t
if
ic
ia
l
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
AI
)
V
ol
.
14
, N
o.
4
,
A
ugus
t
2025
, pp.
2991
~
3002
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
14
.i
4
.pp
2991
-
3002
2991
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
O
p
t
i
m
i
z
i
n
g t
r
af
f
i
c
l
i
gh
t
s at
u
n
b
al
an
c
e
d
i
n
t
e
r
se
c
t
i
on
s u
si
n
g
d
e
e
p
r
e
i
n
f
or
c
e
m
e
n
t
l
e
ar
n
i
n
g
D
u
m
an
C
ar
e
K
h
r
is
n
e
1
,
2
, M
ad
e
S
u
d
ar
m
a
2
, I
d
a A
yu
D
w
i
G
ir
i
an
t
ar
i
2
, D
e
w
a M
ad
e
Wi
h
ar
t
a
2
1
D
oc
t
or
a
l
P
r
ogr
a
m
of
E
ngi
ne
e
r
i
ng S
c
i
e
nc
e
, F
a
c
ul
t
y of
E
ngi
ne
e
r
i
ng, U
da
ya
na
U
ni
ve
r
s
i
t
y, B
a
l
i
, I
ndone
s
i
a
2
D
e
pa
r
t
m
e
nt
of
E
l
e
c
t
r
i
c
a
l
E
ngi
ne
e
r
i
ng,
F
a
c
ul
t
y of
E
ngi
ne
e
r
i
ng, U
da
ya
na
U
ni
ve
r
s
i
t
y, B
a
l
i
, I
ndone
s
i
a
A
r
t
ic
le
I
n
f
o
A
B
S
T
R
A
C
T
A
r
ti
c
le
h
is
to
r
y
:
R
e
c
e
iv
e
d
J
a
n
23
,
2025
R
e
vi
s
e
d
J
un
10
,
2025
A
c
c
e
pt
e
d
J
ul
10
,
2025
Unbalanced
intersectional
traffic
flow
increases
vehicle
delays
,
fuel
consumpt
ion,
and
pollut
ion.
This
study
investi
gates
the
applicati
on
o
f
deep
reinforcement
learni
ng
(DRL)
to
optimize
traffic
signal
timing
at
the
Pamelisa
n
interse
ction
in
Denpa
sar,
Indone
sia.
Real
-
world
traffic
dat
a
were
incorpora
ted
into
a
SUMO
microsimulation
environmen
t
to
train
DRL
agents
using
the
deep
Q
-
network
(DQN)
algorithm.
Experimental
results
show
that
DRL
-
based
optimization
reduced
the
average
vehicle
waitin
g
time
from
594.4
9
seconds
(static
control)
to
169.44
seconds
and
173.10
s
econds
for
agents
trained
without
and
with
noise,
respectively.
The
average
vehicle
speed
remained
stable
at
5.6
–
5.97
m/s
across
all
scenarios,
ind
icating
enhanced
traffic
efficiency
without
adverse
effects.
The
finding
s
und
erscore
the effe
ctiveness a
nd adapta
bility of DRL in
addressin
g traffic
ineffici
encies,
optimizing
them,
and
offering
a
robust
solution
for
dynamic
traffic
manageme
nt at unbala
nced tra
ffic inter
sections in ur
ban are
as.
K
e
y
w
o
r
d
s
:
D
e
e
p r
e
in
f
or
c
e
m
e
nt
l
e
a
r
ni
ng
O
pt
im
iz
e
S
im
ul
a
ti
on
T
r
a
f
f
ic
s
ig
na
l
U
nba
la
nc
e
d t
r
a
f
f
ic
W
a
it
in
g t
im
e
This is an
open
acce
ss artic
le unde
r the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
D
um
a
n C
a
r
e
K
hr
is
ne
D
oc
to
r
a
l
P
r
ogr
a
m
of
E
ng
in
e
e
r
in
g S
c
ie
nc
e
, F
a
c
ul
ty
of
E
ngi
ne
e
r
i
ng, Uda
ya
na
U
ni
ve
r
s
it
y
B
a
li
,
I
ndone
s
ia
E
m
a
il
:
dum
a
n@
unud.a
c
.i
d
1.
I
N
T
R
O
D
U
C
T
I
O
N
T
r
a
ns
por
ta
ti
on
pr
obl
e
m
s
a
r
e
s
ti
ll
of
te
n
c
a
us
e
d
by
tr
a
f
f
ic
c
onge
s
ti
on,
w
hi
c
h
ha
s
a
n
im
pa
c
t
on
tr
a
f
f
ic
a
c
c
id
e
nt
s
,
pol
lu
ti
on,
a
nd
e
c
onomi
c
lo
s
s
e
s
[
1]
,
[
2]
.
P
r
e
vi
ous
r
e
s
e
a
r
c
h
ha
s
s
uc
c
e
e
de
d
in
s
um
m
a
r
iz
in
g
s
e
v
e
r
a
l
te
c
hni
que
s
th
a
t
c
a
n
be
us
e
d
to
s
ol
ve
tr
a
f
f
ic
pr
obl
e
m
s
.
P
r
e
vi
ous
r
e
s
e
a
r
c
h
ha
s
s
uc
c
e
e
de
d
in
s
um
m
a
r
iz
in
g
s
e
ve
r
a
l
te
c
hni
que
s
f
or
s
ol
vi
ng
tr
a
f
f
ic
pr
obl
e
m
s
ba
s
e
d
on
th
e
ir
c
om
pl
e
ti
on
ti
m
e
[
3]
.
T
he
s
e
te
c
hni
que
s
a
r
e
gr
oupe
d
in
to
lo
ng
-
te
r
m
,
m
e
di
um
-
te
r
m
,
a
nd
s
hor
t
-
te
r
m
s
ol
ut
io
ns
.
O
ne
of
th
e
s
hor
t
-
te
r
m
or
r
e
a
l
-
ti
m
e
te
c
hni
que
s
is
c
a
r
r
ie
d
out
th
r
ough
good
m
a
na
ge
m
e
nt
of
tr
a
f
f
ic
f
lo
w
a
t
in
te
r
s
e
c
ti
ons
[
4]
,
[
5]
.
A
tr
a
f
f
ic
li
ght
s
ys
te
m
c
a
n
m
a
na
ge
s
hor
t
-
te
r
m
tr
a
f
f
ic
f
lo
w
a
t
in
te
r
s
e
c
ti
ons
[
5]
,
[
6]
.
A
da
pt
iv
e
s
ig
na
l
c
ont
r
ol
m
e
th
ods
,
s
uc
h
a
s
s
pl
it
,
c
yc
le
a
nd
of
f
s
e
t
opt
im
iz
a
ti
on
te
c
hni
que
(
S
C
O
O
T
)
[
7]
a
n
d
S
ydne
y
c
oor
di
na
te
d
a
da
pt
iv
e
tr
a
f
f
ic
(
S
C
A
T
)
[
8]
a
r
e
w
id
e
ly
us
e
d
in
tr
a
f
f
ic
li
ght
m
a
na
ge
m
e
nt
s
ys
te
m
s
.
T
h
e
y
m
os
tl
y
r
e
ly
on
m
a
nua
ll
y
s
c
h
e
dul
e
d
s
ig
na
l
pha
s
e
s
a
nd
w
or
k
w
e
ll
w
he
n
tr
a
f
f
ic
f
lo
w
is
ne
a
r
ly
e
qua
l
in
a
ll
di
r
e
c
ti
ons
.
T
hi
s
s
c
h
e
dul
e
c
ha
nge
s
dyna
m
ic
a
ll
y
by
lo
oki
ng
onl
y
a
t
tr
a
f
f
ic
vol
um
e
us
in
g
in
duc
ti
on
lo
op
s
e
ns
or
s
.
A
s
a
r
e
s
ul
t,
s
ig
n
a
ls
c
a
nnot
s
e
e
a
nd
r
e
a
c
t
to
c
ha
nge
s
in
tr
a
f
f
ic
pa
tt
e
r
ns
in
r
e
a
l
ti
m
e
,
a
nd
tr
a
ns
por
ta
ti
on
ope
r
a
to
r
s
of
te
n
ha
ve
to
m
a
nua
ll
y
c
ha
nge
s
ig
na
l
pha
s
e
s
to
ke
e
p
up
w
it
h
tr
a
f
f
ic
c
ondi
ti
ons
[
9
]
.
F
u
r
th
e
r
m
or
e
,
it
is
of
te
n
pos
s
ib
le
to
f
in
d
m
or
e
tr
a
f
f
ic
in
on
e
di
r
e
c
ti
on t
ha
n t
he
ot
he
r
(
unba
la
nc
e
d t
r
a
f
f
ic
f
lo
w
)
.
T
he
tr
a
di
ti
o
na
l
s
ys
te
m
l
a
c
k
s
in
t
e
ll
ig
e
nt
m
a
n
a
ge
m
e
nt
,
w
hi
c
h
r
e
s
ul
t
s
in
pe
opl
e
w
a
it
in
g,
r
e
g
a
r
dl
e
s
s
of
th
e
a
bs
e
nc
e
of
v
e
hi
c
le
s
f
r
om
th
e
op
pos
i
te
di
r
e
c
ti
on.
T
hi
s
in
e
vi
ta
bl
e
w
a
it
in
g
ti
m
e
s
om
e
ti
m
e
s
m
a
ke
s
p
e
opl
e
r
e
s
tl
e
s
s
, of
t
e
n e
ndi
ng
in
vi
ol
a
ti
on
of
r
ul
e
s
a
nd a
c
c
i
de
nt
s
[
10]
.
F
u
r
th
e
r
m
or
e
, t
hi
s
le
a
ds
t
o m
or
e
f
ue
l
c
on
s
um
pt
i
on
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
.
14
, N
o.
4
,
A
ugus
t
20
25
:
2991
-
3002
2992
a
nd
po
ll
ut
e
s
t
he
s
ur
r
oun
di
ng
e
nvi
r
o
nm
e
nt
.
T
he
r
e
f
or
e
,
i
nt
e
ll
i
ge
nt
or
dyn
a
m
i
c
tr
a
f
f
ic
li
ght
c
on
tr
ol
n
e
e
d
s
to
be
c
ont
in
u
ous
ly
im
pr
ove
d,
e
s
pe
c
i
a
ll
y
w
h
e
n
lo
oki
n
g
a
t
ot
he
r
f
a
c
t
or
s
,
s
u
c
h
a
s
w
a
it
in
g
ti
m
e
a
t
in
t
e
r
s
e
c
ti
ons
a
nd
t
he
tr
a
f
f
ic
vo
lu
m
e
f
a
c
to
r
,
w
hi
c
h
i
s
c
ur
r
e
nt
ly
w
id
e
ly
us
e
d.
A
r
ti
f
ic
ia
l
in
t
e
ll
ig
e
nc
e
,
w
hi
c
h
h
a
s
de
ve
lo
p
e
d
in
th
is
de
c
a
de
,
pr
ovi
de
s
ho
pe
f
or
th
e
e
m
e
r
g
e
nc
e
of
s
y
s
t
e
m
s
w
it
h
hi
g
h
i
nt
e
ll
ig
e
nc
e
a
nd a
d
a
pt
a
ti
on. R
e
s
e
a
r
c
h
[
11]
–
[
1
5]
us
e
s
r
e
in
f
or
c
e
m
e
nt
l
e
a
r
ni
n
g
(
R
L
)
a
nd
d
e
e
p
r
e
in
f
or
c
e
m
e
nt
le
a
r
n
in
g
(
D
R
L
)
a
p
pr
oa
c
he
s
to
pr
ov
id
e
s
ol
ut
i
ons
t
o
ove
r
c
om
e
t
r
a
f
f
i
c
c
onge
s
ti
o
n.
T
he
R
L
m
od
e
l
di
r
e
c
tl
y t
r
ie
s
t
o
a
da
pt
t
o s
ol
ve
t
he
pr
obl
e
m
s
i
m
po
s
e
d
on t
h
e
m
od
e
l,
in
c
lu
di
ng
tr
a
f
f
ic
c
o
nge
s
ti
on
pr
ob
le
m
s
.
D
e
s
pi
t
e
i
ts
s
uc
c
e
s
s
, R
L
s
t
il
l
h
a
s
s
hor
tc
o
m
in
g
s
. W
h
e
n de
a
li
ng
w
it
h
s
ta
t
e
-
a
c
ti
on
s
p
a
c
e
s
t
ha
t
a
r
e
t
oo
l
a
r
g
e
, R
L
a
lg
or
it
hm
s
of
t
e
n r
e
q
ui
r
e
m
a
nua
l
di
v
is
io
n of
s
pa
c
e
i
nt
o s
m
a
ll
e
r
a
nd
s
e
p
a
r
a
t
e
pa
r
ts
t
o
r
e
pr
e
s
e
nt
di
f
f
e
r
e
nt
s
ta
te
s
(
s
t
a
te
-
a
c
ti
o
n
s
pa
c
e
di
s
c
r
e
ti
z
a
ti
on)
.
D
i
s
c
r
e
ti
z
a
ti
on
of
th
e
a
c
ti
on
-
s
t
a
te
s
p
a
c
e
c
a
u
s
e
s
t
h
e
c
om
pl
e
xi
ty
of
th
e
pr
obl
e
m
s
R
L
c
a
n
s
ol
v
e
t
o
be
l
im
it
e
d a
nd
ti
m
e
-
c
on
s
um
in
g
[
16]
.
D
R
L
c
om
e
s
a
s
a
de
ve
lo
pm
e
nt
of
c
onve
nt
io
na
l
R
L
by
a
ddi
ng
a
de
e
p
ne
ur
a
l
ne
twor
k
(
D
N
N
)
to
R
L
.
D
R
L
ha
s
s
uc
c
e
e
de
d
in
ove
r
c
om
in
g
s
e
ve
r
a
l
w
e
a
kne
s
s
e
s
of
c
on
ve
nt
io
na
l
R
L
.
O
ne
of
th
e
m
a
in
r
e
a
s
on
s
is
th
a
t
D
R
L
us
e
s
D
N
N
,
w
hi
c
h
c
a
n
ove
r
c
om
e
hi
ghe
r
pr
obl
e
m
c
om
pl
e
xi
ty
a
nd
r
e
pr
e
s
e
nt
m
or
e
c
om
pl
e
x
va
lu
e
f
unc
ti
ons
or
pol
ic
ie
s
[
17]
.
T
hus
,
D
R
L
c
a
n
a
ddr
e
s
s
pr
obl
e
m
s
w
it
h
la
r
ge
r
di
m
e
ns
io
ns
a
nd
m
or
e
c
om
pl
e
x
e
nvi
r
onm
e
nt
s
,
w
hi
c
h
a
r
e
di
f
f
ic
ul
t
to
ha
ndl
e
by
c
onve
nt
io
na
l
R
L
.
I
n
a
ddi
ti
on,
D
N
N
in
D
R
L
c
a
n
a
ut
om
a
ti
c
a
ll
y
le
a
r
n
m
or
e
m
e
a
ni
ngf
ul
f
e
a
tu
r
e
r
e
p
r
e
s
e
nt
a
ti
ons
f
r
om
in
put
da
ta
,
a
ll
ow
in
g
a
ge
nt
s
to
r
e
c
ogni
z
e
m
or
e
c
om
pl
e
x
pa
tt
e
r
ns
a
nd
m
a
ke
be
tt
e
r
de
c
is
io
ns
[
18]
–
[
20]
.
D
R
L
,
w
hi
c
h
h
a
s
th
e
a
dva
nt
a
ge
of
ha
ndl
in
g
la
r
ge
-
s
c
a
le
a
nd
hi
gh
-
c
om
pl
e
xi
ty
pr
obl
e
m
s
,
m
a
ke
s
it
a
n
a
tt
r
a
c
ti
ve
c
hoi
c
e
f
o
r
c
o
ve
r
in
g
th
e
w
e
a
kne
s
s
e
s
of
c
onve
nt
io
na
l
R
L
in
r
e
s
e
a
r
c
h t
o buil
d a
m
or
e
a
da
pt
iv
e
t
r
a
f
f
ic
l
ig
ht
c
ont
r
ol
s
ys
te
m
.
A
tr
a
f
f
ic
s
im
ul
a
to
r
is
of
te
n
us
e
d
to
e
va
lu
a
te
tr
a
f
f
ic
c
ont
r
ol
s
tr
a
te
gi
e
s
[
21]
,
e
m
pha
s
iz
in
g
s
us
ta
in
a
bi
li
ty
, s
a
f
e
ty
,
a
nd
tr
a
f
f
ic
e
f
f
ic
ie
nc
y
p
e
r
f
or
m
a
nc
e
in
di
c
a
t
or
s
.
R
e
s
e
a
r
c
he
r
s
ha
ve
us
e
d
two
m
a
in
m
e
th
od
s
to
t
e
s
t
tr
a
f
f
ic
s
im
ul
a
to
r
s
:
m
a
c
r
os
c
opi
c
a
nd mi
c
r
os
c
opi
c
. S
e
ve
r
a
l
s
tu
di
e
s
ha
ve
us
e
d m
a
c
r
os
c
opi
c
s
im
ul
a
ti
on
s
t
o
m
im
ic
r
e
a
l
-
w
or
ld
tr
a
f
f
ic
dyna
m
ic
s
[
22]
,
[
23]
.
H
ow
e
ve
r
,
m
or
e
a
nd
m
or
e
s
tu
di
e
s
a
r
e
tu
r
ni
ng
to
m
ic
r
os
c
opi
c
s
im
ul
a
ti
ons
,
s
uc
h
a
s
S
U
M
O
,
V
I
S
S
I
M
,
a
nd
A
I
M
S
U
N
,
w
hi
c
h
of
f
e
r
a
m
or
e
c
om
pr
e
he
ns
iv
e
de
pi
c
ti
on
of
c
om
pl
e
x t
r
a
f
f
ic
dyna
m
ic
s
, i
nc
lu
di
ng t
he
s
to
c
ha
s
ti
c
c
ha
r
a
c
te
r
of
dr
iv
in
g a
nd r
out
e
c
hoi
c
e
s
[
21]
. S
U
M
O
, a
s
one
of
th
e
m
ic
r
os
c
opi
c
s
im
ul
a
to
r
s
,
is
w
id
e
ly
us
e
d
to
e
va
lu
a
te
tr
a
f
f
ic
c
ont
r
ol
s
tr
a
te
gi
e
s
.
H
ow
e
ve
r
,
to
th
e
a
ut
hor
s
'
knowle
dge
, no S
U
M
O
s
im
ul
a
ti
on ha
s
be
e
n buil
t
us
in
g
a
c
tu
a
l
tr
a
f
f
ic
f
lo
w
da
ta
(
a
nd r
e
a
l
-
w
o
r
ld
r
oa
d
ne
twor
ks
)
to
de
m
ons
tr
a
te
t
he
unba
la
nc
e
d t
r
a
f
f
ic
f
lo
w
s
ta
te
.
F
ur
th
e
r
m
or
e
,
a
c
c
or
di
ng
to
T
a
n
e
t
al
.
[
18
]
,
m
os
t
D
R
L
w
o
r
k
is
s
ti
ll
not
r
e
a
dy
f
or
d
ir
e
c
t
a
ppl
ic
a
ti
on
in
r
e
a
l
-
w
or
ld
tr
a
f
f
ic
be
c
a
us
e
,
unt
il
now
,
th
e
D
R
L
a
ge
nt
is
a
s
s
u
m
e
d
to
ha
ve
p
e
r
f
e
c
t
knowle
dge
of
th
e
tr
a
f
f
ic
e
nvi
r
onm
e
nt
.
I
n
r
e
a
li
ty
,
a
c
onge
s
ti
on
d
e
te
c
ti
on
or
pr
e
di
c
ti
on
s
ys
te
m
i
s
hi
ghl
y
de
s
ir
e
d
to
e
s
ti
m
a
te
tr
a
f
f
ic
c
ondi
ti
ons
w
it
h
s
ig
ni
f
ic
a
nt
di
s
tu
r
ba
nc
e
s
,
one
of
w
hi
c
h
is
unba
l
a
nc
e
d
tr
a
f
f
ic
f
lo
w
.
T
he
r
e
f
or
e
,
in
th
is
s
tu
dy,
a
n
a
da
pt
iv
e
tr
a
f
f
ic
c
ont
r
ol
s
y
s
te
m
w
a
s
bui
lt
u
s
in
g
de
e
p
Q
-
ne
twor
k
(
D
Q
N
)
[
24]
.
T
hi
s
D
R
L
a
lg
or
it
hm
is
us
e
d
to
opt
im
iz
e
ve
hi
c
le
w
a
it
in
g
ti
m
e
a
t
s
ig
na
li
z
e
d
in
te
r
s
e
c
ti
ons
by
opt
im
iz
in
g
c
ha
nge
s
in
tr
a
f
f
ic
l
ig
ht
ti
m
e
s
.
T
he
D
Q
N
in
th
is
s
tu
dy
w
a
s
tr
a
in
e
d
us
in
g
S
U
M
O
m
ic
r
os
c
opi
c
s
im
u
la
ti
on
da
ta
w
it
h
c
ha
r
a
c
te
r
is
ti
c
s
of
unba
la
nc
e
d
tr
a
f
f
ic
f
lo
w
a
nd
pe
r
tu
r
ba
ti
on
o
f
que
ue
l
e
ngt
h.
2.
M
E
T
H
O
D
T
hi
s
r
e
s
e
a
r
c
h
is
bui
lt
w
it
h
f
our
m
a
in
s
te
p
s
.
F
ir
s
t,
th
e
num
be
r
of
ve
hi
c
le
s
pa
s
s
in
g
th
r
ough
one
la
n
e
a
t
a
n
in
te
r
s
e
c
ti
on
w
a
s
c
a
lc
ul
a
te
d
us
in
g
Y
O
L
O
v8.
N
e
xt
,
a
n
in
te
r
s
e
c
ti
on
s
im
ul
a
ti
on
w
il
l
be
bui
lt
us
in
g
th
e
pr
e
vi
ous
c
a
lc
ul
a
ti
on
da
ta
us
in
g
S
U
M
O
m
ic
r
os
im
ul
a
ti
on,
w
hi
c
h
w
il
l
be
c
ont
in
ue
d
by
tr
a
in
in
g
D
R
L
a
ge
nt
s
us
in
g
S
U
M
O
s
im
ul
a
ti
on
a
s
in
put
.
F
in
a
ll
y,
th
e
opt
im
iz
a
ti
on
th
a
t
th
e
D
R
L
a
ge
nt
di
d w
il
l
be
a
na
ly
z
e
d.
F
ig
ur
e
1
s
how
s
us
t
h
e
r
e
s
e
a
r
c
h f
lo
w
di
a
gr
a
m
.
CCT
V
F
o
o
t
a
g
e
o
f
Pa
m
e
l
i
s
a
n
I
n
t
e
r
s
e
c
t
i
o
n
,
D
e
n
p
a
s
a
r
Co
u
n
t
i
n
g
Ve
h
i
c
l
e
s
i
n
e
a
c
h
l
a
n
e
u
s
i
n
g
Y
O
L
O
v
8
B
u
i
l
d
i
n
g
a
S
U
M
O
s
i
m
u
l
a
t
i
o
n
f
o
r
t
h
e
p
a
m
e
l
i
s
a
n
i
n
t
e
rs
e
c
t
i
o
n
L
e
v
e
ra
g
i
n
g
t
h
e
S
U
M
O
-
R
L
a
n
d
SB
3
l
i
b
ra
ri
e
s
t
o
t
ra
i
n
R
L
a
ge
n
t
s
w
i
t
h
t
h
e
D
Q
N
(
D
e
e
p
Q
-
N
e
t
w
o
r
k
)
a
l
go
ri
t
h
m
T
ra
i
n
e
d
R
L
A
g
e
n
t
s
f
o
r
I
n
t
e
rs
e
c
t
i
o
n
O
p
t
i
m
i
z
a
t
i
o
n
(
SP
.
P
A
M
E
L
I
S
A
N
)
E
va
l
u
a
t
i
o
n
F
ig
ur
e
1. R
e
s
e
a
r
c
h s
t
a
ge
s
f
lo
w
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
O
pt
imi
z
in
g t
r
af
fi
c
l
ig
ht
s
at
unbalanc
e
d i
nt
e
r
s
e
c
ti
ons
u
s
in
g de
e
p
r
e
in
fo
r
c
e
m
e
nt
…
(
D
um
an C
a
r
e
K
h
r
is
ne
)
2993
2
.1.
C
al
c
u
la
t
in
g ve
h
ic
le
f
lo
w
T
he
da
ta
us
e
d
to
bui
ld
th
e
in
te
r
s
e
c
ti
on
s
im
ul
a
ti
on
w
it
h
unba
la
nc
e
d
tr
a
f
f
ic
f
lo
w
w
a
s
ta
ke
n
f
r
om
th
e
D
in
a
s
P
e
r
hubunga
n
P
r
ovi
ns
i
B
a
li
or
B
a
li
P
r
ovi
nc
ia
l
T
r
a
n
s
p
or
ta
ti
on
A
ge
nc
y
,
w
hi
c
h
ut
il
iz
e
d
s
ur
ve
il
la
nc
e
c
a
m
e
r
a
s
in
s
t
a
ll
e
d
a
t
th
e
P
a
m
e
li
s
a
n
in
te
r
s
e
c
ti
on.
T
he
da
ta
obt
a
in
e
d
is
a
vi
de
o
r
e
c
or
di
ng
of
th
e
s
ur
ve
il
la
nc
e
c
a
m
e
r
a
.
T
he
vi
de
o
is
us
e
d
a
s
in
put
to
c
a
lc
ul
a
te
ve
hi
c
le
s
pa
s
s
in
g
th
r
ough
th
e
la
ne
s
a
t
th
e
P
a
m
e
li
s
a
n
in
te
r
s
e
c
ti
on.
T
he
c
onf
ig
ur
a
ti
on
of
th
e
pos
it
io
n
of
th
e
v
e
hi
c
le
f
l
ow
c
ount
e
r
-
poi
nt
on
th
e
la
ne
a
t
th
e
P
a
m
e
li
s
a
n
in
te
r
s
e
c
ti
on
is
s
how
n
in
F
ig
ur
e
2.
F
ig
ur
e
2
s
how
s
th
a
t
s
ix
c
oun
te
r
-
poi
nt
s
w
e
r
e
us
e
d
to
c
a
lc
ul
a
t
e
ve
hi
c
le
f
lo
w
in
e
a
c
h l
a
ne
a
t
th
e
i
nt
e
r
s
e
c
ti
on. T
a
bl
e
1 pr
e
s
e
nt
s
a
m
or
e
c
onc
is
e
r
e
la
ti
ons
hi
p be
twe
e
n t
he
pos
it
io
n of
t
he
po
in
t
a
nd
th
e
di
r
e
c
ti
on
of
ve
hi
c
le
f
lo
w
c
a
lc
ul
a
te
d
a
t
e
a
c
h
c
ount
e
r
-
poi
nt
.
T
o
c
ount
ve
hi
c
le
s
,
w
e
u
s
e
th
e
Y
O
L
O
v8
obj
e
c
t
de
te
c
ti
on
a
lg
or
it
hm
[
25]
.
Y
O
L
O
v8
is
u
s
e
d
to
de
te
c
t
ve
hi
c
le
s
p
a
s
s
in
g
th
r
ough
a
l
a
ne
.
A
f
te
r
th
a
t,
w
e
us
e
a
n
obj
e
c
t
tr
a
c
ki
ng
a
nd
c
ount
in
g
a
lg
or
it
hm
m
a
de
e
xpl
ic
it
ly
f
or
tr
a
c
ki
ng
obj
e
c
ts
f
r
om
th
e
r
e
s
ul
ts
of
Y
O
L
O
v8
de
te
c
ti
on t
o c
ount
t
he
t
r
a
f
f
ic
f
lo
w
. T
he
f
lo
w
c
ha
r
t
of
t
he
t
r
a
c
ki
ng a
nd c
ount
in
g a
lg
or
it
hm
i
s
s
how
n i
n F
ig
ur
e
3.
F
ig
ur
e
2. C
onf
ig
ur
a
ti
on of
t
he
ve
hi
c
le
f
lo
w
c
ount
e
r
poi
nt
T
a
bl
e
1. S
um
m
a
r
y of
ve
hi
c
le
f
lo
w
di
r
e
c
ti
on a
t
e
a
c
h c
ount
in
g po
in
t
L
a
be
l
C
ount
i
ng
poi
nt
V
e
hi
c
l
e
f
l
ow
di
r
e
c
t
i
on
●
N
gur
a
h R
a
i
1
B
ypa
s
s
N
gur
a
h R
a
i
→ P
a
m
e
l
i
s
a
n (
E
a
s
t
-
N
or
t
h)
;
B
ypa
s
N
gur
a
h R
a
i
→ B
ypa
s
s
N
gur
a
h R
a
i
(
E
a
s
t
-
W
e
s
t
)
●
N
gur
a
h R
a
i
2
B
ypa
s
s
N
gur
a
h R
a
i
→ B
ypa
s
s
N
gur
a
h
R
a
i
(
E
a
s
t
-
W
e
s
t
)
;
B
ypa
s
N
gu
r
a
h R
a
i
→
P
a
m
e
l
i
s
a
n
(
E
a
s
t
-
S
ou
t
h
)
&
B
y
pa
s
s
N
gu
r
a
h R
a
i
→ B
ypa
s
s
N
gu
r
a
h
R
a
i
(
E
a
s
t
-
E
a
s
t
/
U
-
t
ur
n
)
●
N
gur
a
h R
a
i
3
B
ypa
s
s
N
gur
a
h R
a
i
→ B
ypa
s
s
N
gur
a
h
R
a
i
(
W
e
s
t
-
E
a
s
t
)
;
B
yp
a
s
N
gu
r
a
h
R
a
i
→
P
a
m
e
l
i
s
a
n
(
W
e
s
t
-
N
o
r
t
h
)
&
B
y
pa
s
s
N
g
u
r
a
h R
a
i
→
B
y
pa
s
s
N
g
u
r
a
h
R
a
i
(
W
e
s
t
-
W
e
s
t
/
U
-
t
u
r
n
)
●
N
gur
a
h R
a
i
4
B
ypa
s
s
N
gur
a
h R
a
i
→ P
a
m
e
l
i
s
a
n (
W
e
s
t
-
S
out
h)
&
B
ypa
s
N
gur
a
h R
a
i
→ B
yp
a
s
s
N
gur
a
h R
a
i
(
W
e
s
t
-
E
a
s
t
)
●
P
a
m
e
l
i
s
a
n 1
P
a
m
e
l
i
s
a
n → B
a
ypa
s
s
N
gur
a
h
R
a
i
(
S
out
h
-
E
a
s
t
)
, P
a
m
e
l
i
s
a
n → B
a
ypa
s
s
N
gur
a
h
R
a
i
(
S
out
h
-
W
e
s
t
)
;
P
a
m
e
l
i
s
a
n →
P
a
m
e
l
i
s
a
n (
S
out
h
-
N
or
t
h)
●
P
a
m
e
l
i
s
a
n 2
P
a
m
e
l
i
s
a
n → B
a
ypa
s
s
N
gur
a
h
R
a
i
(
N
or
t
h
-
E
a
s
t
)
, P
a
m
e
l
i
s
a
n → B
a
ypa
s
s
N
gur
a
h R
a
i
(
N
or
t
h
-
W
e
s
t
)
;
P
a
m
e
l
i
s
a
n → P
a
m
e
l
i
s
a
n (
N
or
t
-
S
out
h)
2
.2.
B
u
il
d
in
g a S
U
M
O
s
im
u
la
t
io
n
A
f
te
r
obt
a
in
in
g
th
e
ve
hi
c
le
f
lo
w
da
ta
th
a
t
pa
s
s
e
s
th
r
ough
e
a
c
h
r
oa
d
la
ne
a
t
th
e
P
a
m
e
li
s
a
n
in
te
r
s
e
c
ti
on,
th
e
ne
xt
s
te
p
is
tr
a
ns
la
ti
ng
th
e
v
e
hi
c
le
f
lo
w
in
to
th
e
S
U
M
O
m
ic
r
os
im
ul
a
ti
on
[
26]
.
I
n
S
U
M
O
,
th
e
ve
hi
c
le
f
lo
w
is
c
onve
r
te
d
in
to
a
ve
hi
c
le
e
m
e
r
ge
nc
e
s
im
ul
a
ti
on
us
in
g
th
e
r
out
e
s
f
unc
ti
on
.
S
U
M
O
is
a
w
e
ll
-
known
ope
n
-
s
our
c
e
tr
a
f
f
ic
s
im
ul
a
to
r
th
a
t
p
r
ovi
de
s
pr
a
c
ti
c
a
l
gr
a
phi
c
a
l
us
e
r
in
te
r
f
a
c
e
s
(
G
U
I
s
)
a
nd
a
ppl
ic
a
ti
on
pr
ogr
a
m
m
in
g
in
te
r
f
a
c
e
s
(
A
P
I
s
)
f
or
e
f
f
ic
ie
nt
ly
m
a
na
gi
ng
a
nd
m
ode
li
ng
r
oa
d
ne
twor
ks
.
I
t
of
f
e
r
s
a
vi
s
ua
l
gr
a
phi
c
a
l
in
te
r
f
a
c
e
f
or
c
r
e
a
ti
ng dif
f
e
r
e
nt
r
oa
d ne
twor
k a
r
c
hi
te
c
t
ur
e
s
i
n m
a
ny gr
id
f
or
m
a
ts
a
nd a
ll
ow
s
dyna
m
ic
r
out
in
g
[
16]
.
A
ddi
ti
ona
ll
y,
S
U
M
O
s
uppor
ts
O
pe
nS
tr
e
e
tM
a
p
(
O
S
M
)
.
A
f
ul
l
s
c
e
na
r
io
m
a
y
be
c
r
e
a
te
d
qui
c
kl
y
a
nd
e
a
s
il
y
w
it
h
th
e
he
lp
of
th
e
O
S
M
s
c
r
ip
t.
T
yp
e
m
a
ps
a
nd
s
e
t
ti
ngs
a
ppr
opr
ia
te
f
or
th
e
c
hos
e
n
tr
a
f
f
ic
m
ode
s
w
il
l
be
im
por
te
d
in
to
th
e
ne
twor
k.
F
ur
th
e
r
m
or
e
,
S
U
M
O
c
a
n
c
ont
r
ol
e
a
c
h
in
te
r
s
e
c
ti
on'
s
tr
a
f
f
ic
li
ght
s
us
in
g
us
e
r
-
de
f
in
e
d
pol
ic
ie
s
.
S
U
M
O
m
a
ke
s
it
pos
s
ib
le
to
ta
ke
pi
c
tu
r
e
s
a
t
e
ve
r
y
s
im
ul
a
ti
on
s
ta
ge
,
gi
vi
ng
us
th
e
s
ta
te
da
ta
f
or
our
s
tu
dy.
S
U
M
O
s
im
ul
a
ti
on
f
or
th
e
P
a
m
e
li
s
a
n
in
te
r
s
e
c
ti
on
w
a
s
bui
lt
w
it
h
d
a
ta
c
a
lc
ul
a
te
d
f
r
om
th
e
a
c
tu
a
l
tr
a
f
f
ic
f
lo
w
obt
a
in
e
d
in
th
e
pr
e
vi
ous
s
te
p.
U
s
in
g
th
is
da
ta
,
w
e
bui
ld
a
s
im
ul
a
t
e
d
e
nvi
r
onm
e
nt
th
a
t
im
it
a
te
s
r
e
a
l
-
w
or
ld
tr
a
f
f
ic
f
lo
w
a
t
th
e
P
a
m
e
li
s
a
n
in
te
r
s
e
c
ti
on,
w
hi
c
h
ha
s
a
n
unba
la
nc
e
d
tr
a
f
f
ic
f
lo
w
a
nd
s
ta
ti
c
tr
a
f
f
ic
l
ig
ht
pha
s
e
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
.
14
, N
o.
4
,
A
ugus
t
20
25
:
2991
-
3002
2994
S
t
a
r
t
I
n
it
ia
liz
a
t
io
n
O
b
j
e
c
t
(
Y
O
L
O
v
8
)
C
e
n
t
e
r
P
o
in
t
=
{}
I
D
C
o
u
n
t
=
0
o
b
j
e
c
t
B
B
I
D
=
[]
A
ll
O
b
j
e
c
t
s
(
Y
O
L
O
v
8
)
c
h
e
c
k
e
d
?
C
a
l
c
u
la
t
e
O
b
j
e
c
t
C
e
n
t
e
r
N
e
w
o
b
j
e
c
t
d
e
t
e
c
t
e
d
?
A
p
p
e
n
d
t
o
C
e
n
t
e
r
P
o
in
t
C
r
e
a
t
e
a
N
e
w
I
D
a
d
d
it
t
o
o
b
j
e
c
t
B
B
I
D
No
T
h
e
la
s
t
o
b
j
e
c
t
in
C
e
n
t
e
r
P
o
in
t
?
Y
e
s
C
a
l
c
u
la
t
e
Dis
t
a
n
c
e
fr
o
m
T
r
a
c
k
e
d
O
b
j
e
c
t
P
o
s
it
io
n
t
o
p
o
in
t
in
C
e
n
t
e
r
P
o
in
t
Dis
t
a
n
c
e
<
35
?
No
U
p
d
a
t
e
T
r
a
c
k
e
d
O
b
j
e
c
t
p
o
s
it
io
n
t
o
C
e
n
t
e
r
P
o
in
t
A
d
d
o
b
j
e
c
t
p
o
s
it
i
o
n
(
b
y
ID
)
t
o
o
b
j
e
c
t
B
B
I
D
R
e
t
u
r
n
V
a
lu
e
o
b
j
e
c
t
B
B
I
D
Y
e
s
St
o
p
Y
e
s
No
F
ig
ur
e
3. F
lo
w
c
ha
r
t
of
t
r
a
c
ki
ng obje
c
t
a
lg
or
it
hm
2
.
3
.
D
e
e
p
r
e
in
f
or
c
e
m
e
n
t
l
e
ar
n
in
g age
n
t
t
r
ai
n
in
g
I
n
th
is
s
tu
dy,
th
e
D
R
L
a
ge
nt
w
a
s
bui
lt
us
in
g
th
e
S
U
M
O
-
R
L
li
br
a
r
y
[
27
]
.
S
U
M
O
-
R
L
pr
ovi
de
s
a
s
im
pl
e
in
te
r
f
a
c
e
to
c
r
e
a
te
a
R
L
e
nvi
r
onm
e
nt
w
it
h
S
U
M
O
f
or
tr
a
f
f
ic
s
ig
na
l
c
ont
r
ol
.
T
he
D
R
L
a
ge
nt
bui
lt
is
a
n
a
ge
nt
th
a
t
us
e
s
th
e
D
Q
N
a
lg
or
it
hm
f
or
th
e
t
r
a
in
in
g
pr
oc
e
s
s
to
opt
im
iz
e
ve
hi
c
le
w
a
it
in
g
ti
m
e
a
t
th
e
P
a
m
e
li
s
a
n
in
te
r
s
e
c
ti
on. DQ
N
i
n S
U
M
O
-
R
L
w
a
s
bui
lt
us
in
g t
h
e
s
ta
bl
e
b
a
s
e
li
ne
s
3 (
S
B
3)
l
ib
r
a
r
y
[
28]
.
D
Q
N
gi
ve
n
in
put
a
s
a
s
im
ul
a
ti
on
ge
ne
r
a
te
d
in
th
e
p
r
e
vi
ous
s
ta
ge
a
s
a
n
e
nvi
r
onm
e
nt
(
S
U
M
O
e
nvi
r
onm
e
nt
)
.
B
e
c
a
us
e
th
is
s
tu
dy
onl
y
opt
im
iz
e
s
one
t
r
a
f
f
ic
li
ght
(
P
a
m
e
li
s
a
n
in
te
r
s
e
c
ti
on)
,
th
e
D
R
L
a
ge
nt
us
e
d
is
one
(
s
in
gl
e
a
ge
nt
)
.
T
he
a
ge
nt
pe
r
f
or
m
s
opt
im
iz
a
ti
on
us
in
g
th
e
M
a
r
kov
de
c
is
io
n
pr
oc
e
s
s
(
M
D
P
)
m
ode
l
w
it
h t
hr
e
e
c
om
pone
nt
s
:
obs
e
r
va
ti
on, a
c
ti
on, a
nd r
e
w
a
r
d
.
F
or
o
bs
e
r
va
ti
on
s
pa
c
e
,
D
Q
N
us
e
s
D
N
N
,
w
hi
c
h
ha
s
15
in
put
s
ge
ne
r
a
te
d
f
r
om
obs
e
r
va
ti
ons
in
th
e
e
nvi
r
onm
e
nt
,
na
m
e
ly
two
gr
e
e
n
pha
s
e
s
(
nor
th
-
s
out
h
a
nd
e
a
s
t
-
w
e
s
t
gr
e
e
n
li
ght
s
)
,
one
tr
a
ns
it
io
n
pha
s
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
O
pt
imi
z
in
g t
r
af
fi
c
l
ig
ht
s
at
unbalanc
e
d i
nt
e
r
s
e
c
ti
ons
u
s
in
g de
e
p
r
e
in
fo
r
c
e
m
e
nt
…
(
D
um
an C
a
r
e
K
h
r
is
ne
)
2995
(
ye
ll
ow
l
ig
ht
)
, a
nd de
ns
it
y a
nd que
ue
va
lu
e
s
on s
ix
l
a
ne
s
of
t
he
r
oa
d a
t
th
e
P
a
m
e
li
s
a
n i
nt
e
r
s
e
c
ti
on (
12 i
nput
s
)
.
I
n
a
ddi
ti
on
to
s
ta
nda
r
d
obs
e
r
va
ti
ons
in
tr
a
in
in
g,
obs
e
r
va
ti
ons
w
it
h
noi
s
e
w
e
r
e
m
a
de
,
w
hi
c
h
w
e
r
e
c
a
r
r
ie
d
out
by
c
ha
ngi
ng
th
e
que
ue
le
ngt
h
va
lu
e
.
T
hi
s
is
a
w
a
y
to
r
e
pr
e
s
e
nt
r
e
a
l
-
w
or
ld
c
ondi
ti
ons
w
he
r
e
la
r
ge
ve
hi
c
le
s
s
om
e
ti
m
e
s
bl
oc
k
s
ur
ve
il
la
nc
e
c
a
m
e
r
a
s
,
w
hi
c
h
c
a
u
s
e
s
c
ha
ng
e
s
in
th
e
que
ue
l
e
ngt
h
va
lu
e
a
t
in
te
r
s
e
c
ti
ons
.
A
c
ti
on
s
p
a
c
e
g
e
ne
r
a
te
d
a
s
out
put
f
r
om
D
N
N
. T
he
r
e
a
r
e
2
a
c
ti
on
s
p
a
c
e
s
i
n
th
i
s
s
tu
dy
,
a
s
il
l
us
tr
a
te
d
in
F
ig
ur
e
4
.
O
ne
f
or
N
gur
a
h
R
a
i
r
oa
d
(
E
a
s
t
-
W
e
s
t)
,
de
pi
c
te
d
in
F
ig
ur
e
4(
a
)
, a
nd
one
f
or
th
e
gr
e
e
n
pha
s
e
a
t
P
a
m
e
li
s
a
n
r
oa
d
(
N
or
th
-
S
out
h
)
,
de
pi
c
te
d
in
F
ig
ur
e
4(
b
)
.
T
he
r
e
w
a
r
d
f
unc
ti
o
n
is
c
a
lc
ul
a
te
d
us
in
g
c
ha
nge
s
in
c
um
ul
a
ti
ve
ve
hi
c
le
de
la
y.
I
n
ot
he
r
w
or
ds
,
th
e
r
e
w
a
r
d
is
how
m
uc
h
th
e
to
ta
l
de
la
y
(
th
e
s
um
of
th
e
w
a
it
in
g
ti
m
e
s
of
a
ll
a
ppr
oa
c
hi
ng ve
hi
c
le
s
)
c
ha
ng
e
s
c
on
c
e
r
ni
ng t
he
pr
e
vi
ous
t
im
e
s
te
p.
(
a
)
(
b)
F
ig
ur
e
4. P
a
m
e
li
s
a
n i
nt
e
r
s
e
c
ti
on i
n S
U
M
O
w
it
h (
a
)
N
gur
a
h R
a
i
gr
e
e
n pha
s
e
a
nd (
b)
P
a
m
e
li
s
a
n gr
e
e
n
ph
a
s
e
2
.4. E
val
u
at
io
n
T
he
f
in
a
l
s
ta
ge
of
th
is
r
e
s
e
a
r
c
h
is
th
e
e
va
lu
a
ti
on
pr
oc
e
s
s
c
a
r
r
ie
d
out
on
th
e
D
R
L
m
ode
l,
w
hi
c
h
w
a
s
tr
a
in
e
d
in
th
e
pr
e
vi
ous
s
ta
ge
.
T
w
o
e
va
lu
a
ti
on
m
e
tr
ic
s
w
e
r
e
us
e
d
to
m
e
a
s
ur
e
th
e
a
ge
nt
'
s
opt
im
iz
a
ti
on
a
bi
li
ty
:
th
e
a
c
c
um
ul
a
te
d
w
a
it
in
g
ti
m
e
a
t
th
e
in
te
r
s
e
c
ti
on
(
f
or
a
ll
r
oa
d
la
ne
s
)
a
nd
th
e
a
ve
r
a
ge
v
e
hi
c
le
s
p
e
e
d
a
t
th
e
in
te
r
s
e
c
ti
on.
V
e
hi
c
le
w
a
it
in
g
ti
m
e
is
de
f
in
e
d
a
s
th
e
ti
m
e
(
in
s
e
c
onds
)
s
pe
nt
be
lo
w
0.1
m
/s
s
in
c
e
th
e
la
s
t
ti
m
e
th
e
ve
hi
c
le
tr
a
ve
l
e
d
f
a
s
t
e
r
th
a
n
0.1
m
/
s
.
(
T
h
e
ve
hi
c
le
w
a
it
in
g
ti
m
e
is
r
e
s
e
t
to
0
e
ve
r
y
ti
m
e
th
e
ve
hi
c
le
m
ove
s
)
.
I
n (
1)
c
a
lc
ul
a
te
s
t
he
t
ot
a
l
ve
hi
c
le
w
a
it
in
g t
im
e
a
t
a
n i
nt
e
r
s
e
c
ti
on
.
_
=
∑
∑
=
1
=
1
(
1)
W
h
e
r
e
to
ta
l
_W
i
s
t
h
e
t
ot
a
l
w
a
it
i
ng
ti
m
e
in
a
ll
l
a
n
e
s
of
t
h
e
r
oa
d i
n
th
e
i
n
te
r
s
e
c
ti
o
n,
L
i
s
th
e
n
um
b
e
r
of
la
ne
s
of
t
he
r
oa
d i
n t
he
i
n
te
r
s
e
c
ti
on,
V
i
is
t
he
nu
m
b
e
r
of
v
e
h
ic
le
s
i
n
l
a
n
e
i
a
nd
W
ij
i
s
th
e
w
a
it
i
ng
ti
m
e
of
ve
hi
c
le
j
in
l
a
n
e
i
.
T
he
a
ve
r
a
ge
ve
hi
c
le
s
pe
e
d
a
t
a
n
in
te
r
s
e
c
ti
on
is
c
a
lc
ul
a
te
d
by
f
in
di
ng
th
e
a
ve
r
a
ge
s
pe
e
d
of
th
e
ve
hi
c
le
s
a
t
th
e
in
te
r
s
e
c
ti
on,
nor
m
a
li
z
e
d
by
th
e
m
a
xi
m
um
s
pe
e
d
a
ll
ow
e
d
f
or
e
a
c
h
ve
hi
c
le
.
I
f
th
e
r
e
a
r
e
no
ve
hi
c
le
s
a
t
th
e
in
te
r
s
e
c
ti
on,
th
is
f
unc
ti
on
r
e
tu
r
ns
1.
In
(
2
)
c
a
lc
ul
a
te
s
th
e
nor
m
a
li
z
e
d
a
ve
r
a
ge
ve
hi
c
l
e
s
pe
e
d
a
t
a
n i
nt
e
r
s
e
c
ti
on.
_
=
1
∑
,
=
1
(
2)
W
he
r
e
av
g_s
pe
e
d
is
th
e
nor
m
a
li
z
e
d
a
ve
r
a
ge
s
pe
e
d
f
or
a
ll
ve
hi
c
le
s
a
t
th
e
in
te
r
s
e
c
ti
on,
N
is
th
e
num
be
r
o
f
ve
hi
c
le
s
a
t
th
e
in
te
r
s
e
c
ti
on,
S
i
is
th
e
s
pe
e
d
of
ve
hi
c
le
i
a
t
th
e
ti
m
e
of
obs
e
r
va
ti
on
,
a
nd
S
m
ax
,
i
is
th
e
m
a
xi
m
um
s
pe
e
d pe
r
m
it
te
d f
or
ve
hi
c
le
i
(
in
t
he
s
im
ul
a
ti
on e
a
c
h t
ype
of
ve
h
ic
le
i
s
s
e
t
to
a
m
a
xi
m
um
pe
r
m
it
te
d s
pe
e
d)
.
T
o
e
va
lu
a
te
th
e
a
ge
nt
'
s
opt
im
iz
a
ti
on
c
a
pa
bi
li
ty
,
e
a
c
h
a
ve
r
a
ge
w
a
it
in
g
ti
m
e
a
nd
a
ve
r
a
ge
s
pe
e
d
pr
oduc
e
d a
f
te
r
t
r
a
in
in
g
r
e
s
ul
ts
a
r
e
c
om
pa
r
e
d w
it
h t
he
i
ni
ti
a
l
s
im
ul
a
ti
on da
ta
'
s
a
ve
r
a
ge
w
a
it
in
g t
im
e
a
nd s
pe
e
d.
S
in
c
e
th
e
in
it
ia
l
s
im
ul
a
ti
on
da
ta
us
e
d
is
s
ta
ti
c
tr
a
f
f
ic
li
ght
s
,
c
om
pa
r
in
g
th
e
a
ve
r
a
ge
w
a
it
in
g
ti
m
e
a
nd
s
pe
e
d
be
f
or
e
a
nd a
f
te
r
t
r
a
in
in
g c
a
n s
how
t
he
c
ha
nge
s
i
n t
he
a
ve
r
a
ge
w
a
it
in
g t
im
e
a
nd s
pe
e
d a
t
th
e
i
nt
e
r
s
e
c
ti
on w
he
n
us
in
g
s
ta
ti
c
ti
m
e
a
nd
th
e
D
R
L
a
ge
nt
.
W
e
a
ls
o
di
d
a
not
he
r
e
va
lu
a
ti
on
to
s
e
e
w
he
th
e
r
D
R
L
a
ge
nt
s
ha
ve
s
om
e
di
f
f
e
r
e
nt
r
e
s
ul
ts
w
he
n
f
a
c
e
d
w
it
h
tr
a
in
in
g
w
it
h
noi
s
e
. T
hi
s
e
va
l
ua
ti
on
w
il
l
c
om
pa
r
e
D
R
L
a
ge
nt
s
tr
a
in
e
d
u
s
in
g
noi
s
y da
ta
(
p
e
r
tu
r
ba
ti
on)
t
o a
ge
nt
s
t
ha
t
do not e
xpe
r
ie
nc
e
noi
s
e
in
t
he
ir
t
r
a
in
in
g.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
.
14
, N
o.
4
,
A
ugus
t
20
25
:
2991
-
3002
2996
3.
R
E
S
U
L
T
S
A
N
D
D
I
S
C
U
S
S
I
O
N
3.1.
V
e
h
ic
le
f
lo
w
c
ou
n
t
in
g r
e
s
u
lt
V
id
e
o
d
a
ta
ob
ta
in
e
d
f
r
om
B
a
l
i
P
r
ov
in
c
ia
l
T
r
a
ns
p
or
ta
ti
on
A
ge
nc
y
,
p
r
od
uc
e
d
f
r
om
s
u
r
ve
il
la
nc
e
c
a
m
e
r
a
s
a
t
th
e
P
a
m
e
li
s
a
n
i
nt
e
r
s
e
c
ti
on.
B
e
c
a
us
e
th
e
r
e
a
r
e
no
s
ur
ve
i
ll
a
nc
e
c
a
m
e
r
a
s
w
i
th
a
w
id
e
v
ie
w
a
t
th
e
P
a
m
e
li
s
a
n
in
t
e
r
s
e
c
t
io
n
,
a
n
d
o
nl
y
pa
n
-
ti
lt
-
z
oo
m
(
P
T
Z
)
c
a
m
e
r
a
s
a
r
e
a
va
i
la
b
le
,
da
ta
f
o
r
th
e
e
n
ti
r
e
la
ne
c
a
nn
ot
be
t
a
ke
n
a
t
o
nc
e
.
T
he
vi
de
o
w
a
s
ta
ke
n
f
o
r
4
d
a
ys
f
r
o
m
J
a
n
u
a
r
y
7
th
,
2
024
t
o
J
a
n
ua
r
y
11
th
,
2
024
,
to
ob
ta
i
n
vi
de
o
r
e
c
o
r
d
in
gs
o
f
ve
hi
c
le
s
p
a
s
s
in
g
t
h
r
ou
gh
e
a
c
h
la
ne
a
t
t
he
P
a
m
e
li
s
a
n
in
t
e
r
s
e
c
t
io
n
in
D
e
n
pa
s
a
r
,
o
ne
r
oa
d
a
r
m
f
o
r
o
ne
da
y
o
f
r
e
c
o
r
d
in
g.
F
o
r
4
da
ys
,
r
e
c
or
di
ngs
w
e
r
e
o
bt
a
in
e
d
f
r
o
m
e
a
c
h
r
oa
d
a
r
m
a
nd
e
a
c
h
l
a
ne
o
f
th
e
r
o
a
d
c
ou
ld
be
c
a
lc
u
la
te
d.
I
n
th
is
s
tu
dy
,
t
o
r
e
p
r
e
s
e
nt
th
e
c
ond
it
io
ns
a
n
d
a
ve
r
a
ge
de
ns
it
y
o
f
e
a
c
h
la
n
e
,
vi
de
o
r
e
c
o
r
di
n
gs
w
e
r
e
s
e
le
c
te
d
f
or
7,
20
0
s
e
c
o
nds
(
1
20
m
i
nut
e
s
)
a
t
th
e
s
a
m
e
ti
m
e
o
n
d
i
f
f
e
r
e
nt
da
ys
(
r
e
c
o
r
d
in
gs
o
f
one
a
r
m
a
nd
t
he
ot
he
r
w
e
r
e
on
d
if
f
e
r
e
n
t
da
ys
)
.
T
he
s
e
le
c
te
d
vi
de
o
t
im
e
w
a
s
13
.0
0
to
15.
00
W
I
T
A
.
T
he
obt
a
in
e
d
vi
de
o
c
a
lc
ul
a
te
d
ve
hi
c
le
f
lo
w
us
in
g
Y
O
L
O
v8
by
c
a
lc
ul
a
ti
ng
ve
hi
c
le
s
pa
s
s
in
g
th
r
ough
th
e
pe
r
im
e
te
r
box c
onf
ig
u
r
e
d a
s
i
n F
ig
u
r
e
2. T
he
ve
hi
c
le
c
la
s
s
e
s
t
ha
t
a
r
e
s
e
t
to
be
r
e
c
ogni
z
e
d by YO
L
O
v8
a
r
e
c
a
r
s
,
bus
e
s
, a
nd
tr
uc
ks
.
I
n
th
i
s
s
tu
dy,
w
e
di
d
not
c
a
lc
ul
a
te
th
e
f
l
ow
of
2
-
w
he
e
le
d
ve
hi
c
l
e
s
(
m
ot
or
c
yc
le
s
)
.
A
s
a
r
e
s
ul
t,
1,504
ve
hi
c
le
s
pa
s
s
e
d
th
r
ough
N
gur
a
h
R
a
i
1
a
nd
N
gur
a
h
R
a
i
2
la
ne
s
,
1,013
ve
hi
c
le
s
pa
s
s
e
d
th
r
ough
N
gur
a
h
R
a
i
3
a
nd
N
gur
a
h
R
a
i
4
la
ne
s
,
1,017
ve
hi
c
le
s
pa
s
s
e
d
th
r
ough
P
a
m
e
li
s
a
n
2
a
nd
236
ve
hi
c
le
s
pa
s
s
e
d
th
r
ough
P
a
m
e
li
s
a
n
1
la
ne
s
. T
a
bl
e
2
pr
e
s
e
nt
s
th
e
num
be
r
of
ve
h
ic
le
f
lo
w
s
(
ba
s
e
d
on
ob
s
e
r
va
ti
on
pos
it
io
n)
a
nd
th
e
de
s
ti
na
ti
on
of
ve
hi
c
le
s
pa
s
s
in
g
th
r
ough
th
e
P
a
m
e
li
s
a
n
in
te
r
s
e
c
ti
on.
I
t
s
houl
d
be
not
e
d
th
a
t
ob
s
e
r
va
ti
ons
w
e
r
e
not
c
a
r
r
ie
d
out
a
t
on
c
e
, s
o
f
or
N
gur
a
h
R
a
i
1
a
nd
N
gur
a
h R
a
i
2
obs
e
r
va
ti
ons
w
e
r
e
c
a
r
r
ie
d
out
on
th
e
f
ir
s
t
da
y,
N
gur
a
h
R
a
i
3
a
nd
N
gur
a
h
R
a
i
4
on
th
e
s
e
c
ond
da
y,
P
a
m
e
li
s
a
n
1
on
th
e
th
ir
d
da
y
a
nd
th
e
f
our
th
da
y
f
or
P
a
m
e
li
s
a
n 2.
T
a
bl
e
2. V
e
hi
c
l
e
f
lo
w
c
ount
in
g r
e
s
ul
t
D
e
pa
r
t
l
a
ne
A
r
r
i
va
l
l
a
ne
V
e
hi
c
l
e
c
ount
T
ot
a
l
P
a
m
e
l
i
s
a
n 1 (
S
out
h)
N
gur
a
h R
a
i
1 (
E
a
s
t
)
48
236
N
gur
a
h R
a
i
2 (
E
a
s
t
)
44
N
gur
a
h R
a
i
4 (
W
e
s
t
)
77
P
a
m
e
l
i
s
a
n 1 (
N
or
t
h)
67
P
a
m
e
l
i
s
a
n 2 (
N
or
t
h)
N
gur
a
h R
a
i
1 (
E
a
s
t
)
234
1017
N
gur
a
h R
a
i
1 (
E
a
s
t
)
#
2
N
gur
a
h R
a
i
3 (
W
e
s
t
)
687
N
gur
a
h R
a
i
3 (
W
e
s
t
)
#
23
P
a
m
e
l
i
s
a
n 2 (
S
out
h)
71
N
gur
a
h R
a
i
3 (
E
a
s
t
)
*
N
gur
a
h R
a
i
3 (
W
e
s
t
)
39
289
P
a
m
e
l
i
s
a
n 1
(
N
or
t
h)
250
N
gur
a
h R
a
i
4 (
E
a
s
t
)
*
N
gur
a
h R
a
i
4 (
W
e
s
t
)
652
724
P
a
m
e
l
i
s
a
n 2 (
S
out
h)
72
N
gur
a
h R
a
i
2 (
W
e
s
t
)
*
N
gur
a
h R
a
i
2 (
E
a
s
t
)
247
324
P
a
m
e
l
i
s
a
n 2 (
S
out
h)
77
N
gur
a
h R
a
i
1 (
W
e
s
t
)
*
N
gur
a
h R
a
i
1 (
E
a
s
t
)
418
1180
P
a
m
e
l
i
s
a
n 1 (
N
or
t
h)
762
# L
a
ne
c
ha
ngi
ng oc
c
ur
* A
s
t
he
pos
i
s
i
t
i
on of
de
pa
r
t
i
ng i
s
c
l
o
s
e
t
oge
t
he
r
, w
e
a
s
s
um
e
t
he
de
p
a
r
t
i
ng poi
nt
f
r
om
a
r
r
i
va
l
l
a
ne
3.2.
S
U
M
O
s
im
u
la
t
io
n
T
hi
s
pa
p
e
r
pr
e
s
e
nt
s
a
c
a
s
e
s
tu
dy
of
r
e
a
l
-
w
or
ld
tr
a
f
f
ic
a
t
P
a
m
e
li
s
a
n
i
nt
e
r
s
e
c
ti
on,
th
e
in
te
r
s
e
c
ti
on
of
P
a
m
e
li
s
a
n
R
oa
d
a
nd
N
gur
a
h
R
a
i
R
oa
d
in
D
e
npa
s
a
r
C
it
y.
W
e
s
im
ul
a
te
d
th
e
P
a
m
e
li
s
a
n
in
te
r
s
e
c
ti
on
us
in
g
th
e
S
U
M
O
s
im
ul
a
to
r
.
T
he
in
te
r
s
e
c
ti
on
la
yout
in
S
U
M
O
is
de
pi
c
te
d
in
F
ig
ur
e
4.
P
a
m
e
li
s
a
n
in
te
r
s
e
c
ti
on
f
e
a
tu
r
e
s
f
our
di
r
e
c
ti
ons
,
a
nd
N
gur
a
h
R
a
i
R
oa
d
ha
s
two
la
ne
s
.
T
he
le
f
t
m
os
t
la
ne
in
N
gur
a
h
R
a
i
is
de
s
ig
na
t
e
d
f
or
le
f
t
tu
r
ns
a
nd
goi
ng
s
tr
a
ig
ht
,
a
nd
th
e
r
ig
ht
m
os
t
la
ne
is
r
e
s
e
r
ve
d
f
or
r
ig
ht
tu
r
ns
,
goi
ng
s
tr
a
ig
ht
,
a
nd
u
-
tu
r
n.
P
a
m
e
li
s
a
n R
oa
d ha
s
one
l
a
ne
w
it
h no de
s
ig
na
te
d ve
hi
c
le
t
ur
ns
(
f
r
e
e
f
or
a
ll
t
u
r
ns
)
. I
n
a
ddi
ti
on t
o
bui
ld
in
g
r
oa
d
la
ne
s
, S
U
M
O
s
im
ul
a
ti
on a
ls
o s
im
ul
a
te
s
ve
hi
c
le
f
lo
w
s
,
a
s
de
s
c
r
i
be
d i
n T
a
bl
e
2.
T
he
P
a
m
e
li
s
a
n
in
te
r
s
e
c
ti
on
th
a
t
w
e
s
im
ul
a
te
d
u
s
in
g
S
U
M
O
ha
s
two
gr
e
e
n
li
ght
pha
s
e
s
a
nd
two
tr
a
ns
it
io
n
pha
s
e
s
.
O
ne
gr
e
e
n
li
ght
pha
s
e
is
f
or
N
gur
a
h
R
a
i
R
oa
d,
de
pi
c
t
e
d
in
F
ig
ur
e
4(
a
)
,
a
nd
one
f
or
P
a
m
e
li
s
a
n
R
oa
d,
de
pi
c
te
d
in
F
ig
ur
e
4(
b)
.
T
he
s
e
two
gr
e
e
n
pha
s
e
s
a
l
s
o
w
or
k
a
s
a
c
ti
on
s
ta
te
f
or
D
Q
N
.
E
a
c
h
pha
s
e
ha
s
a
tr
a
ns
it
io
n
pha
s
e
(
ye
ll
ow
li
ght
)
th
a
t
c
yc
le
be
twe
e
n
gr
e
e
n
a
nd
ye
ll
ow
ph
a
s
e
.
T
he
s
e
gr
e
e
n
a
nd
ye
ll
ow
pha
s
e
s
ha
ve
th
e
ir
dur
a
ti
on
a
nd
a
r
e
c
a
ll
e
d
s
ig
na
l
pha
s
e
dur
a
ti
on.
F
or
th
e
P
a
m
e
li
s
a
n
in
te
r
s
e
c
ti
on,
th
e
s
ta
nda
r
d (
f
ix
-
ti
m
e
d)
s
ig
na
l
pha
s
e
dur
a
ti
on i
s
s
how
n i
n T
a
bl
e
3.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
O
pt
imi
z
in
g t
r
af
fi
c
l
ig
ht
s
at
unbalanc
e
d i
nt
e
r
s
e
c
ti
ons
u
s
in
g de
e
p
r
e
in
fo
r
c
e
m
e
nt
…
(
D
um
an C
a
r
e
K
h
r
is
ne
)
2997
T
a
bl
e
3. P
a
m
e
li
s
a
n i
nt
e
r
s
e
c
ti
on f
ix
-
ti
m
e
d s
ig
na
l
pha
s
e
dur
a
ti
on
P
ha
s
e
D
ur
a
t
i
on (
s
e
c
ond)
C
yc
l
e
l
e
ngt
h (
s
e
c
ond)
N
gur
a
h R
a
i
gr
e
e
n pha
s
e
39
90
N
gur
a
h R
a
i
ye
l
l
ow
pha
s
e
6
P
a
m
e
l
i
s
a
n
gr
e
e
n pha
se
39
P
a
m
e
l
i
s
a
n
ye
l
l
ow
pha
s
e
6
3.
3
.
A
ge
n
t
t
r
ai
n
in
g
W
e
pe
r
f
or
m
a
ge
nt
tr
a
in
in
g
us
in
g
m
ul
ti
-
la
ye
r
pe
r
c
e
pt
r
on
(
M
LP
)
p
ol
ic
y
f
r
om
S
B
3,
hype
r
pa
r
a
m
e
te
r
s
f
or
th
e
m
ode
l
c
a
n
be
s
e
e
n
in
T
a
bl
e
4.
W
e
pe
r
f
or
m
tr
a
in
in
g
us
in
g
72,000
ti
m
e
s
te
ps
on
two
obs
e
r
va
ti
on
m
ode
s
e
tt
in
gs
.
T
he
f
ir
s
t
s
e
tt
in
g
is
us
in
g
obs
e
r
va
ti
ons
w
it
hout
noi
s
e
a
nd
th
e
s
e
c
ond
s
e
tt
in
g
is
us
in
g
obs
e
r
va
ti
ons
w
it
h
noi
s
e
.
T
h
e
s
e
tt
in
g
w
it
hout
noi
s
e
unde
r
ta
ke
s
th
e
ve
hi
c
l
e
f
lo
w
va
lu
e
in
T
a
bl
e
2
w
it
hout
c
ha
nge
s
.
T
h
e
s
e
tt
in
g
w
it
h
noi
s
e
pe
r
tu
r
ba
ti
on
is
c
a
r
r
ie
d
out
by
in
s
e
r
ti
ng
noi
s
e
in
to
th
e
l
e
ngt
h
of
th
e
ve
hi
c
le
que
ue
a
t
th
e
in
te
r
s
e
c
ti
on
s
o
th
a
t
th
e
r
e
is
a
c
ha
ng
e
in
th
e
ob
s
e
r
va
ti
on.
T
he
i
ns
e
r
te
d
noi
s
e
is
ge
n
e
r
a
te
d
r
a
ndoml
y
us
in
g
th
e
r
a
ndom G
a
us
s
ia
n f
unc
ti
on, with m
e
a
n =
0 a
nd s
ta
nd
a
r
d de
vi
a
ti
on =
1.
T
a
bl
e
4. H
ype
r
pa
r
a
m
e
te
r
va
lu
e
f
or
D
Q
N
m
ode
l
P
a
r
a
m
e
t
e
r
V
a
l
ue
L
e
a
r
ni
ng
r
a
t
e
1e
-
3
L
e
a
r
ni
ng
s
t
a
r
t
5
T
r
a
i
ni
ng
f
r
e
que
nc
y
4
G
a
m
m
a
0.9
E
xpl
or
a
t
i
on
f
r
a
c
t
i
on
0.1
E
xpl
or
a
t
i
on
f
i
na
l
e
pi
s
ode
0.05
T
a
r
ge
t
upda
t
e
i
nt
e
r
va
l
500
R
e
pl
a
y
buf
f
e
r
s
i
z
e
50000
A
f
te
r
tr
a
in
in
g
f
or
50
e
pi
s
ode
s
,
th
e
a
ve
r
a
g
e
r
e
w
a
r
d
obt
a
in
e
d
by
th
e
a
ge
nt
pe
r
e
pi
s
ode
is
s
um
m
a
r
iz
e
d
in
F
ig
ur
e
5.
F
ig
ur
e
5(
a
)
s
how
s
th
e
a
ve
r
a
ge
r
e
w
a
r
d
obt
a
in
e
d
b
y
th
e
a
ge
nt
w
he
n
tr
a
in
e
d
in
a
s
e
tt
in
g
w
it
hout
noi
s
e
,
a
t
th
e
be
gi
nni
ng
of
tr
a
in
in
g
th
e
r
e
w
a
r
d
obt
a
in
e
d
is
r
e
la
ti
ve
ly
s
m
a
ll
but
a
s
th
e
tr
a
in
in
g
e
pi
s
ode
s
in
c
r
e
a
s
e
,
hi
ghe
r
r
e
w
a
r
ds
a
r
e
obt
a
in
e
d
a
nd
te
nd
to
be
s
ta
bl
e
.
F
i
gur
e
5(
b
)
s
how
s
th
e
a
ve
r
a
ge
r
e
w
a
r
d
obt
a
in
e
d
by t
he
a
ge
nt
t
hr
ough tr
a
in
in
g w
it
h a
s
e
tt
in
g a
ddi
ng nois
e
. S
in
c
e
t
he
be
gi
nni
ng of
t
r
a
in
in
g, t
he
r
e
w
a
r
d obta
in
e
d
by
th
e
a
ge
nt
is
m
or
e
s
ta
bl
e
in
th
is
s
e
tt
in
g.
S
ta
bl
e
a
ge
nt
r
e
w
a
r
d
dur
in
g
a
ge
nt
tr
a
in
in
g
w
it
h
noi
s
e
be
c
a
us
e
th
e
a
ge
nt
le
a
r
ns
not
to
m
oni
to
r
th
e
que
ue
le
ngt
h
in
th
e
e
nvi
r
on
m
e
nt
to
o
m
uc
h
a
t
th
e
be
gi
nni
ng
of
t
r
a
in
in
g.
A
lt
hough
th
e
r
e
is
a
di
f
f
e
r
e
nc
e
in
it
ia
ll
y,
bot
h
a
ge
nt
s
ge
t
s
ta
bl
e
r
e
w
a
r
ds
a
t
th
e
e
nd
of
tr
a
in
in
g.
T
hi
s
r
e
w
a
r
d
s
how
s
th
a
t
th
e
a
ge
nt
ha
s
le
a
r
ne
d
how
to
opt
im
iz
e
a
t
th
is
in
te
r
s
e
c
ti
on
(
w
it
h
a
n
unba
la
nc
e
d
tr
a
f
f
ic
f
lo
w
)
.
E
vi
de
nc
e
f
or
th
is
s
ta
te
m
e
nt
c
a
n
be
s
e
e
n
in
F
ig
ur
e
6,
w
hi
c
h
s
how
s
th
a
t
th
e
a
ve
r
a
ge
w
a
it
in
g
ti
m
e
a
t
th
e
in
te
r
s
e
c
ti
on be
twe
e
n a
g
e
nt
s
l
ooks
c
lo
s
e
t
oge
th
e
r
.
(
a
)
(
b)
F
ig
ur
e
5. A
ge
nt
r
e
w
a
r
d w
hi
le
t
r
a
in
in
g (
a
)
w
it
hout
noi
s
e
pe
r
tu
r
b
a
ti
on a
nd (
b)
w
it
h pe
r
tu
r
ba
ti
on
-
1
6
0
-
1
4
0
-
1
2
0
-
1
0
0
-
8
0
-
6
0
-
4
0
-
2
0
0
4
8
12
16
20
24
28
32
36
40
44
48
50
M
ea
n
a
gen
t
r
ew
a
r
d
N
u
m
b
er
o
f
ep
i
s
o
d
es
M
e
a
n
a
g
e
n
t
r
e
w
a
r
d du
r
i
n
g
t
r
a
i
n
i
n
g
w
i
t
h
ou
t
p
e
r
t
u
r
b
a
t
i
o
n
-
2
0
-
1
8
-
1
6
-
1
4
-
1
2
-
1
0
-8
-6
-4
-2
0
4
8
12
16
20
24
28
32
36
40
44
48
50
M
ea
n
a
gen
t
r
ew
a
r
d
N
u
m
b
er
o
f
ep
i
s
o
d
es
M
e
a
n
a
g
e
n
t
r
e
wa
r
d du
r
i
n
g
t
r
a
i
n
i
n
g
wi
t
h
pe
r
t
u
r
ba
t
i
on
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
.
14
, N
o.
4
,
A
ugus
t
20
25
:
2991
-
3002
2998
F
i
g
u
r
e
6
.
I
n
t
e
r
s
e
c
t
i
o
n
a
v
e
r
a
g
e
w
a
i
t
i
n
g
t
i
m
e
d
u
r
i
n
g
t
r
a
i
n
i
n
g
w
i
t
h
o
u
t
n
o
i
s
e
p
e
r
t
u
r
b
a
t
i
o
n
a
n
d
t
r
a
i
n
i
n
g
w
i
t
h
p
e
r
t
u
r
b
a
t
i
o
n
T
r
a
in
in
g hi
s
to
r
y i
n F
ig
ur
e
6
a
ls
o s
how
s
t
ha
t
th
e
a
ge
nt
t
r
a
in
e
d w
it
hout
us
in
g nois
e
a
t
th
e
be
gi
nni
ng of
tr
a
in
in
g ha
s
di
f
f
ic
ul
ty
f
in
d
in
g a
ge
ne
r
a
li
z
a
ti
on pa
tt
e
r
n
of
w
a
it
in
g t
im
e
t
o
que
ue
l
e
ngt
h. T
hi
s
di
f
f
ic
ul
ty
c
a
us
in
g
th
e
a
ge
nt
to
e
xpl
or
e
a
t
th
e
be
gi
nni
ng
of
tr
a
in
in
g.
A
f
te
r
ge
tt
in
g
th
e
e
xpl
or
a
ti
on
pa
tt
e
r
n,
th
e
a
ge
nt
s
how
s
qui
te
good
e
xpl
oi
ta
ti
on,
s
o
th
a
t
a
t
th
e
e
nd
of
t
r
a
in
in
g
th
e
w
a
it
in
g
ti
m
e
a
t
th
e
in
te
r
s
e
c
ti
on
is
e
ve
n
be
tt
e
r
th
a
n
th
e
a
ge
nt
t
r
a
in
e
d us
in
g nois
e
. O
n t
he
ot
he
r
ha
nd, t
he
a
ge
nt
t
r
a
in
e
d
us
in
g nois
e
ha
s
a
m
uc
h s
m
a
ll
e
r
w
a
it
in
g t
im
e
a
t
th
e
be
gi
nni
ng
of
tr
a
in
in
g,
but
a
s
tr
a
in
in
g
p
r
ogr
e
s
s
e
s
,
th
e
w
a
it
in
g
ti
m
e
va
lu
e
te
nds
to
r
e
m
a
in
th
e
s
a
m
e
.
T
hi
s
s
how
s
t
ha
t
if
w
e
w
a
nt
f
a
s
te
r
e
xpl
oi
ta
ti
on, we
c
a
n us
e
t
r
a
in
in
g w
it
h nois
e
, but
i
f
w
e
w
a
nt
a
hi
ghe
r
w
a
it
in
g t
im
e
r
e
w
a
r
d, w
e
c
a
n us
e
t
r
a
in
in
g w
it
hout
noi
s
e
.
3.4. E
val
u
at
in
g op
t
im
i
z
at
io
n
r
e
s
u
lt
E
va
lu
a
ti
on
of
th
e
opt
im
iz
a
ti
on
pe
r
f
or
m
e
d
by
th
e
D
R
L
a
ge
nt
a
t
th
e
P
a
m
e
li
s
a
n
in
te
r
s
e
c
ti
on
is
m
e
a
s
ur
e
d
us
in
g
S
U
M
O
s
im
ul
a
ti
on
a
nd
bui
lt
us
in
g
a
n
unba
la
n
c
e
d
tr
a
f
f
ic
f
lo
w
.
F
ur
th
e
r
m
or
e
,
th
e
s
im
ul
a
ti
on
r
e
s
ul
ts
of
th
e
in
te
r
s
e
c
ti
on
opt
im
iz
e
d
u
s
in
g
th
e
D
R
L
a
ge
nt
w
e
r
e
c
om
pa
r
e
d
w
it
h
th
os
e
of
th
e
in
te
r
s
e
c
ti
on
s
im
ul
a
ti
on
w
it
hout
opt
im
iz
a
ti
on.
T
e
s
ti
ng
w
a
s
a
ls
o
c
onduc
te
d
us
in
g
a
s
im
ul
a
ti
on
c
ont
a
in
in
g
noi
s
e
to
s
e
e
w
he
th
e
r
th
e
a
ge
nt
c
oul
d
ove
r
c
om
e
noi
s
e
a
t
th
e
in
te
r
s
e
c
ti
on.
T
he
noi
s
e
u
s
e
d
dur
in
g
th
e
e
va
lu
a
ti
on
w
a
s
ge
ne
r
a
te
d i
n t
he
s
a
m
e
w
a
y a
s
dur
in
g t
r
a
in
in
g.
A
ge
nt
s
w
e
r
e
e
va
lu
a
te
d
on
one
s
im
ul
a
ti
on
e
pi
s
ode
w
it
h
noi
s
e
a
nd
one
w
it
hout
noi
s
e
(
bot
h
w
it
h
im
ba
la
nc
e
d
f
lo
w
s
)
.
T
hi
s
s
im
ul
a
ti
on
w
a
s
in
tr
oduc
e
d
to
a
ge
nt
s
tr
a
in
e
d
w
it
h
noi
s
e
a
nd
a
g
e
nt
s
tr
a
in
e
d
w
it
hout
noi
s
e
.
F
ig
ur
e
7
s
how
s
th
e
e
va
lu
a
ti
on
r
e
s
ul
ts
of
e
a
c
h
a
ge
nt
f
or
opt
im
iz
a
ti
on
w
it
h
w
a
it
in
g
ti
m
e
a
s
th
e
m
e
a
s
ur
e
d
pa
r
a
m
e
te
r
. F
ig
ur
e
7(
a
)
de
pi
c
ts
t
he
w
a
it
in
g t
im
e
va
lu
e
be
f
or
e
a
g
e
nt
opt
im
iz
a
ti
on, with a
n a
ve
r
a
ge
w
a
it
in
g t
im
e
of
594.49
s
e
c
ond
s
a
t
e
a
c
h
in
te
r
s
e
c
ti
on.
W
he
n
opt
im
iz
in
g
a
s
i
m
ul
a
ti
on
w
it
h
noi
s
e
,
F
ig
ur
e
7(
b)
de
pi
c
t
s
th
e
opt
im
iz
a
ti
on
out
c
om
e
s
obt
a
in
e
d
by
a
n
a
g
e
nt
tr
a
in
e
d
w
it
hout
noi
s
e
dur
in
g
it
s
tr
a
in
in
g
pe
r
io
d.
F
ig
ur
e
7(
c
)
de
pi
c
ts
th
e
opt
im
iz
a
ti
on
r
e
s
ul
ts
a
c
hi
e
ve
d
by
a
n
a
ge
nt
tr
a
in
e
d
on
noi
s
y
da
ta
dur
in
g
it
s
tr
a
in
in
g
pha
s
e
.
W
he
n
opt
im
iz
a
ti
on i
s
pe
r
f
or
m
e
d on a
noi
s
e
-
f
r
e
e
s
im
ul
a
ti
on, F
ig
ur
e
7(
d)
di
s
pl
a
ys
t
he
opt
im
iz
a
ti
on r
e
s
ul
ts
of
a
n a
ge
nt
tr
a
in
e
d
w
it
hout
noi
s
e
dur
in
g
it
s
t
r
a
in
in
g
pe
r
io
d.
F
ig
ur
e
7(
e
)
di
s
pl
a
ys
th
e
opt
im
iz
a
ti
on
out
c
om
e
s
of
a
n
a
ge
nt
tr
a
in
e
d w
it
h nois
y da
ta
dur
in
g i
ts
t
r
a
in
in
g pe
r
io
d.
A
c
c
or
di
ng
to
F
ig
ur
e
7,
th
e
a
ve
r
a
ge
w
a
it
in
g
ti
m
e
a
t
th
e
in
te
r
s
e
c
ti
on
is
lo
w
e
r
f
or
F
i
gur
e
s
7(
b)
to
7(
e
)
,
(
248.5,
173.1,
169.4,
a
nd
186.5
s
e
c
onds
r
e
s
pe
c
ti
ve
ly
)
th
a
n
f
or
th
e
in
te
r
s
e
c
ti
on
w
it
hout
opt
im
iz
a
ti
on
F
ig
ur
e
7(
a
)
.
F
ur
th
e
r
m
or
e
;
w
e
c
a
n
c
om
pa
r
e
F
ig
ur
e
s
7(
b)
a
nd
7(
c
)
to
s
e
e
th
e
a
ge
nt
'
s
opt
im
iz
a
ti
on
a
bi
li
ty
w
he
n
f
a
c
e
d
w
it
h
a
s
im
ul
a
ti
on
w
it
h
noi
s
e
on
th
e
que
ue
l
e
ngt
h.
F
r
om
th
e
s
e
two
im
a
g
e
s
,
w
e
c
a
n
s
e
e
th
a
t
th
e
a
g
e
nt
tr
a
in
e
d
w
it
h
noi
s
e
on
th
e
tr
a
in
in
g
da
ta
in
F
ig
ur
e
7(
c
)
,
c
a
n
be
c
om
e
a
c
c
us
to
m
e
d
to
opt
im
iz
in
g
a
t
th
e
in
te
r
s
e
c
ti
on
e
a
r
li
e
r
.
A
lt
hough
th
e
a
g
e
nt
in
F
ig
ur
e
7(
b)
w
a
s
s
im
il
a
r
ly
s
uc
c
e
s
s
f
ul
in
opt
im
iz
in
g,
it
is
c
le
a
r
th
a
t
th
e
r
e
w
e
r
e
m
ul
ti
pl
e
in
s
t
a
nc
e
s
w
he
r
e
th
e
a
g
e
nt
be
c
a
m
e
c
onf
us
e
d
by
th
e
noi
s
e
th
a
t
a
r
os
e
,
in
c
r
e
a
s
in
g
th
e
a
ve
r
a
ge
w
a
it
in
g
ti
m
e
.
F
ig
ur
e
s
7(
d)
a
nd
7
(
e
)
de
m
ons
tr
a
te
th
e
a
ge
nt
'
s
e
va
lu
a
ti
on
in
th
e
s
im
ul
a
ti
on
w
it
h
no
noi
s
e
on
qu
e
ue
le
ngt
h.
I
f
th
e
a
ge
nt
ha
d
di
f
f
ic
ul
ty
de
a
li
ng
w
it
h
noi
s
e
in
th
e
pr
e
vi
ous
e
va
lu
a
ti
on,
th
e
r
e
w
a
s
no
noi
s
e
in
th
e
s
im
ul
a
ti
on
th
is
ti
m
e
,
s
o
th
e
a
g
e
nt
tr
a
in
e
d
w
it
hout
noi
s
e
F
ig
ur
e
7
(
d)
a
nd
w
it
h
noi
s
e
F
ig
ur
e
7
(
e
)
ha
d
no
di
f
f
ic
ul
ty
opt
im
iz
in
g;
it
'
s
ju
s
t
th
a
t
th
e
a
ge
nt
in
e
va
lu
a
ti
on
F
ig
ur
e
7
(
d)
a
ppe
a
r
e
d
to
opt
im
iz
e
be
tt
e
r
w
he
n
m
e
a
s
ur
e
d
by
it
s
a
v
e
r
a
ge
w
a
it
in
g
ti
m
e
.
M
e
a
nw
hi
le
,
if
w
e
lo
ok
a
t
it
,
F
ig
ur
e
7
(
e
)
is
not
m
uc
h
di
f
f
e
r
e
nt
f
r
om
F
ig
ur
e
7
(
d
)
,
a
nd
th
e
opt
im
iz
a
ti
on
is
not
m
uc
h
di
f
f
e
r
e
nt
;
it
'
s
ju
s
t
th
a
t
be
c
a
u
s
e
it
is
tr
a
in
e
d
us
in
g
tr
a
in
in
g
da
ta
c
ont
a
in
in
g
noi
s
e
,
th
e
a
g
e
nt
is
m
or
e
c
a
r
e
f
ul
a
nd
ha
s
di
f
f
ic
ul
ti
e
s
a
t
th
e
b
e
gi
nni
ng
of
th
e
e
pi
s
ode
,
but
be
c
om
e
s
m
or
e
a
c
c
us
to
m
e
d a
t
th
e
e
nd of
t
he
e
va
lu
a
ti
on e
pi
s
od
e
.
1
4
5
9
.9
0
50
100
150
200
250
0
100
200
300
400
500
600
1
3
5
7
9
11
13
15
17
19
21
23
25
27
29
31
33
35
37
39
41
43
45
47
49
W
a
i
t
i
n
g
t
i
m
e
w
i
t
h
n
o
i
s
e
(
s
)
W
a
i
t
i
n
g
t
i
m
e
w
i
t
h
o
u
t
n
o
i
s
e
(
s
)
E
p
i
s
o
d
es
A
v
erage
wai
ti
n
g
ti
m
e
du
ri
n
g
trai
n
in
g
A
v
er
a
g
e
o
f
WT
Wi
t
h
o
u
t
N
o
i
se
T
r
a
i
n
i
n
g
A
v
er
a
g
e
o
f
WT
Wi
t
h
N
o
i
se
T
r
a
i
n
i
n
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
O
pt
imi
z
in
g t
r
af
fi
c
l
ig
ht
s
at
unbalanc
e
d i
nt
e
r
s
e
c
ti
ons
u
s
in
g de
e
p
r
e
in
fo
r
c
e
m
e
nt
…
(
D
um
an C
a
r
e
K
h
r
is
ne
)
2999
(
a
)
(
b)
(
c
)
(
d
)
(
e
)
F
ig
ur
e
7. W
a
it
in
g
ti
m
e
a
t
P
a
m
e
li
s
a
n i
nt
e
r
s
e
c
ti
on
unde
r
di
f
f
e
r
e
nt
s
tr
a
te
gi
e
s
:
(
a
)
w
it
hout
opt
im
iz
a
ti
on
,
(
b)
us
in
g D
R
L
a
ge
nt
t
r
a
in
e
d w
it
hout
noi
s
e
pe
r
tu
r
ba
ti
on
, (
c
)
us
in
g D
R
L
a
ge
nt
t
r
a
in
e
d w
it
h nois
e
pe
r
tu
r
ba
ti
on
,
(
d)
us
in
g D
R
L
a
ge
nt
t
r
a
in
e
d w
it
hout
noi
s
e
pe
r
tu
r
ba
ti
on i
n non nois
y t
r
a
f
f
ic
, a
nd
(
e
)
us
in
g D
R
L
a
ge
nt
t
r
a
in
e
d
w
it
h nois
e
pe
r
tu
r
ba
ti
on i
n non nois
y t
r
a
f
f
ic
A
pa
r
t
f
r
om
w
a
it
in
g
ti
m
e
,
e
va
lu
a
ti
on
is
a
ls
o
done
by
c
on
s
id
e
r
in
g
th
e
a
ve
r
a
ge
s
p
e
e
d
of
ve
hi
c
le
s
a
t
th
e
in
te
r
s
e
c
ti
on.
F
ig
ur
e
8
s
how
s
th
e
a
ve
r
a
ge
s
pe
e
d
of
ve
hi
c
le
s
w
he
n
th
e
e
va
lu
a
ti
on
s
im
ul
a
ti
on
is
c
a
r
r
ie
d
out
.
F
or
th
is
s
c
e
na
r
io
,
F
ig
ur
e
8(
a
)
s
ho
w
s
w
he
n
th
e
in
te
r
s
e
c
ti
on
i
s
not
opt
im
iz
e
d
us
in
g
a
n
a
ge
nt
.
F
ig
ur
e
8(
b)
s
how
s
th
e
a
ve
r
a
ge
s
pe
e
d
of
ve
hi
c
le
s
w
he
n
th
e
in
te
r
s
e
c
ti
on
is
opt
im
iz
e
d
us
in
g
a
n
a
ge
nt
tr
a
in
e
d
w
it
hout
noi
s
e
,
th
e
in
te
r
s
e
c
ti
on
s
im
ul
a
ti
on
is
not
a
f
f
e
c
te
d
by
noi
s
e
in
th
e
qu
e
ue
.
F
ig
ur
e
8(
c
)
s
how
s
th
e
a
ve
r
a
ge
s
p
e
e
d
of
ve
hi
c
le
s
a
t
a
n
in
te
r
s
e
c
ti
on
opt
im
iz
e
d
u
s
in
g
a
n
a
ge
nt
tr
a
in
e
d
u
s
in
g
noi
s
e
in
th
e
que
u
e
le
ngt
h,
th
e
e
va
lu
a
ti
on
M
e
a
n
:
594.4913901
0
200
400
600
800
1000
1200
1400
1600
0
515
1030
1545
2060
2575
3090
3605
4120
4635
5150
5665
6180
6695
W
a
i
t
i
n
g
t
i
m
e
(
s
)
S
i
m
u
l
a
t
i
on
t
i
m
e
s
t
e
p
(
s
)
W
a
i
t
i
n
g
t
i
m
e
on
s
t
a
t
i
c
t
r
a
f
f
i
c
l
i
g
h
t
t
i
m
e
(
w
i
t
h
ou
t
a
g
e
n
t
opt
i
m
i
z
a
t
i
on
)
M
e
a
n
:
248.5905621
0
500
1000
1500
2000
2500
3000
3500
0
455
910
1365
1820
2275
2730
3185
3640
4095
4550
5005
5460
5915
6370
6825
W
a
i
t
i
n
g
t
i
m
e
(
s
)
S
i
m
u
l
a
t
i
on
t
i
m
e
s
t
e
p
(
s
)
W
a
i
t
i
n
g
t
i
m
e
on
t
r
a
f
f
i
c
l
i
g
h
t
u
s
i
n
g
a
g
e
n
t
t
r
a
i
n
e
d
w
i
t
h
ou
t
n
oi
s
e
pe
r
t
u
r
ba
r
t
i
on
i
n
n
oi
s
y
t
r
a
f
f
i
c
M
e
a
n
:
173.1006246
0
100
200
300
400
500
600
700
800
900
0
485
970
1455
1940
2425
2910
3395
3880
4365
4850
5335
5820
6305
6790
W
a
i
t
i
n
g
t
i
m
e
(
s
)
S
i
m
u
l
a
t
i
on
t
i
m
e
s
t
e
p
(
s
)
W
ai
ti
n
g
ti
m
e
on
traffi
c
l
i
ght
u
si
n
g
agen
t
trai
n
ed
wi
th
n
oi
se
pertu
rb
arti
on
i
n
n
oi
sy
traff
i
c
M
e
a
n
:
169.4420541
0
100
200
300
400
500
600
700
800
900
1000
0
455
910
1365
1820
2275
2730
3185
3640
4095
4550
5005
5460
5915
6370
6825
W
a
i
t
i
n
g
t
i
m
e
(
s
)
S
i
m
u
l
a
t
i
on
t
i
m
e
s
t
e
p
(
s
)
W
a
i
t
i
n
g
t
i
m
e
on
t
r
a
f
f
i
c
l
i
g
h
t
u
s
i
n
g
a
g
e
n
t
t
r
a
i
n
e
d
w
i
t
h
ou
t
n
oi
s
e
pe
r
t
u
r
ba
r
t
i
on
i
n
n
on
n
oi
s
y
t
r
a
f
f
i
c
M
e
a
n
:
186.5052047
0
200
400
600
800
1000
1200
10
490
970
1450
1930
2410
2890
3370
3850
4330
4810
5290
5770
6250
6730
W
a
i
t
i
n
g
t
i
m
e
(
s
)
S
i
m
u
l
a
t
i
on
t
i
m
e
s
t
e
p
(
s
)
W
a
it
in
g
ti
m
e
on
tr
a
f
f
ic
li
g
h
t
u
s
in
g
a
g
e
n
t
tr
a
in
e
d
w
i
t
h
n
oi
s
e
pe
r
t
u
r
ba
r
t
i
on
i
n
n
on
n
oi
s
y
t
r
a
f
f
i
c
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
.
14
, N
o.
4
,
A
ugus
t
20
25
:
2991
-
3002
3000
in
te
r
s
e
c
ti
on
s
im
ul
a
ti
on
is
a
ls
o
a
f
f
e
c
te
d
by
th
e
s
a
m
e
noi
s
e
a
s
th
e
noi
s
e
dur
in
g
tr
a
in
in
g.
T
he
a
ve
r
a
ge
s
pe
e
d
of
c
a
r
s
tr
a
ve
li
ng
th
r
ough
th
e
in
te
r
s
e
c
ti
ons
,
w
h
e
th
e
r
opt
im
iz
e
d
or
not
,
is
not
s
ig
ni
f
ic
a
nt
ly
di
f
f
e
r
e
nt
,
a
s
s
how
n
in
F
ig
ur
e
s
8(
a
)
to
8(
c
)
(
5.6
-
5.9
m
/s
or
20.16
-
21.24
km
/h
)
.
G
iv
e
n
th
e
de
c
r
e
a
s
e
s
in
a
v
e
r
a
ge
w
a
it
in
g
ti
m
e
a
t
th
e
in
te
r
s
e
c
ti
on,
it
in
di
c
a
te
s
th
a
t
opt
im
iz
a
ti
on
ha
s
s
u
c
c
e
e
d
e
d
in
in
c
r
e
a
s
in
g
th
e
f
lo
w
of
ve
hi
c
le
s
pa
s
s
in
g
th
r
ough
th
e
in
te
r
s
e
c
ti
on
w
it
hout
r
e
qui
r
in
g
a
n
in
c
r
e
a
s
e
in
a
ve
r
a
ge
ve
hi
c
l
e
s
pe
e
d.
T
hi
s
s
ugge
s
ts
th
a
t
th
e
a
v
e
r
a
ge
s
pe
e
d
a
t
th
e
i
nt
e
r
s
e
c
ti
on doe
s
not
c
h
a
nge
m
uc
h, e
v
e
n w
it
h opti
m
iz
a
ti
on a
ge
nt
s
pr
e
s
e
nt
or
not
.
(
a
)
(
b)
(
c
)
F
ig
ur
e
8. A
ve
r
a
ge
s
pe
e
d a
t
in
te
r
s
e
c
ti
on w
it
h di
f
f
e
r
e
nt
s
tr
a
te
gy,
(
a
)
w
it
hout
opt
im
iz
a
ti
on,
(
b)
us
in
g D
R
L
a
ge
nt
tr
a
in
e
d w
it
hout
noi
s
e
pe
r
tu
r
ba
ti
on, a
nd
(
c
)
us
in
g D
R
L
a
ge
nt
t
r
a
in
e
d w
it
h nois
e
pe
r
tu
r
ba
ti
on
4.
C
O
N
C
L
U
S
I
O
N
T
hi
s
s
tu
dy
e
f
f
e
c
ti
ve
ly
us
e
d
th
e
D
Q
N
a
lg
or
it
hm
in
c
onj
unc
ti
on
w
it
h
D
R
L
to
opt
im
iz
e
tr
a
f
f
ic
s
ig
na
l
ti
m
e
a
t
th
e
P
a
m
e
li
s
a
n
in
te
r
s
e
c
ti
on
w
it
h
unba
la
nc
e
d
tr
a
f
f
ic
f
lo
w
.
V
e
hi
c
le
w
a
it
in
g
ti
m
e
w
a
s
s
ig
ni
f
ic
a
nt
ly
r
e
duc
e
d,
a
c
c
or
di
ng
to
e
xpe
r
im
e
nt
a
l
d
a
ta
.
W
it
h
th
e
D
R
L
a
ge
nt
t
r
a
in
e
d
w
it
hout
noi
s
e
,
th
e
a
v
e
r
a
ge
w
a
it
in
g
ti
m
e
dr
oppe
d
f
r
om
594.49
s
e
c
onds
in
th
e
ba
s
e
li
ne
s
ta
ti
c
tr
a
f
f
ic
s
ig
na
l
s
ys
te
m
to
a
s
lo
w
a
s
169.44
s
e
c
ond
s
,
a
nd
w
it
h
th
e
D
R
L
a
ge
nt
tr
a
in
e
d
w
it
h
noi
s
e
,
it
dr
oppe
d
to
173.
10
s
e
c
ond
s
.
T
h
e
s
e
e
nha
nc
e
m
e
nt
s
s
how
th
a
t
in
te
r
s
e
c
ti
on
e
f
f
ic
ie
nc
y
is
e
f
f
ic
ie
nt
ly
opt
im
iz
e
d
by
bot
h
noi
s
e
-
tr
a
in
e
d
a
nd
noi
s
e
-
f
r
e
e
D
R
L
a
ge
nt
s
,
w
it
h
th
e
noi
s
e
-
tr
a
in
e
d
a
ge
nt
de
m
ons
tr
a
ti
ng
s
upe
r
io
r
r
obus
tn
e
s
s
in
noi
s
y
e
nvi
r
onm
e
nt
s
.
F
ur
th
e
r
m
o
r
e
,
th
e
a
ve
r
a
ge
ve
hi
c
le
s
p
e
e
d
(
a
bout
5.6
–
5.97
m
/s
)
s
ta
ye
d
c
ons
ta
nt
a
c
r
os
s
s
i
tu
a
ti
ons
,
de
m
ons
tr
a
ti
ng
th
a
t
th
e
opt
im
iz
a
ti
on
e
nha
nc
e
d
tr
a
f
f
ic
f
lo
w
w
it
hout
s
a
c
r
if
ic
in
g
s
m
oot
hne
s
s
or
s
a
f
e
ty
.
T
he
s
e
f
in
di
ngs
de
m
ons
tr
a
te
how
D
R
L
-
ba
s
e
d
s
ys
te
m
s
h
a
ve
th
e
a
bi
li
ty
to
c
om
pl
e
te
ly
tr
a
ns
f
or
m
a
da
pt
iv
e
t
r
a
f
f
ic
c
ont
r
ol
in
in
tr
ic
a
te
a
nd
e
ve
r
-
c
ha
ngi
ng
m
e
tr
opol
it
a
n
s
e
tt
in
gs
.
B
y
us
in
g
D
R
L
-
ba
s
e
d
D
Q
N
,
th
e
in
te
r
s
e
c
ti
on
e
f
f
ic
ie
nc
y
is
s
ig
ni
f
ic
a
nt
ly
im
pr
ove
d,
de
m
ons
tr
a
ti
ng t
he
pot
e
nt
ia
l
f
or
i
m
pl
e
m
e
nt
a
ti
on i
n r
e
a
l
-
w
or
ld
s
c
e
na
r
io
s
.
F
U
N
D
I
N
G
I
N
F
O
R
M
A
T
I
O
N
T
he
F
a
c
ul
ty
of
E
ngi
ne
e
r
in
g
a
t
U
da
ya
na
U
ni
ve
r
s
it
y
a
nd
th
e
I
n
s
ti
tu
te
f
or
R
e
s
e
a
r
c
h
a
nd
C
om
m
uni
ty
S
e
r
vi
c
e
U
da
ya
na
U
ni
ve
r
s
it
y f
undi
ng t
hi
s
w
or
k w
it
h gr
a
nt
numbe
r
B
/2
55.3/UN14.4.A
/P
T
.01.03/2024.
M
e
a
n
:
5.629757273
0
2
4
6
8
10
12
14
16
18
0
481
962
1443
1924
2405
2886
3367
3848
4329
4810
5291
5772
6253
6734
A
v
e
r
a
g
e
s
pe
e
d
(
m
/s
)
S
i
m
u
l
a
t
i
on
t
i
m
e
s
t
e
p
(
s
)
A
v
e
r
a
g
e
s
pe
e
d
on
s
t
a
t
i
c
t
r
a
f
f
i
c
l
i
g
h
t
(
w
i
t
h
ou
t
a
g
e
n
t
opt
i
m
i
z
a
t
i
on
)
M
e
a
n
:
5.649423594
0
2
4
6
8
10
12
0
485
970
1455
1940
2425
2910
3395
3880
4365
4850
5335
5820
6305
6790
A
v
e
r
a
g
e
s
pe
e
d
(
m
/
s
)
S
i
m
u
l
a
t
i
on
t
i
m
e
s
t
e
p
(
s
)
A
v
e
r
a
g
e
s
pe
e
d
on
t
r
a
f
f
i
c
l
i
g
h
t
u
s
i
n
g
a
g
e
n
t
t
r
a
i
n
e
d
w
i
t
h
ou
t
n
oi
s
e
pe
r
t
u
r
ba
r
t
i
on
M
ea
n
:
5
.9
7
0
6
9
8
2
3
9
0
2
4
6
8
10
12
14
0
555
1
1
1
0
1
6
6
5
2
2
2
0
2
7
7
5
3
3
3
0
3
8
8
5
4
4
4
0
4
9
9
5
5
5
5
0
6
1
0
5
6
6
6
0
A
v
er
a
ge
s
p
ee
d
(
m
/
s
)
S
i
m
u
l
a
t
i
o
n
t
i
m
e
s
t
ep
(
s
)
A
v
er
a
ge
s
p
ee
d
o
n
t
r
a
f
f
i
c
l
i
gh
t
u
s
i
n
g
a
gen
t
t
r
a
i
n
ed
w
i
t
h
n
o
i
s
e
p
er
t
u
r
b
a
r
t
i
o
n
Evaluation Warning : The document was created with Spire.PDF for Python.