I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
pu
t
er
E
ng
ineering
(
I
J
E
CE
)
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
,
p
p
.
2748
~
2
7
5
7
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
15
i
3
.
pp
2
7
4
8
-
2
7
5
7
2748
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
Indo
o
r
na
v
ig
a
tio
n f
o
r mo
bile
robo
ts bas
ed on d
eep
reinforceme
nt
l
ea
rning
wit
h
conv
o
lutiona
l neural n
et
wo
rk
K
ho
a
Ng
uy
en
Da
ng
1
,
Va
n T
ra
n T
hi
2
,
Ng
uy
en
Va
n T
ha
n
g
3
1
F
a
c
u
l
t
y
o
f
En
g
i
n
e
e
r
i
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
I
n
t
e
r
n
a
t
i
o
n
a
l
S
c
h
o
o
l
,
V
i
e
t
n
a
m
N
a
t
i
o
n
a
l
U
n
i
v
e
r
si
t
y
,
H
a
n
o
i
,
V
i
e
t
n
a
m
2
F
a
c
u
l
t
y
o
f
G
e
n
e
r
a
l
E
d
u
c
a
t
i
o
n
,
U
n
i
v
e
r
si
t
y
o
f
La
b
o
u
r
a
n
d
S
o
c
i
a
l
A
f
f
a
i
r
s
,
H
a
n
o
i
,
V
i
e
t
n
a
m
3
U
n
i
v
e
r
s
i
t
y
o
f
E
n
g
i
n
e
e
r
i
n
g
a
n
d
Te
c
h
n
o
l
o
g
y
,
V
i
e
t
n
a
m
N
a
t
i
o
n
a
l
U
n
i
v
e
r
s
i
t
y
,
H
a
n
o
i
,
V
i
e
t
n
a
m
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
l 1
5
,
2
0
2
4
R
ev
is
ed
Dec
2
3
,
2
0
2
4
Acc
ep
ted
J
an
1
6
,
2
0
2
5
Th
e
m
o
b
il
e
r
o
b
o
t
is
a
n
i
n
telli
g
e
n
t
d
e
v
ice
th
a
t
c
a
n
a
c
h
iev
e
m
a
n
y
tas
k
s
in
li
fe
.
F
o
r
a
u
to
n
o
m
o
u
s,
n
a
v
ig
a
ti
o
n
b
a
s
e
d
o
n
th
e
li
n
e
o
n
th
e
g
r
o
u
n
d
is
o
ften
u
se
d
b
e
c
a
u
se
it
h
e
lp
s
th
e
r
o
b
o
t
to
m
o
v
e
a
lo
n
g
a
p
re
d
e
fin
e
d
p
a
t
h
,
sim
p
li
fies
th
e
p
a
th
p
la
n
n
i
n
g
,
a
n
d
re
d
u
c
e
s
th
e
c
o
m
p
u
tati
o
n
a
l
l
o
a
d
.
T
h
is
p
a
p
e
r
p
re
se
n
ts
a
m
e
th
o
d
fo
r
n
a
v
i
g
a
ti
n
g
th
e
f
o
u
r
-
w
h
e
e
l
m
o
b
i
le
ro
b
o
t
t
o
trac
k
a
li
n
e
b
a
se
d
o
n
a
d
e
e
p
Q
-
n
e
two
rk
a
s
a
c
o
n
tro
l
a
lg
o
rit
h
m
to
d
e
sire
th
e
a
c
ti
o
n
o
f
th
e
m
o
b
il
e
ro
b
o
t
a
n
d
a
c
a
m
e
ra
a
s
a
fe
e
d
b
a
c
k
se
n
so
r
t
o
d
e
tec
t
t
h
e
li
n
e
.
Th
e
c
o
n
tro
l
a
lg
o
rit
h
m
u
se
s
a
c
o
n
v
o
lu
t
io
n
n
e
u
ra
l
n
e
two
rk
(CNN
)
t
o
g
e
n
e
ra
te
t
h
e
m
o
b
il
e
ro
b
o
t
a
c
ti
o
n
,
d
e
fin
e
d
a
s
a
n
a
g
e
n
t
o
f
d
e
e
p
Q
-
n
e
two
r
k
.
CNN
u
se
s
ima
g
e
s
fro
m
th
e
c
a
m
e
ra
to
d
e
fin
e
th
e
sta
te
o
f
th
e
d
e
e
p
Q
n
e
two
r
k
.
T
h
e
sim
u
latio
n
s
a
re
p
e
rfo
rm
e
d
b
a
se
d
o
n
G
a
z
e
b
o
so
ftwa
re
wh
ich
in
c
lu
d
e
s
a
3
D
e
n
v
iro
n
m
e
n
t,
m
o
b
il
e
r
o
b
o
t
m
o
d
e
l,
li
n
e
,
a
n
d
P
y
t
h
o
n
p
r
o
g
ra
m
m
in
g
.
Th
e
re
su
l
ts
d
e
m
o
n
stra
te
th
e
h
i
g
h
-
p
e
rfo
rm
a
n
c
e
trac
k
in
g
o
f
m
o
b
il
e
ro
b
o
ts
wit
h
c
o
m
p
lex
li
n
e
traje
c
to
ries
,
a
c
h
iev
in
g
e
rr
o
rs
o
f
l
e
ss
th
a
n
1
0
0
p
x
,
wh
ic
h
is
c
o
m
p
a
re
d
with
th
e
trad
it
io
n
a
l
v
isi
o
n
m
e
th
o
d
(V
NS),
t
h
e
M
S
E
o
f
th
e
p
r
o
p
o
sa
l
m
e
th
o
d
is
0
.
0
2
6
4
l
o
we
r
th
a
n
VN
S
wit
h
0
.
0
4
0
6
.
S
h
o
wc
a
se
s
p
ro
v
e
d
c
o
n
v
in
c
in
g
l
y
th
a
t
e
ffe
c
ti
v
e
n
e
ss
su
g
g
e
ste
d
a
c
o
n
tr
o
l
a
p
p
ro
a
c
h
.
K
ey
w
o
r
d
s
:
C
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
s
Dee
p
Q
-
n
etwo
r
k
Gaz
eb
o
L
in
e
tr
ac
k
in
g
Mo
b
ile
r
o
b
o
t
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Kh
o
a
Ng
u
y
e
n
Dan
g
Facu
lty
o
f
E
n
g
in
ee
r
in
g
an
d
T
e
ch
n
o
lo
g
y
,
I
n
ter
n
atio
n
al
Sch
o
o
l,
Vietn
am
Natio
n
al
Un
iv
er
s
ity
1
4
4
Xu
a
n
T
h
u
y
R
o
ad
,
C
au
Gi
ay
,
Ha
No
i 1
0
0
0
0
0
,
Vietn
am
E
m
ail:
k
h
o
an
d
@
v
n
u
is
.
ed
u
.
v
n
1.
I
NT
RO
D
UCT
I
O
N
Mo
b
ile
r
o
b
o
ts
(
MR)
h
av
e
b
ee
n
r
ap
id
ly
d
ev
el
o
p
in
g
in
m
an
y
f
ield
s
in
life
s
u
c
h
as
in
d
u
s
tr
y
[
1
]
,
m
ilit
ar
y
[
2
]
,
a
n
d
m
e
d
icin
e
[
3
]
with
th
e
p
u
r
p
o
s
e
o
f
tr
an
s
p
o
r
tatio
n
[
4
]
,
ass
is
tan
ce
[
5
]
,
r
escu
e
[
6
]
,
ca
v
e
ex
p
lo
r
atio
n
[
7
]
,
an
d
o
p
er
atio
n
in
p
o
llu
ted
e
n
v
ir
o
n
m
en
ts
[
8
]
to
ch
ec
k
th
e
p
o
llu
tio
n
lev
el
o
f
th
e
en
v
ir
o
n
m
e
n
t.
B
ec
au
s
e
th
ey
ar
e
v
e
r
y
f
lex
ib
le
,
p
er
f
o
r
m
m
an
y
task
s
,
ar
e
au
t
o
n
o
m
o
u
s
,
an
d
ca
n
f
r
ee
u
p
h
u
m
an
h
an
d
s
.
Fo
r
ea
c
h
m
is
s
io
n
,
MR
is
d
esig
n
ed
with
d
if
f
er
en
t
t
y
p
es
as
two
-
wh
ee
l
m
o
b
ile
r
o
b
o
ts
(
T
W
MR),
an
d
f
o
u
r
-
w
h
ee
l
m
o
b
ile
r
o
b
o
ts
(
FW
MR),
[
9
]
,
[
1
0
]
.
E
ac
h
ty
p
e
h
as
ad
v
a
n
tag
es
an
d
d
is
ad
v
an
tag
es
in
t
h
e
s
p
ec
if
i
c
ap
p
licatio
n
s
.
Fo
r
ex
am
p
le,
T
W
MR
is
o
f
ten
s
m
all
in
s
ize,
a
n
d
s
m
all
p
ay
lo
ad
[
1
1
]
.
W
h
ile
FW
MR
h
as
m
o
r
e
r
ig
id
ity
,
lar
g
er
p
ay
lo
ad
,
b
etter
d
r
iv
ab
ilit
y
,
a
n
d
m
o
r
e
s
tab
ilit
y
in
co
r
n
er
i
n
g
[
1
2
]
.
I
n
g
en
e
r
al,
th
e
FW
M
R
is
o
f
ten
u
s
ed
to
d
em
o
n
s
tr
ate
co
n
tr
o
l
alg
o
r
ith
m
s
as
we
ll
a
s
ap
p
ly
th
em
to
r
ea
l
life
.
Ad
d
itio
n
s
,
th
e
F
W
M
R
is
s
im
i
lar
to
a
ca
r
v
eh
icle,
a
n
d
au
to
n
o
m
o
u
s
v
eh
icle
s
y
s
tem
s
ar
e
o
f
in
ter
est
b
o
t
h
in
d
o
o
r
s
an
d
o
u
td
o
o
r
s
n
o
wad
ay
s
.
W
ith
o
u
td
o
o
r
,
th
e
v
eh
icle
co
u
ld
u
s
e
th
e
g
lo
b
al
p
o
s
itio
n
in
g
s
y
s
tem
f
o
r
n
av
i
g
atin
g
.
Ho
wev
er
,
th
e
in
d
o
o
r
is
lack
s
th
is
,
wh
ich
h
as to
u
s
e
th
e
lan
d
m
ar
k
o
r
r
o
a
d
m
ar
k
in
g
s
to
co
m
p
u
te
its
p
o
s
itio
n
an
d
tr
ac
k
in
g
.
B
y
u
s
in
g
a
g
lo
b
al
s
en
s
o
r
lik
e
a
ca
m
er
a
[
1
3
]
,
th
e
b
r
o
ad
ca
s
tin
g
s
ig
n
al
is
s
en
t
to
all
m
o
b
ile
r
o
b
o
ts
,
an
d
th
e
p
o
s
itio
n
o
f
MR
is
d
etec
ted
b
ased
o
n
t
h
e
b
r
o
ad
ca
s
t
co
n
tr
o
l.
T
h
is
s
y
s
tem
co
u
ld
s
u
p
p
o
r
t
th
e
p
o
s
itio
n
f
o
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
I
n
d
o
o
r
n
a
vig
a
tio
n
fo
r
mo
b
ile
r
o
b
o
ts
b
a
s
ed
o
n
d
ee
p
r
ein
fo
r
ce
men
t le
a
r
n
in
g
…
(
K
h
o
a
N
g
u
ye
n
Da
n
g
)
2749
m
an
y
MRs
in
th
is
n
etwo
r
k
b
u
t
th
e
co
n
f
ig
u
r
atio
n
co
m
p
le
x
an
d
th
e
l
o
w
s
ig
n
al
tr
a
n
s
f
er
af
f
ec
t
to
p
e
r
f
o
r
m
an
ce
o
f
ea
ch
MR.
T
o
lo
ca
lized
co
n
tr
o
l
MR,
en
er
g
y
co
n
s
u
m
p
tio
n
co
m
p
ar
is
o
n
[
1
4
]
r
elatin
g
to
tr
ac
k
in
g
ac
cu
r
ac
y
is
u
s
ed
to
s
tr
ict
cr
iter
io
n
.
T
h
is
wo
r
k
s
h
o
ws
th
e
h
ig
h
p
er
f
o
r
m
an
ce
o
f
MR
in
tr
ac
k
in
g
p
o
s
itio
n
f
ield
,
b
u
t
m
ain
tain
in
g
en
e
r
g
y
a
n
d
its
co
m
p
ar
is
o
n
ar
e
d
if
f
icu
lt
wo
r
k
.
I
t
n
ee
d
ed
h
ig
h
ac
cu
r
ac
y
to
d
ec
i
d
e
th
e
co
n
tr
o
ller
f
o
r
MR.
I
f
th
is
p
er
f
o
r
m
an
ce
is
r
ed
u
ce
d
f
o
r
s
o
m
e
r
ea
s
o
n
,
th
e
tr
a
jecto
r
y
o
f
MR
is
af
f
ec
ted
an
d
d
o
es
n
o
t
f
o
llo
w
th
e
p
u
r
p
o
s
e.
Oth
er
r
esear
ch
r
elate
d
to
th
e
co
n
t
r
o
l
o
f
ea
c
h
ac
tu
a
to
r
with
d
ea
d
-
z
o
n
e
in
p
u
t
[
1
5
]
is
to
m
ain
tain
th
e
tr
ajec
to
r
y
o
f
MR,
wh
ich
co
u
ld
p
r
o
v
e
th
e
s
tab
ilit
y
o
f
th
e
L
y
a
p
u
n
o
v
f
u
n
ctio
n
b
u
t
it
is
d
if
f
ic
u
lt
to
ap
p
ly
t
h
em
to
r
ea
l
d
ev
ices
an
d
en
v
ir
o
n
m
en
t
s
.
B
esid
es,
an
ad
ap
tiv
e
co
n
tr
o
l
alg
o
r
ith
m
b
ased
o
n
s
lid
in
g
m
o
d
e
c
o
n
tr
o
l
[
1
6
]
was
d
ev
elo
p
ed
f
o
r
th
e
later
al
s
tab
ilit
y
o
f
m
o
b
ile
r
o
b
o
ts
in
b
o
th
s
im
u
latio
n
an
d
ex
p
er
im
en
tal.
Her
ein
,
th
e
MR
co
u
ld
well
tr
ac
k
th
e
S
p
ath
tr
a
jecto
r
y
.
Ho
wev
er
,
t
h
e
v
ib
r
ate
d
ap
p
ea
r
s
in
th
e
y
aw
r
ate
an
d
s
lip
an
g
le
in
th
e
f
ast
u
p
d
ate.
No
r
m
ally
,
th
e
c
o
n
tr
o
l
alg
o
r
ith
m
s
b
ased
o
n
lan
d
m
ar
k
s
ar
e
co
m
p
le
x
an
d
n
ee
d
m
o
r
e
s
en
s
o
r
s
to
d
etec
t
th
e
en
v
ir
o
n
m
e
n
t.
T
h
e
r
ef
o
r
e
,
u
s
in
g
lin
e
m
ar
k
in
g
s
s
u
ch
as
co
l
o
r
lin
es
an
d
m
a
g
n
etis
m
lin
es
t
o
g
u
i
d
e
MR
m
o
v
es
to
th
e
tar
g
et
is
o
f
ten
co
n
s
id
er
e
d
in
m
an
y
r
esear
ch
a
n
d
ap
p
lic
atio
n
s
.
Her
ein
,
th
e
co
s
t o
f
th
e
m
ag
n
etis
m
lin
e
an
d
its
s
en
s
o
r
f
o
r
d
etec
tin
g
ar
e
m
o
r
e
ex
p
en
s
iv
e
th
an
th
e
c
o
lo
r
l
in
e.
I
n
th
is
p
ap
er
,
we
s
elec
t
t
h
e
b
lack
lin
e
o
n
th
e
g
r
o
u
n
d
as th
e
m
a
r
k
s
f
o
r
th
e
M
R
.
T
h
e
d
esig
n
in
g
a
u
to
m
atic
n
a
v
ig
atio
n
alg
o
r
ith
m
s
f
o
r
MR
b
ased
o
n
th
e
b
lack
lin
e
h
av
e
b
ee
n
im
p
lem
en
ted
b
y
u
s
in
g
an
in
f
r
ar
ed
s
en
s
o
r
(
I
R
)
[
1
7
]
an
d
a
ca
m
er
a
s
en
s
o
r
[
1
8
]
.
B
o
th
s
en
s
o
r
s
d
eliv
er
h
i
g
h
ef
f
ec
tiv
en
ess
f
o
r
ea
ch
ap
p
licatio
n
.
Ho
wev
er
,
I
R
is
o
f
ten
af
f
ec
ted
b
y
lig
h
t
in
ten
s
ity
an
d
t
h
e
co
n
d
itio
n
o
f
th
e
en
v
ir
o
n
m
en
t,
wh
ile
th
e
ca
m
er
a
h
as m
o
r
e
u
s
ef
u
l in
f
o
r
m
atio
n
b
y
th
e
lar
g
er
v
iew
o
f
th
e
p
o
in
t
.
A
co
m
b
in
atio
n
o
f
ar
tific
ial
in
tellig
en
ce
(
AI
)
tr
e
n
d
s
is
d
ev
elo
p
in
g
in
all
f
ield
s
o
f
life
.
T
h
en
AI
h
elp
s
MR
to
n
av
ig
ate
lin
es
is
f
o
cu
s
ed
in
s
o
m
e
r
esear
ch
[
1
9
]
,
[
2
0
]
wh
ich
u
s
es
an
ar
tific
ial
n
eu
r
al
n
etwo
r
k
(
ANN)
to
co
m
p
u
te
th
e
s
p
ee
d
o
f
ea
ch
ac
tu
ato
r
o
f
MR.
W
ith
th
i
s
s
tr
u
ctu
r
e,
t
h
e
d
ataset
f
o
r
tr
ai
n
in
g
t
h
e
weig
h
t o
f
ANN
n
ee
d
s
a
lar
g
er
s
ize
an
d
is
d
ef
in
ed
in
d
etail.
T
o
o
v
er
co
m
e
th
is
lim
itatio
n
,
r
ein
f
o
r
ce
m
en
t
lear
n
i
n
g
(
R
L
)
is
th
e
m
o
s
t
u
s
ed
in
au
to
m
atic
n
av
ig
atio
n
b
ec
au
s
e
o
f
its
s
elf
-
lear
n
in
g
,
e
x
p
lo
r
atio
n
,
an
d
d
is
co
v
er
y
o
f
th
e
en
v
ir
o
n
m
en
t
[
2
1
]
.
Ho
wev
er
,
it
h
as
lim
itatio
n
s
r
elate
d
to
th
e
m
em
o
r
y
in
th
e
co
n
tin
u
o
u
s
en
v
ir
o
n
m
en
t
d
u
e
to
th
e
s
elf
-
lear
n
in
g
p
r
o
ce
s
s
wh
ich
co
u
ld
b
e
s
o
lv
ed
b
y
a
c
o
m
b
in
atio
n
o
f
R
L
an
d
d
ee
p
lear
n
in
g
(
DL
)
ca
lled
d
ee
p
r
ei
n
f
o
r
ce
m
en
t
lea
r
n
in
g
(
DR
L
)
wh
ich
u
s
es
th
e
lo
o
p
b
etwe
en
ag
e
n
t
an
d
en
v
ir
o
n
m
en
t
to
f
in
is
h
th
e
co
n
tr
o
l
task
s
.
Her
ein
,
en
v
ir
o
n
m
en
t
p
r
o
v
id
es
th
e
State
an
d
r
ewa
r
d
-
lik
e
f
ee
d
b
ac
k
co
n
tr
o
l
f
o
r
th
e
ag
en
t
.
Af
ter
h
av
in
g
th
is
in
f
o
r
m
atio
n
,
th
e
a
g
en
t
will
g
en
er
ate
th
e
ac
tio
n
co
r
r
esp
o
n
d
in
g
to
s
en
d
b
ac
k
to
th
e
en
v
i
r
o
n
m
e
n
t
lik
e
th
e
co
n
tr
o
l
s
y
s
tem
.
A
tab
le
in
ag
en
t
to
m
ap
b
etwe
en
s
tate
in
p
u
t
an
d
ac
tio
n
o
u
tp
u
t.
Ho
we
v
er
,
th
e
m
em
o
r
y
s
p
ac
e
n
ee
d
s
to
b
e
lar
g
e
an
d
q
u
ick
ly
ac
ce
s
s
ib
le.
Fo
r
th
is
r
ea
s
o
n
,
a
n
eu
r
al
n
etw
o
r
k
(
NN)
is
u
s
ed
to
r
ep
lace
th
e
m
ap
p
in
g
tab
le
wh
ich
is
ca
lled
a
d
ee
p
Q
n
etwo
r
k
(
DQN)
[
2
2
]
.
T
o
p
r
ese
n
t
N
N
in
DQN
f
o
r
c
o
n
t
r
o
lli
n
g
t
h
e
m
o
b
il
e
r
o
b
o
t,
t
h
e
ANN
[
2
3
]
,
s
i
n
g
le
s
h
o
t
m
u
lt
ib
o
x
d
ete
ct
o
r
[
2
2
]
an
d
c
o
n
v
o
l
u
t
io
n
al
n
e
u
r
a
l
n
etw
o
r
k
[
2
4
]
h
a
v
e
b
ee
n
c
o
n
s
id
e
r
ed
u
s
in
g
t
h
e
r
ed
,
g
r
e
e
n
,
b
l
u
e
-
d
e
p
t
h
(
R
GB
-
D
)
an
d
li
d
a
r
s
en
s
o
r
s
to
d
et
ec
t
t
h
e
m
ar
k
s
in
t
h
e
e
n
v
i
r
o
n
m
e
n
ts
.
So
,
t
h
e
DQN
i
n
th
e
n
av
ig
ati
o
n
f
i
el
d
f
o
r
a
m
o
b
il
e
r
o
b
o
t
b
ase
d
o
n
li
n
e
tr
ac
k
i
n
g
is
a
n
e
w
p
o
i
n
t
in
t
h
is
p
ap
er
.
I
n
p
ar
t
ic
u
la
r
,
C
NN
is
a
n
ar
ch
ite
ct
u
r
e
i
n
d
e
ep
le
ar
n
i
n
g
th
a
t
co
u
l
d
r
ea
ch
h
i
g
h
ac
c
u
r
a
cy
i
n
class
if
i
ca
ti
o
n
a
n
d
s
e
g
m
en
ta
tio
n
b
ase
d
o
n
im
a
g
es
f
r
o
m
a
c
am
e
r
a
[
2
5
]
.
I
n
t
h
is
p
a
p
e
r
,
t
h
e
ca
m
e
r
a
is
u
s
ed
to
c
ap
t
u
r
e
th
e
i
m
a
g
e
f
r
o
m
t
h
e
e
n
v
ir
o
n
m
e
n
t
a
n
d
a
n
al
y
ze
i
t
t
o
d
et
er
m
i
n
e
t
h
e
c
o
n
t
r
o
l
s
ig
n
al
f
o
r
MR
b
ase
d
o
n
DR
L
.
A
n
d
C
N
N
is
s
ele
cte
d
f
o
r
t
h
e
DQN
s
tr
u
ct
u
r
e
in
DR
L
.
F
u
r
t
h
er
m
o
r
e,
t
h
e
m
o
b
i
le
r
o
b
o
t
m
u
s
t
b
e
s
i
m
u
lat
ed
b
e
f
o
r
e
a
p
p
ly
in
g
it
t
o
r
e
al
e
x
p
er
im
en
ts
t
o
a
v
o
i
d
t
r
o
u
b
le
-
r
ela
te
d
co
n
t
r
o
l
al
g
o
r
it
h
m
s
,
p
a
r
a
m
e
te
r
s
,
a
n
d
e
n
v
ir
o
n
m
e
n
ta
l
c
o
n
d
iti
o
n
s
.
T
h
e
s
im
u
l
ati
o
n
s
t
ep
c
o
u
l
d
h
el
p
r
e
d
u
c
e
th
e
c
o
s
t
o
f
t
h
e
d
e
v
el
o
p
m
e
n
t
p
r
o
d
u
ct
as
w
ell
as
test
t
h
e
co
n
t
r
o
l
al
g
o
r
it
h
m
.
T
h
is
r
ese
ar
ch
u
s
es
t
h
e
i
m
a
g
e
f
r
o
m
th
e
ca
m
er
a
f
o
r
t
r
a
in
in
g
t
h
e
C
NN
s
tr
u
ct
u
r
e
t
o
m
a
k
e
it
m
o
r
e
in
tell
ig
e
n
t.
Fo
r
s
im
u
l
ati
o
n
,
th
e
Ga
ze
b
o
s
i
m
u
lat
io
n
s
o
f
tw
ar
e
is
s
u
g
g
este
d
to
p
r
ese
n
t
t
h
e
e
n
v
i
r
o
n
m
e
n
t
,
s
e
n
s
o
r
m
o
d
el
(
ca
m
e
r
a
)
,
a
n
d
m
o
b
il
e
r
o
b
o
t
m
o
d
e
l
b
as
ed
o
n
p
h
y
s
ics
en
g
i
n
es
[
2
6
]
.
I
n
th
is
p
ap
er
,
we
p
r
o
p
o
s
e
t
o
u
s
e
th
e
class
ical
DQN
s
tr
u
ctu
r
e
i
n
DR
L
to
g
e
n
er
ate
ac
tio
n
f
o
r
n
av
ig
atin
g
th
e
m
o
b
ile
r
o
b
o
t
with
o
n
e
R
GB
ca
m
er
a
to
f
o
ll
o
w
th
e
lin
e.
T
h
e
DQN
is
d
ev
elo
p
ed
b
ased
o
n
th
e
C
NN
s
tr
u
ctu
r
e.
An
d
s
im
u
latio
n
is
b
u
ilt
b
ased
o
n
Gaz
eb
o
s
o
f
twar
e
to
p
r
esen
t
th
e
3
D
en
v
i
r
o
n
m
en
ts
,
m
o
d
elin
g
m
o
b
ile
r
o
b
o
ts
,
a
n
d
s
en
s
o
r
s
(
c
am
er
a)
.
T
h
e
s
im
u
latio
n
r
esu
lt
s
s
h
o
w
th
e
e
f
f
ec
tiv
en
ess
o
f
th
e
p
r
o
p
o
s
al
m
eth
o
d
th
at
MB c
o
u
ld
tr
ac
k
th
e
b
lack
lin
e
with
s
m
all
er
r
o
r
s
(
less
th
an
1
0
0
p
x
)
an
d
c
o
m
p
lex
d
esire
d
tr
ajec
to
r
y
.
T
h
e
r
e
s
t
o
f
t
h
e
p
a
p
e
r
i
s
p
r
e
s
e
n
te
d
a
s
f
o
l
l
o
w
s
:
s
e
ct
i
o
n
2
i
s
m
o
b
i
l
e
r
o
b
o
t
m
o
d
e
l
i
n
g
w
h
i
l
e
t
h
e
o
v
e
r
a
l
l
R
L
a
n
d
D
Q
N
a
l
g
o
r
i
t
h
m
a
r
e
p
r
e
s
en
t
e
d
i
n
s
e
c
t
i
o
n
3
.
N
e
x
t
,
t
h
e
p
r
o
p
o
s
e
d
c
o
n
t
r
o
l
a
l
g
o
r
i
t
h
m
f
o
r
m
o
b
i
l
e
r
o
b
o
t
s
u
s
i
n
g
DQN
a
g
e
n
t
b
y
t
h
e
c
o
n
v
o
l
u
t
i
o
n
n
e
u
r
a
l
n
e
t
w
o
r
k
i
s
d
e
v
e
l
o
p
e
d
i
n
s
e
c
t
i
o
n
4
.
S
e
c
t
i
o
n
5
p
r
e
s
e
n
t
s
r
e
s
u
l
t
s
o
b
t
a
i
n
e
d
i
n
t
h
e
G
a
z
e
b
o
s
i
m
u
l
a
ti
o
n
e
n
v
i
r
o
n
m
e
n
t
t
o
v
e
r
i
f
y
t
h
e
c
o
n
t
r
o
l
p
e
r
f
o
r
m
a
n
c
e
a
n
d
s
e
c
ti
o
n
6
i
s
t
h
e
c
o
n
c
l
u
s
i
o
n
.
2.
M
E
T
H
O
D
2
.
1
.
M
o
bil
e
ro
bo
t
m
o
delin
g
A
f
o
u
r
-
w
h
ee
led
m
o
b
ile
r
o
b
o
t
(
FW
MR)
co
n
s
is
t
s
o
f
f
o
u
r
wh
ee
ls
with
a
r
ad
iu
s
o
f
r
.
Her
ein
,
two
wh
ee
ls
in
ea
ch
f
r
o
n
t
an
d
r
ea
r
m
u
s
t
b
e
lin
k
ed
to
en
s
u
r
e
th
e
y
h
av
e
th
e
s
am
e
s
p
ee
d
to
p
r
ev
e
n
t
th
e
m
o
b
ile
r
o
b
o
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
2
7
4
8
-
2
7
5
7
2750
f
r
o
m
s
lip
p
i
n
g
.
T
h
e
m
o
b
ile
r
o
b
o
t
m
o
d
el
a
n
d
its
g
eo
m
etr
y
ar
e
p
r
esen
ted
i
n
Fig
u
r
e
1
[
1
]
,
wh
ic
h
co
u
ld
b
e
d
etec
ted
b
y
t
h
e
lin
k
b
etwe
en
a
g
lo
b
al
f
r
am
e
(
X,
Y)
an
d
a
lo
c
al
f
r
am
e
(
x
,
y
)
; a
n
d
th
e
o
r
ien
ta
tio
n
.
T
h
e
lin
ea
r
v
elo
city
o
f
ea
ch
w
h
ee
l
is
co
m
p
u
ted
as
=
[
1
]
.
W
h
er
e
,
,
an
d
ar
e
lin
ea
r
v
elo
city
,
r
a
d
iu
s
,
an
d
an
g
u
lar
o
f
th
e
i
th
wh
ee
l
(
i=1
.
.
4
)
alo
n
g
th
e
x
-
ax
is
,
r
esp
ec
tiv
ely
.
L
et’
s
d
ef
in
e
th
e
lin
ea
r
v
elo
cities o
f
th
e
lef
t (
v
L
)
an
d
r
ig
h
t sid
e
(
v
R
)
o
f
th
e
m
o
b
ile
r
o
b
o
t a
s
=
1
=
4
an
d
=
2
=
3
.
T
h
e
lin
ea
r
v
elo
city
(
v
x
,
v
y
)
an
d
an
g
u
lar
v
elo
city
(
w
)
o
f
th
e
v
e
h
icle
alo
n
g
th
e
x
-
a
xis
co
u
ld
b
e
r
ep
r
esen
ted
as
=
+
2
,
=
−
+
,
a
n
d
=
0
.
W
h
er
e
2c
is
th
e
d
is
tan
ce
b
etwe
en
t
wo
wh
ee
ls
W
1
an
d
W
2
in
Fig
u
r
e
1
an
d
[
]
=
1
[
]
.
T
h
e
tar
g
et
o
f
co
n
tr
o
l is to
m
o
v
e
th
e
m
o
b
ile
r
o
b
o
t to
a
ta
r
g
et
d
ef
i
n
ed
as
X
,
Y,
an
d
o
r
ien
tatio
n
.
T
h
ey
ar
e
p
r
esen
te
d
b
y
th
e
k
in
em
atic
e
q
u
atio
n
o
f
th
e
FW
MR a
s
in
(
1
)
.
[
̇
̇
̇
]
=
[
−
0
0
0
0
1
]
[
ω
]
(
1
)
I
n
th
is
p
ap
er
,
th
e
FW
MR
m
o
d
el
is
u
s
ed
to
d
em
o
n
s
tr
ate
t
h
e
tr
ac
k
in
g
o
f
th
e
lin
e
alg
o
r
i
th
m
b
ased
o
n
d
ee
p
r
ein
f
o
r
ce
m
e
n
t
lear
n
in
g
with
DQN
ty
p
e.
Her
ein
,
DQL
u
s
e
s
th
e
C
NN
s
tr
u
ctu
r
e
with
im
ag
e
in
p
u
t
wh
ich
is
ca
p
tu
r
ed
f
r
o
m
t
h
e
ca
m
er
a.
DR
L
is
to
g
en
er
ate
th
e
Actio
n
as th
e
co
n
tr
o
l sig
n
al
to
th
e
m
o
b
ile
r
o
b
o
t.
Fig
u
r
e
1
.
Fo
u
r
-
wh
ee
led
m
o
b
il
e
r
o
b
o
t
m
o
d
elin
g
2
.
2
.
Reinf
o
rc
em
ent
lea
rning
a
nd
deep
Q
-
net
wo
rk
I
n
r
ei
n
f
o
r
ce
m
en
t
lear
n
in
g
(
R
L
)
,
th
e
r
elatio
n
s
h
ip
b
etwe
en
th
e
ag
en
t
an
d
en
v
ir
o
n
m
en
t
is
s
h
o
wn
in
Fig
u
r
e
2
.
He
r
ein
,
th
e
ag
en
t
r
e
ce
iv
es
s
tates
(
s
t
)
an
d
r
ewa
r
d
(
r
t
)
f
r
o
m
th
e
en
v
ir
o
n
m
en
t
,
w
h
ich
will
s
elec
t
th
e
ac
tio
n
(
a
t
)
b
ased
o
n
s
t
an
d
p
o
li
cy
as
in
(
2
)
[
2
7
]
.
T
er
m
is
th
e
ep
s
ilo
n
co
ef
f
icie
n
t
wh
ich
p
la
y
s
a
cr
itical
r
o
le
in
m
an
a
g
in
g
th
e
tr
a
d
e
-
o
f
f
b
et
wee
n
ex
p
l
o
r
atio
n
(
ag
e
n
t
tr
y
i
n
g
o
u
t
n
ew
ac
tio
n
s
to
d
is
co
v
e
r
th
eir
ef
f
ec
ts
an
d
r
ewa
r
d
s
)
an
d
e
x
p
lo
itatio
n
(
a
g
en
t
u
s
in
g
its
cu
r
r
en
t
k
n
o
wled
g
e
to
m
ax
im
ize
r
ewa
r
d
)
,
e
n
a
b
lin
g
th
e
ag
e
n
t
to
lear
n
ef
f
ec
tiv
ely
f
r
o
m
its
en
v
ir
o
n
m
en
t,
is
th
e
s
elec
ted
ac
t
io
n
,
m
a
x
(
.
)
is
th
e
in
d
ex
o
f
th
e
m
ax
im
u
m
v
alu
e
an
d
(
)
is
th
e
v
alu
e
at
.
T
h
e
is
o
b
tain
ed
b
y
f
in
d
in
g
t
h
e
m
a
x
o
f
(
,
)
o
f
in
T
ab
le.
1
.
Fig
u
r
e
2
.
R
ein
f
o
r
ce
m
en
t le
ar
n
in
g
s
tr
u
ctu
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
I
n
d
o
o
r
n
a
vig
a
tio
n
fo
r
mo
b
ile
r
o
b
o
ts
b
a
s
ed
o
n
d
ee
p
r
ein
fo
r
ce
men
t le
a
r
n
in
g
…
(
K
h
o
a
N
g
u
ye
n
Da
n
g
)
2751
=
{
m
a
x
(
(
)
)
[
0
,
1
]
<
(
[
1
,
]
)
[
0
,
1
]
≥
ℎ
,
∈
[
1
,
.
.
,
]
(
2
)
(
,
)
=
+
1
+
m
a
x
(
(
+
1
,
+
1
)
)
(
3
)
I
n
T
ab
le
1
,
[
1
,
1
,
…
,
,
]
ar
e
s
tate
-
ac
tio
n
p
air
s
with
n
s
tate
s
an
d
m
ac
tio
n
s
,
r
esp
ec
tiv
ely
.
(
,
)
is
u
p
d
ated
b
ased
o
n
th
e
B
ellm
an
eq
u
atio
n
[
3
]
in
(
3
)
,
wh
e
r
e
is
th
e
d
is
co
u
n
t
f
ac
t
o
r
with
a
v
alu
e
d
ef
in
ed
f
r
o
m
0
to
1
.
B
ellm
an
eq
u
atio
n
p
r
o
v
id
es
a
r
ec
u
r
s
iv
e
d
ec
o
m
p
o
s
itio
n
o
f
th
e
v
alu
e
f
u
n
ctio
n
,
wh
ic
h
is
ce
n
tr
al
to
s
o
lv
in
g
d
ec
is
io
n
-
m
ak
in
g
p
r
o
b
lem
s
wh
er
e
a
n
ag
en
t
s
ee
k
s
to
m
ax
im
ize
cu
m
u
lativ
e
r
ewa
r
d
s
o
v
er
tim
e.
Ho
wev
er
,
th
e
Q
-
tab
le
is
lim
ited
to
b
ein
g
s
to
r
e
d
in
m
e
m
o
r
y
a
n
d
s
tr
u
ctu
r
e
o
r
g
a
n
izatio
n
,
wh
ich
c
o
u
ld
n
o
t
s
av
e
lar
g
e
r
o
ws
an
d
co
lu
m
n
s
.
T
o
o
v
er
co
m
e
th
is
p
o
i
n
t,
th
e
n
eu
r
al
n
etwo
r
k
is
a
r
ep
lace
m
e
n
t
m
eth
o
d
with
th
e
in
p
u
t la
y
er
(
s
tates)
an
d
o
u
t
p
u
t
(
(
,
)
)
wh
ich
is
ca
lled
d
ee
p
Q
-
n
et
wo
r
k
(
DQN)
as in
Fig
u
r
e
3
.
T
h
is
p
ap
er
u
s
es
th
e
ca
m
er
a
to
d
etec
t
th
e
lin
e
d
ef
in
itio
n
o
n
t
h
e
g
r
o
u
n
d
b
ased
o
n
th
e
n
eu
r
a
l
n
etwo
r
k
ty
p
ed
c
o
n
v
o
lu
tio
n
n
eu
r
al
n
et
wo
r
k
(
C
NN)
.
I
t
is
ap
p
lied
to
f
in
d
th
e
p
o
licy
o
f
R
L
as
in
Fig
u
r
e
3
,
wh
ich
is
well
k
n
o
wn
as
a
class
o
f
d
ee
p
n
e
u
r
al
n
etwo
r
k
s
,
m
o
s
t
co
m
m
o
n
ly
ap
p
lied
to
an
aly
zin
g
v
is
u
a
l
im
ag
er
y
.
I
n
th
e
n
av
ig
atio
n
o
f
m
o
b
ile
r
o
b
o
ts
,
C
NN
co
u
ld
b
e
u
s
ed
to
g
en
e
r
ate
th
e
s
tag
e
to
s
elec
t
ac
to
r
s
o
f
D
R
L
.
T
h
e
d
etails
ar
e
p
r
esen
ted
in
th
e
n
ex
t sectio
n
.
T
ab
le
1
.
R
elatio
n
s
h
ip
b
etwe
en
ac
tio
n
an
d
s
tate
1
2
…
1
(
1
,
1
)
(
2
,
1
)
…
(
,
1
)
2
(
1
,
2
)
(
2
,
2
)
…
(
,
2
)
…
…
…
…
…
(
1
,
)
(
2
,
)
…
(
,
)
Fig
u
r
e
3
.
Neu
r
al
n
etwo
r
k
r
ep
l
ac
in
g
f
o
r
Q
-
T
ab
le
in
R
L
2
.
3
.
Dev
el
o
pin
g
deep
Q
-
net
wo
rk
f
o
r
mo
bil
e
ro
bo
t
I
n
th
is
s
ec
tio
n
,
we
r
ep
r
esen
t
th
e
ar
ch
itectu
r
e
o
f
DQN
wh
ich
co
n
tain
s
3
p
ar
ts
:
p
r
ep
r
o
c
e
s
s
o
r
,
C
N
N
lay
er
,
an
d
ac
tio
n
s
et
as
s
h
o
w
n
in
Fig
u
r
e
4
.
State
o
u
tp
u
t
f
r
o
m
th
e
en
v
ir
o
n
m
en
t
is
th
e
im
ag
e
d
escr
ip
tio
n
with
th
r
ee
p
ar
am
eter
s
th
e
wid
th
,
th
e
h
eig
h
t,
an
d
th
e
co
lo
r
.
I
n
t
h
is
ca
s
e,
th
e
wid
th
an
d
h
eig
h
t
a
r
e
4
8
0
p
ix
els
(
p
x
)
,
an
d
3
6
0
p
ix
els,
r
esp
ec
tiv
ely
.
T
h
e
co
lo
r
is
p
r
esen
te
d
with
t
h
r
ee
u
n
its
b
y
r
ed
-
g
r
ee
n
-
b
lu
e
(
R
GB
)
.
T
o
s
im
p
lify
th
e
n
etwo
r
k
ar
ch
itectu
r
e
o
f
C
NN,
in
s
tead
o
f
u
s
in
g
an
R
GB
im
ag
e
with
f
u
ll
c
o
lo
r
(
3
c
h
an
n
els),
th
e
im
ag
e
will
b
e
co
n
v
e
r
ted
to
th
e
g
r
a
y
im
a
g
e
(
d
ef
in
e
d
with
th
e
s
in
g
le
ch
an
n
el)
wh
er
e
ea
c
h
p
ix
el
is
p
r
esen
ted
b
y
a
s
in
g
le
in
ten
s
ity
v
alu
e
(
f
r
o
m
0
to
2
5
5
)
.
T
h
er
ef
o
r
e,
th
e
im
a
g
e
co
u
ld
b
e
p
r
esen
ted
as
(
4
8
0
×
3
6
0
×
1
)
wh
ich
is
r
esized
to
(
8
4
×
8
4
)
t
o
r
e
d
u
ce
th
e
n
u
m
b
er
o
f
p
ix
els
f
r
o
m
t
h
e
im
ag
e
an
d
p
r
esen
t
with
(
8
4
×8
4
×
1
)
as
in
p
u
t
f
o
r
C
NN.
T
h
is
s
tag
e
is
to
r
e
d
u
ce
t
h
e
co
m
p
u
tatio
n
co
m
p
lex
ity
an
d
t
im
e
tr
ain
in
g
o
f
n
eu
r
al
n
etwo
r
k
s
.
T
h
e
ar
ch
itectu
r
e
o
f
C
NN
co
n
s
is
tin
g
an
in
p
u
t
lay
er
o
f
s
ize
8
4
×8
4
×1
as
in
Fig
u
r
e
5
.
T
h
en
th
r
ee
co
n
v
o
lu
tio
n
lay
er
s
ar
e
3
2
f
ilter
s
o
f
th
e
k
er
n
el
m
atr
ix
8
×8
with
4
s
tr
id
es,
6
4
f
ilter
s
o
f
th
e
k
er
n
el
m
atr
ix
4
×4
with
2
s
tr
id
es,
an
d
6
4
f
ilter
s
o
f
k
er
n
el
m
atr
ix
3
×3
with
o
n
e
s
tr
id
e
ar
e
ap
p
lied
to
f
ilter
th
e
in
f
o
r
m
atio
n
an
d
p
r
o
d
u
ce
a
f
ea
t
u
r
e
m
ap
an
d
th
e
n
th
ey
ar
e
co
n
v
er
ted
to
th
e
s
in
g
le
d
im
en
s
io
n
al
v
ec
to
r
[
2
8
]
.
T
h
e
d
ata
is
s
en
t to
th
e
h
id
d
e
n
lay
er
with
1
2
8
n
eu
r
o
n
s
.
B
o
t
h
c
o
n
v
o
l
u
t
i
o
n
l
a
y
e
r
s
a
n
d
h
id
d
e
n
l
a
y
e
r
s
u
s
e
r
e
c
ti
f
i
e
d
l
i
n
ea
r
u
n
i
t
s
(
R
e
L
U
)
f
o
r
t
h
e
a
c
t
i
v
at
i
o
n
f
u
n
c
t
i
o
n
.
T
h
e
o
u
t
p
u
t
l
a
y
e
r
w
i
t
h
t
h
r
e
e
o
u
t
p
u
t
s
p
r
e
s
e
n
ts
t
h
r
e
e
a
c
t
i
o
n
s
t
u
r
n
l
e
f
t
(
L
0
.
5
i
s
0
.
5
r
a
d
/
s
)
,
f
o
r
w
a
r
d
(
F
0
.
2
i
s
0
.
2
m
/
s
)
,
a
n
d
r
i
g
h
t
(
R
0
.
5
i
s
0
.
5
r
a
d
/s
)
a
s
i
n
(
4
)
.
T
h
e
l
i
n
e
a
r
a
c
ti
v
a
t
i
o
n
f
u
n
c
t
i
o
n
i
s
a
p
p
l
i
e
d
t
o
a
li
g
n
o
u
t
p
u
t
i
n
r
a
n
g
e
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
2
7
4
8
-
2
7
5
7
2752
=
{
0
.
5
(
(
)
)
=
0
0
.
2
(
(
)
)
=
1
0
.
5
(
(
)
)
=
2
(
4
)
Fig
u
r
e
4
.
DQN
Ar
ch
itectu
r
e
f
o
r
m
o
b
ile
r
o
b
o
t
Fig
u
r
e
5
.
C
o
n
v
o
lu
tio
n
n
eu
r
al
n
etwo
r
k
s
tr
u
ctu
r
e
T
o
d
ef
in
e
th
e
r
ewa
r
d
f
o
r
DQN,
an
er
r
o
r
o
f
r
o
b
o
t
s
itu
atio
n
s
is
co
m
p
u
ted
b
y
th
e
ab
s
o
lu
te
p
o
s
itio
n
o
f
s
etp
o
in
t
an
d
f
ee
d
b
ac
k
b
ased
o
n
th
e
X
ax
is
.
Fr
o
m
t
h
e
im
ag
e
o
f
th
e
ca
m
er
a,
two
p
o
in
ts
(
f
ee
d
b
ac
k
p
o
in
t
an
d
s
et
p
o
in
t)
co
u
ld
b
e
d
ete
r
m
in
ed
as
in
Fig
u
r
e
6
.
Her
ein
,
Fig
u
r
e
s
6
(
a
)
an
d
6
(
b
)
ar
e
s
am
p
les
p
r
e
s
en
ted
to
th
e
co
r
n
e
r
with
th
e
r
i
g
h
t
s
id
e
an
d
lef
t
s
id
e,
r
esp
ec
tiv
ely
.
Fig
u
r
e
6
(
c
)
p
r
esen
ts
a
s
tr
aig
h
t
s
itu
atio
n
with
o
u
t
an
y
c
o
r
n
er
s
.
T
h
e
f
ee
d
b
ac
k
p
o
i
n
t
is
d
e
f
in
e
d
as
th
e
ce
n
ter
o
f
th
e
ca
m
er
a
v
iew
with
th
e
m
i
d
d
le
c
r
o
s
s
o
f
wid
th
=4
8
0
a
n
d
h
eig
h
t=3
6
0
,
an
d
th
e
s
et
p
o
in
t
i
s
th
e
ce
n
ter
o
f
th
e
r
ec
tan
g
le
c
o
v
er
in
g
th
e
lin
e
tr
ajec
to
r
y
.
T
h
e
r
o
b
o
t
is
f
o
ll
o
win
g
th
e
lin
e
wh
en
X
feedback
=X
setpoint
,
th
e
er
r
o
r
co
u
l
d
b
e
d
ef
in
ed
as
(
5
)
.
=
|
−
|
(
5
)
T
h
e
er
r
o
r
is
p
r
esen
ted
b
y
th
e
d
is
tan
ce
in
p
ix
els.
T
h
en
,
r
ew
ar
d
s
will
b
e
d
ef
in
ed
b
ased
o
n
th
e
er
r
o
r
v
alu
e
as sh
o
wn
in
T
ab
le
2
,
wh
ich
is
d
iv
id
ed
in
to
f
iv
e
lev
els
-
2
,
-
1
,
-
0
.
5
,
0
.
5
,
an
d
1
.
T
h
ese
r
ewa
r
d
s
ar
e
co
u
n
ted
b
ased
o
n
p
r
e
v
io
u
s
ac
tio
n
s
in
(
4
)
a
n
d
t
h
e
cu
r
r
en
t
s
tate
(
e
r
r
o
r
ev
alu
atio
n
s
)
.
Sp
ec
if
ically
,
th
e
er
r
o
r
is
less
th
an
eq
u
al
to
8
0
p
ix
els
an
d
th
e
m
o
b
ile
r
o
b
o
t
is
i
n
th
e
s
tate
f
o
r
wa
r
d
(
s
tr
aig
h
t)
,
th
is
ac
tio
n
(
s
tr
aig
h
t)
is
en
c
o
u
r
a
g
ed
to
m
ain
tain
.
T
h
er
ef
o
r
e,
t
h
e
r
ewa
r
d
is
a
p
o
s
itiv
e
n
u
m
b
er
ad
d
e
d
to
1
.
Oth
e
r
wis
e,
th
e
MB
is
i
n
lef
t/rig
h
t
co
n
tr
o
l,
an
d
th
ese
ac
tio
n
s
ar
e
n
o
t
e
n
c
o
u
r
ag
e
d
in
th
e
n
ex
t
ac
tio
n
.
T
h
e
r
ewa
r
d
is
s
et
to
a
n
eg
ativ
e
n
u
m
b
er
o
f
-
0
.
5
.
Similar
ly
,
if
th
e
er
r
o
r
is
in
th
e
r
an
g
e
o
f
8
0
p
ix
els to
1
2
0
p
i
x
el
s
an
d
all
ac
tio
n
s
in
th
e
p
r
ev
io
u
s
co
n
f
ig
u
r
ed
.
T
h
is
ca
s
e
is
g
en
er
ally
en
c
o
u
r
a
g
ed
,
s
o
th
e
r
ewa
r
d
is
th
e
n
o
r
m
al
v
a
lu
e
ad
d
e
d
to
0
.
5
.
Fin
ally
,
th
e
m
o
b
ile
r
o
b
o
t
s
h
o
u
ld
m
ain
tain
th
e
s
m
allest
er
r
o
r
,
an
d
to
av
o
id
th
e
wo
r
s
t
ca
s
e
wh
ic
h
s
ets
th
e
r
ewa
r
d
to
-
2
if
th
e
e
r
r
o
r
>
220
.
At
th
e
in
itial
tim
e,
th
e
DQN
is
em
p
ty
,
an
d
it
n
ee
d
s
th
e
tr
ain
in
g
p
r
o
ce
s
s
f
o
r
s
tu
d
y
i
n
g
an
d
u
p
d
atin
g
th
e
weig
h
t
to
b
ec
o
m
e
m
o
r
e
in
tellig
en
t.
Ne
x
t p
ar
t,
th
e
tr
ai
n
in
g
p
r
o
ce
s
s
will b
e
co
n
s
id
er
e
d
in
d
etail.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
I
n
d
o
o
r
n
a
vig
a
tio
n
fo
r
mo
b
ile
r
o
b
o
ts
b
a
s
ed
o
n
d
ee
p
r
ein
fo
r
ce
men
t le
a
r
n
in
g
…
(
K
h
o
a
N
g
u
ye
n
Da
n
g
)
2753
(
a)
(
b
)
(
c)
Fig
u
r
e
6
.
C
o
m
p
u
tin
g
e
r
r
o
r
m
e
th
o
d
: (
a)
v
iew
o
f
c
o
r
n
e
r
with
t
h
e
r
ig
h
t sid
e,
(
b
)
v
iew
o
f
c
o
r
n
er
with
th
e
lef
t
s
id
e
,
an
d
(
c)
v
iew
o
f
n
o
c
o
r
n
er
with
th
e
s
tr
aig
h
t sit
u
atio
n
T
ab
le
2
.
R
ewa
r
d
d
e
f
in
itio
n
s
C
o
n
d
i
t
i
o
n
s
R
e
w
a
r
d
v
a
l
u
e
D
e
scri
p
t
i
o
n
s
A
c
t
i
o
n
s
Er
r
o
r
s (p
i
x
e
l
s)
F
0
.
2
≤
80
1
En
c
o
u
r
a
g
e
g
o
st
r
a
i
g
h
t
L0
.
5
,
R
0
.
5
-
0
.
5
N
o
t
e
n
c
o
u
r
a
g
e
A
n
y
a
c
t
i
o
n
s
80
<
≤
120
0
.
5
G
e
n
e
r
a
l
e
n
c
o
u
r
a
g
e
F
0
.
2
120
<
≤
220
-
1
N
o
t
e
n
c
o
u
r
a
g
e
L0
.
5
,
R
0
.
5
0
.
5
G
e
n
e
r
a
l
e
n
c
o
u
r
a
g
e
me
n
t
(
t
u
r
n
l
e
f
t
/
r
i
g
h
t
)
A
n
y
a
c
t
i
o
n
s
>
220
-
2
W
o
r
st
c
a
s
e
T
h
e
tr
ain
in
g
p
r
o
ce
s
s
o
f
th
e
DQN
alg
o
r
ith
m
is
s
h
o
wn
in
Fig
u
r
e
7
.
Her
ein
,
t
h
e
p
r
e
d
ict
an
d
tar
g
et
m
o
d
el
(
PM,
T
M)
ar
e
p
r
esen
te
d
as
th
e
C
NN
ty
p
e,
wh
ich
is
u
s
ed
to
p
r
ed
ict
th
e
ac
tio
n
v
alu
e
s
an
d
ca
lcu
late
th
e
v
alu
e
f
o
r
th
e
n
e
x
t
s
tate
in
B
e
llm
an
(
7
)
,
r
esp
ec
tiv
ely
.
I
n
itially
,
th
e
a
g
en
t
r
ec
eiv
es
th
e
p
ar
am
eter
s
,
,
an
d
,
f
r
o
m
th
e
en
v
ir
o
n
m
en
t
to
ch
o
o
s
e
th
e
ac
tio
n
b
ased
o
n
(
7
)
wh
er
e
(
)
is
th
e
o
u
tp
u
t
o
f
th
e
PM
n
etwo
r
k
.
Af
ter
th
at,
th
e
en
v
ir
o
n
m
e
n
t
will
ex
p
o
r
t
th
e
n
ew
+
1
,
+
1
,
an
d
c
o
n
tin
u
e
a
n
ew
lo
o
p
.
A
b
u
f
f
er
m
em
o
r
y
M
t
is
u
s
ed
to
s
to
r
e
all
d
ata
=
[
,
,
+
1
,
+
1
]
f
o
r
th
e
tr
ain
in
g
o
f
th
e
C
NN
m
o
d
el.
W
h
en
M
t
s
ize
co
m
es
to
a
m
ax
im
u
m
v
alu
e
d
ef
in
itio
n
,
th
e
lo
s
t
f
u
n
c
tio
n
is
ex
ec
u
ted
an
d
is
u
s
ed
to
u
p
d
ate
th
e
weig
h
t
o
f
PM
Af
ter
th
at,
th
e
M
t
is
r
ef
r
esh
ed
f
o
r
n
ew
s
tep
s
to
r
in
g
.
T
h
is
s
tep
en
s
u
r
es
th
e
ef
f
icie
n
cy
o
f
th
e
tr
ai
n
in
g
p
r
o
ce
s
s
with
th
e
en
o
u
g
h
lar
g
e
d
ataset.
T
h
e
s
ize
o
f
M
t
co
u
ld
b
e
d
ef
i
n
ed
as
to
o
lar
g
e
s
o
it
co
u
ld
a
f
f
ec
t
to
co
m
p
u
te
p
r
o
c
ess
an
d
tak
e
lo
n
g
er
tim
e.
T
h
er
ef
o
r
e,
a
b
atch
_
s
ize
i
s
u
s
ed
to
ex
tr
ac
t th
e
n
u
m
b
e
r
o
f
r
an
d
o
m
d
ata
f
r
o
m
M
t
f
o
r
th
e
tr
ain
in
g
p
r
o
ce
s
s
.
T
h
e
weig
h
t u
p
d
atin
g
o
f
PM
b
ased
o
n
th
e
lo
s
s
f
u
n
ctio
n
co
u
ld
b
e
co
m
p
u
ted
b
y
th
e
m
ea
n
s
q
u
ar
ed
er
r
o
r
f
o
llo
win
g
in
(
6
)
.
W
h
er
e
(
+
1
)
is
t
h
e
o
u
tp
u
t
v
alu
es
o
f
th
e
T
M
n
etwo
r
k
at
+
1
an
d
(
,
)
is
th
e
o
u
tp
u
t
o
f
PM
n
etwo
r
k
at
an
d
.
T
h
e
PM
weig
h
ts
ar
e
u
p
d
a
ted
d
u
r
in
g
n
iter
atio
n
an
d
th
e
y
ar
e
co
p
ied
to
u
p
d
ate
to
th
e
weig
h
ts
o
f
T
M.
B
ased
o
n
MSE
,
th
e
tr
ain
in
g
p
r
o
ce
s
s
u
s
es th
e
Ad
am
o
p
tim
iza
tio
n
with
a
lear
n
in
g
r
ate
s
elec
tio
n
[
2
9
]
.
Fig
u
r
e
7
.
T
r
ain
in
g
p
r
o
ce
s
s
f
o
r
C
NN
m
o
d
el
in
ag
en
t
=
1
2
[
(
,
)
−
(
,
)
]
2
wh
er
e
(
,
)
=
+
ma
x
(
(
+
1
)
)
(
6
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
2
7
4
8
-
2
7
5
7
2754
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
I
n
th
is
p
ap
er
,
we
u
s
e
a
Gaz
eb
o
s
im
u
latio
n
en
v
ir
o
n
m
en
t
[
3
0
]
to
s
im
u
late
th
e
f
o
u
r
-
wh
ee
l
m
o
b
ile
r
o
b
o
t
tr
ac
k
in
g
a
lin
e
u
s
in
g
th
e
v
is
io
n
(
ca
m
e
r
a
p
l
u
g
in
in
Gaz
e
b
o
)
as
s
h
o
wn
in
Fig
u
r
e
8
,
an
d
Py
t
h
o
n
p
r
o
g
r
am
m
i
n
g
.
T
h
e
p
ar
a
m
eter
s
o
f
t
h
e
m
o
b
ile
r
o
b
o
t
[
3
1
]
a
n
d
DQN
alg
o
r
i
th
m
[
3
2
]
ar
e
p
r
esen
ted
in
T
a
b
le
3
.
Fo
r
tr
ain
in
g
p
r
o
ce
s
s
in
g
,
th
e
n
u
m
b
er
o
f
t
r
ials
is
s
e
lecte
d
4
0
.
0
0
0
tim
es
ca
lled
ep
is
o
d
e_
s
tep
b
y
e
x
p
er
im
en
tal.
E
ac
h
ep
is
o
d
e_
s
tep
will
h
a
v
e
a
s
tar
t tim
e
(
m
o
b
ile
r
o
b
o
t
g
o
)
a
n
d
r
eset
(
u
n
til
wo
r
s
e
ca
s
e)
d
ef
in
e
d
a
s
n
_
s
tep
tim
e.
As a
r
esu
lt
in
Fig
u
r
e
9
(
a
)
,
th
e
FW
MR
m
o
v
es
with
a
s
h
o
r
t
tim
e
(
s
m
all
n
_
s
tep
,
s
m
all
r
ewa
r
d
)
a
n
d
a
s
h
o
r
t
d
is
tan
ce
in
th
e
f
ir
s
t
ep
is
o
d
e_
s
tep
b
ec
a
u
s
e
all
p
ar
am
eter
s
an
d
weig
h
t
ar
e
in
itial.
T
h
ey
n
ee
d
m
o
r
e
tim
e
f
o
r
tr
ain
i
n
g
an
d
u
p
d
atin
g
an
d
th
en
th
e
n
_
s
tep
an
d
r
ewa
r
d
co
u
l
d
b
e
b
etter
in
ep
is
o
d
e_
s
tep
n
u
m
b
e
r
5
7
as
s
h
o
wn
in
Fig
u
r
e
9
(
b
)
,
wh
ich
also
p
r
esen
ts
th
e
av
er
a
g
e
r
ewa
r
d
v
al
u
e.
Af
ter
th
e
tr
ain
in
g
p
r
o
ce
s
s
,
th
e
test
in
g
p
r
o
ce
s
s
to
p
er
f
o
r
m
e
d
o
n
a
m
o
b
ile
r
o
b
o
t tr
ac
k
in
g
th
e
lin
e
wh
ich
b
u
ild
s
in
Gaz
eb
o
s
im
u
latio
n
.
T
h
e
r
esu
lt
in
Fig
u
r
e
1
0
s
h
o
ws
th
e
p
er
f
o
r
m
a
n
ce
s
o
f
FW
MR,
in
wh
ich
th
e
er
r
o
r
s
ar
e
co
m
p
u
ted
with
n
o
ab
s
o
lu
t
e
in
(
9
)
an
d
s
h
o
wn
in
Fig
u
r
es
1
0
(
a)
an
d
1
0
(
b
)
.
T
h
e
MR
tr
ac
k
in
g
p
er
f
o
r
m
an
c
e
is
s
h
o
wn
in
Fig
u
r
e
1
0
(
c)
.
Ov
er
al
l,
th
e
er
r
o
r
is
in
th
e
r
a
n
g
e
o
f
-
2
0
0
p
x
to
2
0
0
p
x
,
an
d
m
o
s
t
o
f
t
h
e
f
o
c
u
s
is
o
n
-
1
0
0
to
1
0
0
p
x
.
T
h
is
m
ea
n
s
th
at
th
e
er
r
o
r
d
e
f
in
ed
th
e
d
ev
iatio
n
b
etwe
en
th
e
v
is
io
n
v
iew
(
r
o
b
o
t
v
iew)
an
d
th
e
lin
e,
wh
ich
is
s
m
all
(
1
0
0
p
x
is
ar
o
u
n
d
2
.
6
cm
)
an
d
s
u
itab
le
f
o
r
m
ap
p
in
g
in
ea
c
h
d
ir
ec
tio
n
,
s
u
ch
as
f
o
r
war
d
,
lef
t,
an
d
r
ig
h
t.
T
h
e
e
r
r
o
r
is
less
th
an
2
4
0
,
an
d
th
at
m
o
b
ile
r
o
b
o
t
is
in
th
e
h
an
d
le,
n
o
t
in
th
e
w
o
r
s
t
ca
s
e
wh
ich
h
as
th
e
r
ewa
r
d
s
et
b
y
-
2
.
W
ith
th
e
ca
s
e
er
r
o
r
s
to
ar
o
u
n
d
±
2
2
0
p
x
,
th
e
m
o
b
ile
r
o
b
o
t
g
o
es
ar
o
u
n
d
th
e
co
r
n
er
in
th
e
m
ap
.
B
u
t it
m
ay
n
o
t o
f
ten
b
ec
au
s
e
it d
ep
en
d
s
o
n
th
e
p
r
ev
i
o
u
s
d
ir
ec
tio
n
an
d
s
tate
o
f
MR.
T
h
e
co
n
tr
o
l te
n
d
en
c
y
is
to
th
e
er
r
o
r
r
ed
u
ce
d
to
ze
r
o
.
So
th
e
er
r
o
r
s
ex
is
t
with
±
2
2
0
p
x
o
n
ly
m
ain
tain
e
d
d
u
r
in
g
a
s
h
o
r
t
tim
e.
B
esid
es,
m
ea
n
s
q
u
ar
ed
e
r
r
o
r
(
MSE
)
is
co
m
p
u
ted
a
n
d
co
m
p
ar
ed
b
etwe
en
th
e
p
r
o
p
o
s
al
m
eth
o
d
a
n
d
t
h
e
tr
ad
itio
n
al
v
is
io
n
m
eth
o
d
[
3
3
]
(
u
s
in
g
im
ag
e
m
o
m
en
t)
as
s
h
o
wn
in
T
a
b
le
4
.
B
ased
o
n
th
e
er
r
o
r
c
o
n
tr
o
l
in
p
i
x
els,
th
e
co
n
v
er
t
to
th
e
m
eter
is
u
s
ed
an
d
co
m
p
u
t
ed
MSE
(
1
m
eter
=
3
7
7
9
.
5
2
p
ix
els).
T
h
e
MSE
in
th
e
DR
L
m
eth
o
d
is
lo
wer
in
VSN.
I
t p
r
o
v
es th
at
th
e
DR
L
m
eth
o
d
h
as h
ig
h
er
p
er
f
o
r
m
an
c
e
th
an
th
e
VSN
m
eth
o
d
.
Fig
u
r
e
8
.
Fo
u
r
-
wh
ee
l m
o
b
ile
r
o
b
o
t f
o
llo
ws lin
e
u
s
in
g
ca
m
er
a
s
ce
n
ar
io
T
ab
le
3
.
Par
am
eter
o
f
m
o
b
ile
r
o
b
o
t
an
d
DQN
N
a
me
S
i
z
e
_
c
a
mera
r
c
B
a
t
c
h
_
s
i
z
e
Ep
i
s
o
d
e
_
st
e
p
Le
a
r
n
i
n
g
r
a
t
e
V
a
l
u
e
0
.
9
9
1
.
0
4
8
0
×
3
6
0
×
3
2
(
c
m
)
5
(
c
m
)
32
4
0
0
0
0
0
.
0
0
0
1
(
a)
(
b
)
Fig
u
r
e
9
.
T
r
ain
in
g
p
r
o
ce
s
s
r
esu
lts
(
a)
n
_
s
tep
in
o
n
e
e
p
is
o
d
es_
s
tep
an
d
(
b
)
r
ewa
r
d
an
d
av
er
ag
e
r
ewa
r
d
i
n
o
n
e
ep
is
o
d
es_
s
tep
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
I
n
d
o
o
r
n
a
vig
a
tio
n
fo
r
mo
b
ile
r
o
b
o
ts
b
a
s
ed
o
n
d
ee
p
r
ein
fo
r
ce
men
t le
a
r
n
in
g
…
(
K
h
o
a
N
g
u
ye
n
Da
n
g
)
2755
T
ab
le
4
.
MSE
co
m
p
ar
is
o
n
b
etwe
en
p
r
o
p
o
s
al
m
eth
o
d
an
d
tr
a
d
itio
n
al
v
is
io
n
m
eth
o
d
M
e
t
h
o
d
M
S
E
*
1
0
0
(
i
n
me
t
e
r
)
P
r
o
p
o
sa
l
m
e
t
h
o
d
(
D
R
L)
0
.
0
2
6
4
Tr
a
d
i
t
i
o
n
a
l
v
i
s
i
o
n
met
h
o
d
(
V
S
N
)
0
.
0
4
0
6
(
a)
(
b
)
(
c)
Fig
u
r
e
1
0
.
Per
f
o
r
m
a
n
ce
o
f
FW
MR (
a)
er
r
o
r
o
f
p
o
in
t in
im
a
g
e
v
iewin
g
f
r
o
m
t
h
e
r
o
b
o
t in
t
h
e
p
r
o
p
o
s
al
m
eth
o
d
,
(
b
)
er
r
o
r
o
f
p
o
in
t in
im
a
g
e
v
ie
win
g
f
r
o
m
th
e
r
o
b
o
t in
th
e
tr
a
d
itio
n
al
v
is
io
n
m
eth
o
d
,
a
n
d
(
c
)
tr
ac
k
in
g
lin
e
p
er
f
o
r
m
an
ce
o
f
FW
MR
T
h
e
p
er
f
o
r
m
an
ce
o
f
m
o
b
ile
r
o
b
o
t
tr
ac
k
i
n
g
is
s
h
o
wn
in
th
e
d
etails
o
f
Fig
u
r
e
1
0
(
c
)
th
at
FW
MR
co
u
ld
co
m
p
lete
a
tr
ajec
to
r
y
f
r
o
m
th
e
s
tar
t
p
o
in
t
to
th
e
s
to
p
p
o
in
t
in
th
e
m
ap
.
E
asil
y
s
ee
m
s
th
at
th
e
MR
h
as
a
b
ig
g
er
er
r
o
r
in
ea
c
h
co
r
n
er
.
B
u
t
it
is
s
till
b
etter
th
an
th
e
VSN
m
eth
o
d
.
Ho
wev
er
,
in
th
e
s
tr
aig
h
t
li
n
e,
MR
co
u
ld
tr
ac
k
th
e
tr
ajec
to
r
y
with
h
ig
h
er
p
e
r
f
o
r
m
an
ce
(
o
v
er
lap
lin
e,
clo
s
ed
d
is
tan
ce
with
lin
e)
.
T
h
u
s
,
th
e
MR
u
s
es
th
e
C
N
N
b
ased
o
n
th
e
im
ag
e
f
r
o
m
th
e
ca
m
er
a
ca
n
g
en
e
r
ate
th
e
ac
tio
n
f
o
r
ea
ch
s
tate
g
iv
en
b
y
th
e
en
v
ir
o
n
m
en
t
in
th
e
DR
L
s
tr
u
ctu
r
e.
T
h
e
tar
g
et
o
f
in
d
o
o
r
n
av
ig
atio
n
o
f
MR
is
estab
lis
h
ed
s
u
cc
es
s
in
co
m
p
lex
m
ap
p
in
g
an
d
m
ain
tain
in
g
it
in
a
r
ea
l
-
tim
e
s
y
s
tem
.
T
h
is
p
r
o
v
es
o
u
r
p
r
o
p
o
s
al
is
u
s
ef
u
l
an
d
b
r
in
g
s
h
ig
h
e
f
f
ec
tiv
en
ess
f
o
r
th
e
n
av
ig
atio
n
a
n
d
tr
ac
k
in
g
lin
e
o
f
a
m
o
b
ile
r
o
b
o
t.
4.
CO
NCLU
SI
O
N
T
h
is
p
ap
er
p
r
esen
ts
th
e
DQN
in
DR
L
f
o
r
th
e
n
av
ig
atio
n
o
f
a
m
o
b
ile
r
o
b
o
t
f
o
llo
win
g
th
e
l
in
e
o
n
th
e
g
r
o
u
n
d
b
ased
o
n
a
co
n
v
o
lu
tio
n
n
e
u
r
al
n
etwo
r
k
f
o
r
p
r
o
ce
s
s
in
g
im
a
g
es
f
r
o
m
a
ca
m
er
a.
All
s
tep
s
ar
e
d
is
cu
s
s
ed
in
d
etail
to
clea
r
p
er
f
o
r
m
an
ce
an
d
ap
p
ly
to
th
e
s
p
ec
if
ic
o
b
j
ec
t.
T
h
e
r
esu
lts
im
p
r
o
v
ed
t
h
e
ef
f
ec
tiv
en
ess
o
f
th
e
co
n
tr
o
l
alg
o
r
it
h
m
s
o
th
at
th
e
m
o
b
ile
r
o
b
o
t
co
u
ld
f
o
llo
w
th
e
lin
e
with
h
ig
h
ac
cu
r
ac
y
p
r
ese
n
ted
b
y
th
e
er
r
o
r
o
f
less
th
an
1
0
0
p
ix
els
with
d
if
f
er
en
t
t
y
p
es
o
f
tr
ajec
to
r
ies.
T
h
is
p
r
o
p
o
s
al
co
u
ld
b
e
ap
p
li
ed
to
m
o
b
ile
r
o
b
o
t
n
av
ig
atio
n
as we
ll a
s
in
r
o
b
o
ti
cs f
ield
s
in
th
e
f
u
tu
r
e
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Au
th
o
r
s
s
tate
n
o
f
u
n
d
in
g
in
v
o
lv
ed
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
2
7
4
8
-
2
7
5
7
2756
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Kh
o
a
Ng
u
y
e
n
Dan
g
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Van
T
r
an
T
h
i
✓
✓
✓
✓
✓
✓
✓
✓
Ng
u
y
en
Van
T
h
an
g
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
DATA
AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
t
h
at
s
u
p
p
o
r
t
th
e
f
in
d
in
g
s
o
f
t
h
is
s
tu
d
y
ar
e
av
ailab
l
e
f
r
o
m
th
e
c
o
r
r
esp
o
n
d
in
g
au
t
h
o
r
,
K
N
D
,
u
p
o
n
r
ea
s
o
n
ab
le
r
eq
u
est.
R
E
FE
R
E
N
C
E
S
[
1
]
A
.
M
a
r
k
i
s
,
M
.
P
a
p
a
,
D
.
K
a
se
l
a
u
t
z
k
e
,
M
.
R
a
t
h
ma
i
r
,
V
.
S
a
t
t
i
n
g
e
r
,
a
n
d
M
.
B
r
a
n
d
s
t
ö
t
t
e
r
,
“
S
a
f
e
t
y
o
f
m
o
b
i
l
e
r
o
b
o
t
s
y
st
e
ms
i
n
i
n
d
u
st
r
i
a
l
a
p
p
l
i
c
a
t
i
o
n
s
,
”
ARW
& O
AG
M
W
o
rks
h
o
p
2
0
1
9
,
p
p
.
2
6
–
3
1
,
2
0
1
9
.
[
2
]
A
.
Lo
g
a
n
a
t
h
a
n
a
n
d
N
.
S
.
A
h
ma
d
,
“
A
sy
st
e
ma
t
i
c
r
e
v
i
e
w
o
n
r
e
c
e
n
t
a
d
v
a
n
c
e
s
i
n
a
u
t
o
n
o
m
o
u
s
m
o
b
i
l
e
r
o
b
o
t
n
a
v
i
g
a
t
i
o
n
,
”
En
g
i
n
e
e
ri
n
g
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
a
n
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
,
v
o
l
.
4
0
,
p
.
1
0
1
3
4
3
,
A
p
r
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
e
st
c
h
.
2
0
2
3
.
1
0
1
3
4
3
.
[
3
]
C
.
H
.
A
.
H
.
B
.
B
a
sk
o
r
o
,
H
.
M
.
S
a
p
u
t
r
a
,
M
.
M
i
r
d
a
n
i
e
s
,
V
.
S
u
s
a
n
t
i
,
M
.
F
.
R
a
d
z
i
,
a
n
d
R
.
I
.
A
.
A
z
i
z
,
“
A
n
a
u
t
o
n
o
m
o
u
s
m
o
b
i
l
e
r
o
b
o
t
p
l
a
t
f
o
r
m
f
o
r
m
e
d
i
c
a
l
p
u
r
p
o
se,
”
i
n
Pro
c
e
e
d
i
n
g
-
2
0
2
0
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
S
u
s
t
a
i
n
a
b
l
e
E
n
e
r
g
y
En
g
i
n
e
e
ri
n
g
a
n
d
Ap
p
l
i
c
a
t
i
o
n
:
S
u
st
a
i
n
a
b
l
e
En
e
r
g
y
a
n
d
T
ra
n
s
p
o
rt
a
t
i
o
n
:
T
o
w
a
r
d
s
Al
l
-
Re
n
e
w
a
b
l
e
F
u
t
u
r
e
,
I
C
S
EEA
2
0
2
0
,
N
o
v
.
2
0
2
0
,
p
p
.
4
1
–
4
4
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
S
EEA
5
0
7
1
1
.
2
0
2
0
.
9
3
0
6
1
6
1
.
[
4
]
Z.
G
o
n
g
,
Z.
N
i
e
,
Q
.
Li
u
,
a
n
d
X
.
J.
L
i
u
,
“
D
e
si
g
n
a
n
d
c
o
n
t
r
o
l
o
f
a
m
u
l
t
i
-
m
o
b
i
l
e
-
r
o
b
o
t
c
o
o
p
e
r
a
t
i
v
e
t
r
a
n
s
p
o
r
t
sy
st
e
m b
a
s
e
d
o
n
a
n
o
v
e
l
si
x
d
e
g
r
e
e
-
of
-
f
r
e
e
d
o
m
c
o
n
n
e
c
t
o
r
,
”
I
S
A
T
ra
n
s
a
c
t
i
o
n
s
,
v
o
l
.
1
3
9
,
p
p
.
6
0
6
–
6
2
0
,
A
u
g
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
s
a
t
r
a
.
2
0
2
3
.
0
4
.
0
0
6
.
[
5
]
B
.
S
.
O
.
P
a
l
l
a
r
e
s
,
T
.
A
.
M
.
R
o
z
o
,
E
.
C
.
C
a
m
a
c
h
o
,
J
.
G
.
G
u
a
r
n
i
z
o
,
a
n
d
J
.
M
.
C
a
l
d
e
r
o
n
,
“
D
e
s
i
g
n
a
n
d
c
o
n
s
t
r
u
c
t
i
o
n
o
f
a
c
o
s
t
-
o
r
i
e
n
t
e
d
m
o
b
i
l
e
r
o
b
o
t
f
o
r
d
o
m
e
s
t
i
c
a
s
s
i
s
t
a
n
c
e
,
”
I
F
A
C
-
P
a
p
e
r
s
O
n
L
i
n
e
,
v
o
l
.
5
4
,
n
o
.
1
3
,
p
p
.
2
9
3
–
2
9
8
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
f
a
c
o
l
.
2
0
2
1
.
1
0
.
4
6
2
.
[
6
]
S
.
H
a
b
i
b
i
a
n
e
t
a
l
.
,
“
D
e
si
g
n
a
n
d
i
m
p
l
e
men
t
a
t
i
o
n
o
f
a
ma
x
i
-
s
i
z
e
d
mo
b
i
l
e
r
o
b
o
t
(
K
a
r
o
)
f
o
r
r
e
s
c
u
e
m
i
ssi
o
n
s,”
RO
B
O
ME
C
H
J
o
u
r
n
a
l
,
v
o
l
.
8
,
n
o
.
1
,
J
a
n
.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
8
6
/
s
4
0
6
4
8
-
0
2
0
-
0
0
1
8
8
-
9.
[
7
]
H
.
A
z
p
ú
r
u
a
e
t
a
l
.
,
“
A
s
u
r
v
e
y
o
n
t
h
e
a
u
t
o
n
o
mo
u
s
e
x
p
l
o
r
a
t
i
o
n
o
f
c
o
n
f
i
n
e
d
su
b
t
e
r
r
a
n
e
a
n
sp
a
c
e
s:
p
e
r
sp
e
c
t
i
v
e
s
f
r
o
m
r
e
a
l
-
w
o
r
l
d
a
n
d
i
n
d
u
st
r
i
a
l
r
o
b
o
t
i
c
d
e
p
l
o
y
m
e
n
t
s,
”
Ro
b
o
t
i
c
s
a
n
d
Au
t
o
n
o
m
o
u
s
S
y
st
e
m
s
,
v
o
l
.
1
6
0
,
p
.
1
0
4
3
0
4
,
F
e
b
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
r
o
b
o
t
.
2
0
2
2
.
1
0
4
3
0
4
.
[
8
]
V
.
B
u
l
u
t
,
“
P
a
t
h
p
l
a
n
n
i
n
g
o
f
m
o
b
i
l
e
r
o
b
o
t
s
i
n
d
y
n
a
mi
c
e
n
v
i
r
o
n
m
e
n
t
b
a
se
d
o
n
a
n
a
l
y
t
i
c
g
e
o
m
e
t
r
y
a
n
d
c
u
b
i
c
B
é
z
i
e
r
c
u
r
v
e
w
i
t
h
t
h
r
e
e
sh
a
p
e
p
a
r
a
me
t
e
r
s,
”
E
x
p
e
rt
S
y
st
e
m
s
w
i
t
h
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
3
3
,
p
.
1
2
0
9
4
2
,
D
e
c
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
3
.
1
2
0
9
4
2
.
[
9
]
P
.
S
.
Y
a
d
a
v
,
V
.
A
g
r
a
w
a
l
,
J.
C
.
M
o
h
a
n
t
a
,
a
n
d
M
.
D
.
F
a
i
y
a
z
A
h
me
d
,
“
A
r
o
b
u
s
t
sl
i
d
i
n
g
m
o
d
e
c
o
n
t
r
o
l
o
f
me
c
a
n
u
m
w
h
e
e
l
-
c
h
a
i
r
f
o
r
t
r
a
j
e
c
t
o
r
y
t
r
a
c
k
i
n
g
,
”
M
a
t
e
ri
a
l
s
T
o
d
a
y
:
Pro
c
e
e
d
i
n
g
s
,
v
o
l
.
5
6
,
p
p
.
6
2
3
–
6
3
0
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
ma
t
p
r
.
2
0
2
1
.
1
2
.
3
9
8
.
[
1
0
]
D
.
U
.
R
i
j
a
l
u
sa
l
a
m
a
n
d
I
.
I
sw
a
n
t
o
,
“
I
mp
l
e
me
n
t
a
t
i
o
n
k
i
n
e
ma
t
i
c
s
mo
d
e
l
i
n
g
a
n
d
o
d
o
me
t
r
y
o
f
f
o
u
r
o
m
n
i
w
h
e
e
l
m
o
b
i
l
e
r
o
b
o
t
o
n
t
h
e
t
r
a
j
e
c
t
o
r
y
p
l
a
n
n
i
n
g
a
n
d
mo
t
i
o
n
c
o
n
t
r
o
l
b
a
s
e
d
mi
c
r
o
c
o
n
t
r
o
l
l
e
r
,
”
J
o
u
r
n
a
l
o
f
R
o
b
o
t
i
c
s
a
n
d
C
o
n
t
r
o
l
(
J
R
C
)
,
v
o
l
.
2
,
n
o
.
5
,
p
p
.
4
4
8
–
4
5
5
,
2
0
2
1
,
d
o
i
:
1
0
.
1
8
1
9
6
/
j
r
c
.
2
5
1
2
1
.
[
1
1
]
Z.
H
a
n
,
J.
Lo
n
g
,
W
.
W
a
n
g
,
a
n
d
L
.
W
a
n
g
,
“
A
d
a
p
t
i
v
e
t
r
a
c
k
i
n
g
c
o
n
t
r
o
l
o
f
t
w
o
-
w
h
e
e
l
e
d
mo
b
i
l
e
r
o
b
o
t
s
u
n
d
e
r
D
e
n
i
a
l
-
of
-
S
e
r
v
i
c
e
a
t
t
a
c
k
s,
”
I
S
A
T
r
a
n
s
a
c
t
i
o
n
s
,
v
o
l
.
1
4
1
,
p
p
.
3
6
5
–
3
7
6
,
O
c
t
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
s
a
t
r
a
.
2
0
2
3
.
0
6
.
0
2
2
.
[
1
2
]
J
.
S
.
L
i
n
g
L
e
o
n
g
,
K
.
T
.
K
i
n
T
e
o
,
a
n
d
H
.
P
.
Y
o
o
n
g
,
“
F
o
u
r
w
h
e
e
l
e
d
m
o
b
i
l
e
r
o
b
o
t
s
:
a
r
e
v
i
e
w
,
”
i
n
4
t
h
I
E
E
E
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
i
n
E
n
g
i
n
e
e
r
i
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
I
I
C
A
I
E
T
2
0
2
2
,
S
e
p
.
2
0
2
2
,
p
p
.
1
–
6
.
d
o
i
:
1
0
.
1
1
0
9
/
I
I
C
A
I
E
T
5
5
1
3
9
.
2
0
2
2
.
9
9
3
6
8
5
5
.
[
1
3
]
M
.
H
.
M
o
h
a
ma
d
N
o
r
,
Z.
H
.
I
smai
l
,
a
n
d
M
.
A
.
A
h
ma
d
,
“
B
r
o
a
d
c
a
s
t
c
o
n
t
r
o
l
o
f
m
u
l
t
i
-
r
o
b
o
t
s
y
st
e
ms
w
i
t
h
n
o
r
m
-
l
i
m
i
t
e
d
u
p
d
a
t
e
v
e
c
t
o
r
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Ad
v
a
n
c
e
d
R
o
b
o
t
i
c
S
y
s
t
e
m
s
,
v
o
l
.
1
7
,
n
o
.
4
,
Ju
l
.
2
0
2
0
,
d
o
i
:
1
0
.
1
1
7
7
/
1
7
2
9
8
8
1
4
2
0
9
4
5
9
5
8
.
[
1
4
]
A
.
S
t
e
f
e
k
,
T.
v
a
n
P
h
a
m,
V
.
K
r
i
v
a
n
e
k
,
a
n
d
K
.
L
.
P
h
a
m,
“
E
n
e
r
g
y
c
o
m
p
a
r
i
s
o
n
o
f
c
o
n
t
r
o
l
l
e
r
s
u
s
e
d
f
o
r
a
d
i
f
f
e
r
e
n
t
i
a
l
d
r
i
v
e
w
h
e
e
l
e
d
mo
b
i
l
e
r
o
b
o
t
,
”
I
EEE
Ac
c
e
ss
,
v
o
l
.
8
,
p
p
.
1
7
0
9
1
5
–
1
7
0
9
2
7
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
0
.
3
0
2
3
3
4
5
.
[
1
5
]
X
.
Lu
o
,
D
.
M
u
,
Z.
W
a
n
g
,
P
.
N
i
n
g
,
a
n
d
C
.
H
u
a
,
“
A
d
a
p
t
i
v
e
f
u
l
l
-
s
t
a
t
e
c
o
n
st
r
a
i
n
e
d
t
r
a
c
k
i
n
g
c
o
n
t
r
o
l
f
o
r
m
o
b
i
l
e
r
o
b
o
t
i
c
sy
st
e
m
w
i
t
h
u
n
k
n
o
w
n
d
e
a
d
-
z
o
n
e
i
n
p
u
t
,
”
N
e
u
r
o
c
o
m
p
u
t
i
n
g
,
v
o
l
.
5
2
4
,
p
p
.
3
1
–
4
2
,
M
a
r
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
n
e
u
c
o
m.
2
0
2
2
.
1
2
.
0
2
5
.
[
1
6
]
J.
L
i
,
J.
W
a
n
g
,
H
.
P
e
n
g
,
Y
.
H
u
,
a
n
d
H
.
S
u
,
“
F
u
z
z
y
-
t
o
r
q
u
e
a
p
p
r
o
x
i
ma
t
i
o
n
-
e
n
h
a
n
c
e
d
s
l
i
d
i
n
g
mo
d
e
c
o
n
t
r
o
l
f
o
r
l
a
t
e
r
a
l
s
t
a
b
i
l
i
t
y
o
f
mo
b
i
l
e
r
o
b
o
t
,
”
I
EEE
T
ra
n
s
a
c
t
i
o
n
s
o
n
S
y
s
t
e
m
s,
M
a
n
,
a
n
d
C
y
b
e
rn
e
t
i
c
s:
S
y
st
e
m
s
,
v
o
l
.
5
2
,
n
o
.
4
,
p
p
.
2
4
9
1
–
2
5
0
0
,
A
p
r
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
TS
M
C
.
2
0
2
1
.
3
0
5
0
6
1
6
.
[
1
7
]
Y
.
Li
u
,
S
.
W
a
n
g
,
Y
.
X
i
e
,
T.
X
i
o
n
g
,
a
n
d
M
.
W
u
,
“
A
r
e
v
i
e
w
o
f
se
n
si
n
g
t
e
c
h
n
o
l
o
g
i
e
s
f
o
r
i
n
d
o
o
r
a
u
t
o
n
o
m
o
u
s
mo
b
i
l
e
r
o
b
o
t
s
,
”
S
e
n
so
rs
,
v
o
l
.
2
4
,
n
o
.
4
,
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
4
0
4
1
2
2
2
.
[
1
8
]
O
.
K
e
d
z
i
o
r
a
e
t
a
l
.
,
“
A
c
t
i
v
e
sp
e
e
d
a
n
d
c
r
u
i
se
c
o
n
t
r
o
l
o
f
t
h
e
m
o
b
i
l
e
r
o
b
o
t
b
a
s
e
d
o
n
t
h
e
a
n
a
l
y
s
i
s
o
f
t
h
e
p
o
si
t
i
o
n
o
f
t
h
e
p
r
e
c
e
d
i
n
g
v
e
h
i
c
l
e
,
”
i
n
2
0
2
2
2
6
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Me
t
h
o
d
s
a
n
d
M
o
d
e
l
s
i
n
A
u
t
o
m
a
t
i
o
n
a
n
d
R
o
b
o
t
i
c
s,
M
MA
R
2
0
2
2
-
Pro
c
e
e
d
i
n
g
s
,
A
u
g
.
2
0
2
2
,
p
p
.
1
8
1
–
1
8
6
.
d
o
i
:
1
0
.
1
1
0
9
/
M
M
A
R
5
5
1
9
5
.
2
0
2
2
.
9
8
7
4
2
9
8
.
[
1
9
]
R
.
F
a
r
k
h
,
K
.
A
l
Jal
o
u
d
,
S
.
A
l
h
u
w
a
i
m
e
l
,
M
.
T.
Q
u
a
s
i
m,
a
n
d
M
.
K
s
o
u
r
i
,
“
A
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
f
o
r
t
h
e
mo
b
i
l
e
-
r
o
b
o
t
m
o
t
i
o
n
c
o
n
t
r
o
l
sy
s
t
e
m
,
”
I
n
t
e
l
l
i
g
e
n
t
A
u
t
o
m
a
t
i
o
n
a
n
d
S
o
f
t
C
o
m
p
u
t
i
n
g
,
v
o
l
.
2
9
,
n
o
.
2
,
p
p
.
4
2
3
–
4
3
5
,
2
0
2
1
,
d
o
i
:
1
0
.
3
2
6
0
4
/
i
a
s
c
.
2
0
2
1
.
0
1
6
2
1
9
.
[
2
0
]
D
.
Tr
u
j
i
l
l
o
,
L
.
A
.
M
o
r
a
l
e
s
,
D
.
C
h
á
v
e
z
,
a
n
d
D
.
F
.
P
o
z
o
,
“
Tr
a
j
e
c
t
o
r
y
t
r
a
c
k
i
n
g
c
o
n
t
r
o
l
o
f
a
mo
b
i
l
e
r
o
b
o
t
u
si
n
g
n
e
u
r
a
l
n
e
t
w
o
r
k
s,
”
Em
e
rg
i
n
g
S
c
i
e
n
c
e
J
o
u
r
n
a
l
,
v
o
l
.
7
,
n
o
.
6
,
p
p
.
1
8
4
3
–
1
8
6
2
,
D
e
c
.
2
0
2
3
,
d
o
i
:
1
0
.
2
8
9
9
1
/
ESJ
-
2
0
2
3
-
07
-
06
-
0
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
I
n
d
o
o
r
n
a
vig
a
tio
n
fo
r
mo
b
ile
r
o
b
o
ts
b
a
s
ed
o
n
d
ee
p
r
ein
fo
r
ce
men
t le
a
r
n
in
g
…
(
K
h
o
a
N
g
u
ye
n
Da
n
g
)
2757
[
2
1
]
H
.
Li
,
Q
.
Zh
a
n
g
,
a
n
d
D
.
Zh
a
o
,
“
D
e
e
p
r
e
i
n
f
o
r
c
e
m
e
n
t
l
e
a
r
n
i
n
g
-
b
a
s
e
d
a
u
t
o
m
a
t
i
c
e
x
p
l
o
r
a
t
i
o
n
f
o
r
n
a
v
i
g
a
t
i
o
n
i
n
u
n
k
n
o
w
n
e
n
v
i
r
o
n
m
e
n
t
,
”
I
E
EE
T
r
a
n
sa
c
t
i
o
n
s
o
n
N
e
u
r
a
l
N
e
t
w
o
r
k
s
a
n
d
L
e
a
r
n
i
n
g
S
y
st
e
m
s
,
v
o
l
.
3
1
,
n
o
.
6
,
p
p
.
2
0
6
4
–
2
0
7
6
,
Ju
n
.
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
TN
N
LS.
2
0
1
9
.
2
9
2
7
8
6
9
.
[
2
2
]
M
.
F
.
R
.
Le
e
a
n
d
S
.
H
.
Y
u
s
u
f
,
“
M
o
b
i
l
e
r
o
b
o
t
n
a
v
i
g
a
t
i
o
n
u
si
n
g
d
e
e
p
r
e
i
n
f
o
r
c
e
me
n
t
l
e
a
r
n
i
n
g
,
”
Pr
o
c
e
sses
,
v
o
l
.
1
0
,
n
o
.
1
2
,
p
.
2
7
4
8
,
D
e
c
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
p
r
1
0
1
2
2
7
4
8
.
[
2
3
]
J.
Z
h
a
o
,
M
.
K
o
n
g
,
Q
.
Li
,
a
n
d
X
.
S
u
n
,
“
C
o
n
t
r
a
c
t
-
b
a
se
d
c
o
m
p
u
t
i
n
g
r
e
s
o
u
r
c
e
ma
n
a
g
e
m
e
n
t
v
i
a
d
e
e
p
r
e
i
n
f
o
r
c
e
men
t
l
e
a
r
n
i
n
g
i
n
v
e
h
i
c
u
l
a
r
f
o
g
c
o
m
p
u
t
i
n
g
,
”
I
EEE
Ac
c
e
ss
,
v
o
l
.
8
,
p
p
.
3
3
1
9
–
3
3
2
9
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
1
9
.
2
9
6
3
0
5
1
.
[
2
4
]
K
.
Zh
u
a
n
d
T.
Z
h
a
n
g
,
“
D
e
e
p
r
e
i
n
f
o
r
c
e
m
e
n
t
l
e
a
r
n
i
n
g
b
a
se
d
m
o
b
i
l
e
r
o
b
o
t
n
a
v
i
g
a
t
i
o
n
:
A
r
e
v
i
e
w
,
”
T
si
n
g
h
u
a
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
2
6
,
n
o
.
5
,
p
p
.
6
7
4
–
6
9
1
,
O
c
t
.
2
0
2
1
,
d
o
i
:
1
0
.
2
6
5
9
9
/
TST
.
2
0
2
1
.
9
0
1
0
0
1
2
.
[
2
5
]
R
.
P
a
t
e
l
a
n
d
S
.
P
a
t
e
l
,
“
A
c
o
mp
r
e
h
e
n
si
v
e
s
t
u
d
y
o
f
a
p
p
l
y
i
n
g
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
f
o
r
c
o
m
p
u
t
e
r
v
i
s
i
o
n
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Ad
v
a
n
c
e
d
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
2
9
,
n
o
.
6
S
p
e
c
i
a
l
I
ssu
e
,
p
p
.
2
1
6
1
–
2
1
7
4
,
2
0
2
0
.
[
2
6
]
J.
P
l
a
t
t
a
n
d
K
.
R
i
c
k
s,
“
C
o
m
p
a
r
a
t
i
v
e
a
n
a
l
y
s
i
s
o
f
R
O
S
-
U
n
i
t
y
3
D
a
n
d
R
O
S
-
G
a
z
e
b
o
f
o
r
m
o
b
i
l
e
g
r
o
u
n
d
r
o
b
o
t
s
i
m
u
l
a
t
i
o
n
,
”
J
o
u
r
n
a
l
o
f
I
n
t
e
l
l
i
g
e
n
t
a
n
d
R
o
b
o
t
i
c
S
y
st
e
m
s:
T
h
e
o
ry
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
0
6
,
n
o
.
4
,
D
e
c
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
8
4
6
-
0
2
2
-
0
1
7
6
6
-
2.
[
2
7
]
J.
Y
u
,
Y
.
S
u
,
a
n
d
Y
.
Li
a
o
,
“
T
h
e
p
a
t
h
p
l
a
n
n
i
n
g
o
f
mo
b
i
l
e
r
o
b
o
t
b
y
n
e
u
r
a
l
n
e
t
w
o
r
k
s
a
n
d
h
i
e
r
a
r
c
h
i
c
a
l
r
e
i
n
f
o
r
c
e
men
t
l
e
a
r
n
i
n
g
,
”
Fro
n
t
i
e
rs
i
n
N
e
u
ro
r
o
b
o
t
i
c
s
,
v
o
l
.
1
4
,
O
c
t
.
2
0
2
0
,
d
o
i
:
1
0
.
3
3
8
9
/
f
n
b
o
t
.
2
0
2
0
.
0
0
0
6
3
.
[
2
8
]
F
.
F
o
r
o
u
g
h
i
,
Z.
C
h
e
n
,
a
n
d
J.
W
a
n
g
,
“
A
C
N
N
-
b
a
se
d
sy
s
t
e
m
f
o
r
m
o
b
i
l
e
r
o
b
o
t
n
a
v
i
g
a
t
i
o
n
i
n
i
n
d
o
o
r
e
n
v
i
r
o
n
men
t
s
v
i
a
v
i
su
a
l
l
o
c
a
l
i
z
a
t
i
o
n
w
i
t
h
a
sma
l
l
d
a
t
a
se
t
,
”
W
o
rl
d
El
e
c
t
ri
c
V
e
h
i
c
l
e
J
o
u
r
n
a
l
,
v
o
l
.
1
2
,
n
o
.
3
,
p
.
1
3
4
,
A
u
g
.
2
0
2
1
,
d
o
i
:
1
0
.
3
3
9
0
/
w
e
v
j
1
2
0
3
0
1
3
4
.
[
2
9
]
J.
L
.
B
a
a
n
d
D
.
P
.
K
i
n
g
m
a
,
“
A
d
a
m:
a
m
e
t
h
o
d
f
o
r
st
o
c
h
a
s
t
i
c
o
p
t
i
mi
z
a
t
i
o
n
,
”
3
r
d
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
L
e
a
r
n
i
n
g
Re
p
r
e
se
n
t
a
t
i
o
n
s
,
I
C
L
R
2
0
1
5
-
C
o
n
f
e
r
e
n
c
e
T
r
a
c
k
Pr
o
c
e
e
d
i
n
g
s
,
p
p
.
1
–
1
5
,
2
0
1
5
.
[
3
0
]
A
.
F
a
r
l
e
y
,
J.
W
a
n
g
,
a
n
d
J
.
A
.
M
a
r
s
h
a
l
l
,
“
H
o
w
t
o
p
i
c
k
a
mo
b
i
l
e
r
o
b
o
t
s
i
mu
l
a
t
o
r
:
a
q
u
a
n
t
i
t
a
t
i
v
e
c
o
m
p
a
r
i
s
o
n
o
f
C
o
p
p
e
l
i
a
S
i
m
,
G
a
z
e
b
o
,
M
O
R
S
E
a
n
d
W
e
b
o
t
s
w
i
t
h
a
f
o
c
u
s
o
n
a
c
c
u
r
a
c
y
o
f
m
o
t
i
o
n
,
”
S
i
m
u
l
a
t
i
o
n
M
o
d
e
l
l
i
n
g
Pr
a
c
t
i
c
e
a
n
d
T
h
e
o
r
y
,
v
o
l
.
1
2
0
,
p
.
1
0
2
6
2
9
,
N
o
v
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
s
i
mp
a
t
.
2
0
2
2
.
1
0
2
6
2
9
.
[
3
1
]
W
.
Jo
,
J.
K
i
m,
R
.
W
a
n
g
,
J
.
P
a
n
,
R
.
K
.
S
e
n
t
h
i
l
k
u
mara
n
,
a
n
d
B
.
-
C
.
M
i
n
,
“
S
M
A
R
TmB
O
T
:
a
R
O
S
2
-
b
a
s
e
d
l
o
w
-
c
o
st
a
n
d
o
p
e
n
-
s
o
u
r
c
e
mo
b
i
l
e
r
o
b
o
t
p
l
a
t
f
o
r
m
,
”
a
rX
i
v
:
2
2
0
3
.
0
8
9
0
3
,
2
0
2
2
.
[
3
2
]
M
.
K
.
C
h
e
g
e
n
i
,
A
.
R
a
s
h
n
o
,
a
n
d
S
.
F
a
d
a
e
i
,
“
C
o
n
v
o
l
u
t
i
o
n
-
l
a
y
e
r
p
a
r
a
me
t
e
r
s
o
p
t
i
mi
z
a
t
i
o
n
i
n
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s,
”
K
n
o
w
l
e
d
g
e
-
B
a
se
d
S
y
s
t
e
m
s
,
v
o
l
.
2
6
1
,
p
.
1
1
0
2
1
0
,
F
e
b
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
k
n
o
s
y
s.
2
0
2
2
.
1
1
0
2
1
0
.
[
3
3
]
G
.
Y
a
o
,
R
.
S
a
l
t
u
s,
a
n
d
A
.
D
a
n
i
,
“
I
mag
e
mo
me
n
t
-
b
a
se
d
e
x
t
e
n
d
e
d
o
b
j
e
c
t
t
r
a
c
k
i
n
g
f
o
r
c
o
m
p
l
e
x
mo
t
i
o
n
s,”
I
E
EE
S
e
n
so
rs
J
o
u
r
n
a
l
,
v
o
l
.
2
0
,
n
o
.
1
2
,
p
p
.
6
5
6
0
–
6
5
7
2
,
J
u
n
.
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
JS
EN
.
2
0
2
0
.
2
9
7
6
5
4
0
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
K
h
o
a
Ng
u
y
e
n
Da
n
g
re
c
e
iv
e
d
a
n
M
.
Sc
.
d
e
g
re
e
in
in
f
o
rm
a
ti
o
n
tec
h
n
o
lo
g
y
fr
o
m
Ha
n
o
i
U
n
iv
e
rsit
y
o
f
S
c
ien
c
e
a
n
d
Tec
h
n
o
lo
g
y
,
Vie
tn
a
m
,
in
2
0
1
2
a
n
d
a
P
h
.
D.
d
e
g
re
e
i
n
a
e
ro
sp
a
c
e
e
n
g
in
e
e
rin
g
fro
m
Uls
a
n
Un
iv
e
rsity
,
Ko
re
a
.
He
is
a
lec
tu
re
r
a
t
th
e
F
a
c
u
lt
y
o
f
En
g
i
n
e
e
rin
g
a
n
d
Tec
h
n
o
lo
g
y
,
I
n
t
e
rn
a
ti
o
n
a
l
S
c
h
o
o
l
,
Vie
t
n
a
m
Na
ti
o
n
a
l
Un
iv
e
rsity
.
His
re
se
a
rc
h
in
tere
sts
in
c
l
u
d
e
UA
V,
UG
V,
c
o
n
tr
o
l
sy
ste
m
s,
i
n
telli
g
e
n
t
c
o
n
tr
o
l,
Io
T
,
a
n
d
a
rti
ficia
l
in
telli
g
e
n
c
e
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
k
h
o
a
n
d
@
v
n
u
is.ed
u
.
v
n
.
Va
n
Tr
a
n
Th
i
re
c
e
iv
e
d
a
n
M
S
c
.
d
e
g
re
e
in
i
n
fo
rm
a
ti
o
n
tec
h
n
o
l
o
g
y
fr
o
m
Ha
n
o
i
Un
iv
e
rsity
o
f
S
c
ien
c
e
a
n
d
Tec
h
n
o
lo
g
y
,
Vie
tn
a
m
in
2
0
1
2
.
Cu
rre
n
t
ly
,
sh
e
is
a
lec
tu
re
r
a
t
th
e
Un
iv
e
rsity
o
f
Lab
o
u
r
a
n
d
S
o
c
ial
Affa
irs.
He
r
re
se
a
rc
h
in
tere
sts
i
n
c
lu
d
e
UG
V,
c
o
n
tro
l
sy
ste
m
s
,
Io
T,
a
n
d
a
rti
ficia
l
in
telli
g
e
n
c
e
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
tran
v
a
n
tk
4
@
g
m
a
il
.
c
o
m
.
Ng
u
y
e
n
Va
n
Th
a
n
g
re
c
e
iv
e
d
a
n
M
.
Sc
.
d
e
g
re
e
in
e
lec
tri
c
a
l
e
n
g
in
e
e
rin
g
fro
m
L
e
Qu
y
D
o
n
Un
i
v
e
rsity
,
Vie
tn
a
m
i
n
2
0
0
7
,
a
n
d
h
is
P
h
.
D
.
d
e
g
re
e
i
n
t
h
e
sa
m
e
m
a
jo
r
fro
m
t
h
e
Un
iv
e
rsity
o
f
En
g
in
e
e
ri
n
g
a
n
d
Tec
h
n
o
l
o
g
y
,
Vie
tn
a
m
Na
ti
o
n
a
l
U
n
i
v
e
rsity
,
Ha
n
o
i,
Vie
tn
a
m
.
He
is
c
u
rre
n
tl
y
a
lec
tu
re
r
i
n
th
e
F
a
c
u
lt
y
o
f
E
lec
tro
n
ics
a
n
d
Tele
c
o
m
m
u
n
ica
ti
o
n
s,
UET.
Hi
s
in
tere
stin
g
i
n
c
lu
d
e
Io
T,
D
S
P
/se
n
s
o
r
a
p
p
li
c
a
ti
o
n
s,
a
n
d
m
e
d
ica
l
e
q
u
i
p
m
e
n
t.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
n
v
t
h
a
n
g
d
tv
t@v
n
u
.
e
d
u
.
v
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.