Inter
national
J
our
nal
of
Electrical
and
Computer
Engineering
(IJECE)
V
ol.
9,
No.
3,
June
2019,
pp.
1773
1779
ISSN:
2088-8708,
DOI:
10.11591/ijece.v9i3.pp1773-1779
r
1773
Robot
na
vigation
in
unkno
wn
en
vir
onment
with
obstacle
r
ecognition
using
laser
sensor
Neer
endra
K
umar
1
,
Zoltán
Vámossy
2
1,2
John
v
on
Neumann
F
aculty
of
Informatics,
Ób
uda
Uni
v
ersity
,
Hung
ary
1
Department
of
Computer
Science
&
IT
,
Central
Uni
v
ersity
of
Jammu,
India
Article
Inf
o
Article
history:
Recei
v
ed
Jun
5,
2018
Re
vised
No
v
19,
2018
Accepted
Des
15,
2018
K
eyw
ords:
Laser
scan
Obstacle
a
v
oidance
Obstacle
recognition
Robot
na
vig
ation
T
urtlebot
ABSTRA
CT
Robot
na
vig
ation
in
unkno
wn
and
dynamic
en
vironments
may
result
in
aimless
w
an-
dering,
corner
traps
and
repetiti
v
e
path
loops.
T
o
address
these
issues,
this
paper
presents
the
solution
by
compar
ing
the
standard
de
viation
of
the
distance
ranges
of
the
obstacles
appeared
in
the
robot
na
vig
ation
path.
F
or
the
similar
obstacles,
The
standard
de
viations
of
distance
range
v
ectors,
obtained
from
the
laser
range
finder
sensor
of
the
robot
at
similar
pose,
are
v
ery
close
to
each
other
.
Therefore,
the
measurements
of
odometer
sensor
are
also
combined
with
the
standard
de
viation
to
recognize
the
loca-
tion
of
the
obstacles.
A
no
v
el
algorithm,
with
obstacle
detection
feature,
is
presented
for
robot
na
vig
ation
in
unkno
wn
and
dynamic
en
vironments.
The
algorithm
checks
the
similarity
of
the
distance
range
v
ectors
of
the
obstacles
in
the
path
and
uses
this
information
in
combination
with
t
he
odometer
measurements
to
identify
the
obstacles
and
their
locations.
The
e
xperimental
w
ork
is
carried
out
using
Gazebo
simulator
.
Copyright
c
2019
Institute
of
Advanced
Engineering
and
Science
.
All
rights
r
eserved.
Corresponding
A
uthor:
Neerendra
K
umar
,
John
v
on
Neumann
F
aculty
of
Informatics,
Ób
uda
Uni
v
ersity
,
Bécsi
út
96/b,
H-1034
Budapest,
Hung
ary
.
Email:
neerendra.kumar@phd.uni-ob
uda.hu
1.
INTR
ODUCTION
Autonomous
na
vig
ation
of
robots
is
g
aining
popularity
among
the
researchers
day
by
day
.
R
e
g
ardl
ess
of
the
uncertainties
in
the
sensors
observ
ations,
robot
may
na
vig
ate
safely
from
start
to
goal
[1].
Using
range
finder
sensors
(e.g.
laser
scan
sensors),
the
robot
can
na
vig
ate
from
start
to
goal
with
obstacle
a
v
oidance
in
unkno
wn
and
dynamic
en
vironments
en
vironment
[2].
The
obstacle
size
and
v
elocity
v
ector
of
the
unmanned
surf
ace
v
ehicles
can
be
used
to
achie
v
e
the
obstacle
a
v
oidance.
Moreo
v
er
,
fusion
of
more
than
one
na
vig
ation
algorithms
can
result
more
accurate
outcomes
[3].
V
arious
requirements
of
obstacle
a
v
oidance
can
be
satisfied
for
the
changing
distance
between
robot
and
the
obstacles
[4].
The
issue
of
obstacle
a
v
oidance
is
treated
as
an
optimization
problem
in
[5].
On
the
unmark
ed
path,
dynamic
colour
perception
model
for
the
autonomous
na
vig
ation
is
e
xplained
in
[6].
Ne
v
ertheless,
this
model
is
not
suitable
for
the
laser
range
finder
sensors.
The
comple
xity
of
the
object
matching
task
rises
for
similar
objects
[7].
Pre
vious
e
xperiences
during
the
robot
na
vig
ation
are
helpful
to
predict
the
path
with
obstacle
a
v
oidance
of
static
and
dynamic
obstacles
[8].
T
ool
based
learning
for
robots
is
presented
by
[9].
In
case
of
industrial
robots,
an
algorithm
for
condition
monitoring
using
artificial
intelligence
is
studied
in
[10].
An
algorithm
for
obstacles
a
v
oidance
in
unkno
wn
en
vironment
using
b
umper
e
v
ents
of
the
robot
is
proposed
in
[11].
Ho
we
v
er
,
the
algorithm
presented
in
[11]
has
the
limitation
that
the
robot
m
ay
struck
in
the
corners
and
may
follo
w
the
same
loop
of
the
path
ag
ain
and
ag
ain.
A
fuzzy
controller
for
ob
s
tacle
a
v
oidance
in
unkno
wn
en
vironment
is
presented
in
[12].
Although,
the
fuzzy
rules,
for
the
obstacles
in
the
central
part
of
the
laser
scan
area,
are
right
or
left
biased.
Therefore,
the
robot
may
trap
in
a
repetiti
v
e
na
vig
ation
path
loop
in
some
situations.
In
this
paper
,
the
scan
range
v
ectors
are
compared
to
detect
the
pre
viously
occurred
obstacles
on
the
same
locations
of
the
na
vig
ation
path.
J
ournal
homepage:
http://iaescor
e
.com/journals/inde
x.php/IJECE
Evaluation Warning : The document was created with Spire.PDF for Python.
1774
r
ISSN:
2088-8708
2.
PR
OBLEM
DEFINITION
AND
PR
OPOSED
ALGORITHM
During
robot
na
vig
ation
in
unkno
wn
en
vironment,
the
robot
may
trap
in
a
loop
and
may
repeat
the
same
path
ag
ain
and
ag
ain.
This
may
be
due
to
the
presence
of
static
or
dynamic
obst
acles.
T
o
deal
with
this
type
of
problem,
there
is
a
need
to
find
a
method
which
can
recognize
the
pre
viously
occurred
obstacles
in
successi
v
e
repetition
of
the
path.
So
that,
the
robot
can
be
directed
to
re
v
erse
the
angular
v
elocity
applied
so
f
ar
.
In
this
w
ay
,
the
robot
will
come
out
of
the
repetiti
v
e
na
vig
ation
path
loop.
Algorithm
1
Obstacle
recognition
using
laser
sensor
and
odometer
1:
Create
v
ectors:
U
,
V
,
X
,
Y
,
T
,
S
;
2:
Initialize:
R
max
;
g
;
D
t
;
c;
T
s
;
T
f
;
t
;
P
t
;
v
c
;
!
c
;
v
b
;
3:
while
(
T
ime
<
T
f
)
do
4:
R
scan
ranges
recei
v
ed;
5:
f
or
i
1
to
length
of
R
do
6:
if
R
(
i
)
==
N
aN
then
7:
R
(
i
)
R
max
;
8:
end
if
9:
end
f
or
10:
g
g
+
1
;
11:
Collect
odometer
readings
at
the
time
of
scan;
X
(
g
)
X
coordinate
from
odometer;
Y
(
g
)
Y
coordinate
from
odometer;
12:
D
min
minimum
of
R
;
13:
if
D
min
<
D
t
then
14:
c
c
+
1
;
15:
standard
de
viation
of
R
;
16:
S
(
c
)
;
17:
U
(
c
)
X
coordinate
from
odometer;
18:
V
(
c
)
Y
coordinate
from
odometer;
19:
T
(
c
)
time
of
scan;
20:
f
or
j
1
to
(length
of
S
1)
do
21:
D
j
;c
p
(
U
(
c
)
U
(
j
))
2
+
(
V
(
c
)
V
(
j
))
2
;
22:
r
j
(
S
(
c
)
S
(
j
))
j
;
23:
if
(
r
<
t
and
D
j
;c
<
P
t
)
then
24:
Match
found
between
scans
j
and
c
.
Record
corresponding
scan
times
and
positions
of
robot
with
respect
to
j
and
c
.;
25:
end
if
26:
end
f
or
27:
v
j
v
b
j
;
28:
!
!
c
;
29:
else
30:
v
v
c
;
31:
!
0
;
32:
end
if
33:
Send
v
;
!
to
robot;
34:
end
while
The
laser
scanning
is
a
reliable
method
for
obstacles
range
finding.
Generally
,
the
laser
scanner
produces
a
v
ector
of
ranges
(
R
)
of
obstacles.
If
there
are
n
number
of
readings
in
a
scan
then
R
can
be
e
xpressed
by
(1)
as
follo
ws:
R
=
[
d
k
]
;
for
k
=
1
to
n:
(1)
Where,
d
is
a
distance
range.
Int
J
Elec
&
Comp
Eng,
V
ol.
9,
No.
3,
June
2019
:
1773
–
1779
Evaluation Warning : The document was created with Spire.PDF for Python.
Int
J
Elec
&
Comp
Eng
ISSN:
2088-8708
r
1775
Standard
de
viation
(
)
of
R
can
be
e
xpressed
by
(2)
as
gi
v
en
belo
w:
=
s
P
n
k
(
d
k
d
)
2
n
:
(2)
Where,
d
is
the
mean
of
R
.
Standard
de
viations
of
tw
o
range
v
ectors
can
be
compared
to
find
the
similarity
between
them.
If
the
standard
de
viations
of
the
tw
o
scan
range
v
ectors
are
close
enough
then
there
is
the
possibility
that
these
scans
bel
ongs
to
the
same
obstacle.
Further
,
the
standard
de
viation
in
combination
of
the
robot
position
coor
-
dinates
can
be
used
to
confirm
that
the
scan
ranges
belongs
to
the
same
obstacle.
T
o
achie
v
e
this
solution,
the
Algorithm
1
is
presented.
3.
EXPERIMENT
AL
RESUL
TS
AND
DISCUSSION
The
proposed
Algorithm
1
is
implemented
using
a
notebook
computer
equipped
with
4xIntel
Cor
e
i3
CPU
@
230
GHz
.
The
memory
of
the
system
is
3952MB.
Ub
untu
14.04.5
L
TS
operating
system
is
used.
Robotics
System
T
oolbox
The
memory
of
the
system
is
3952MB.
Ub
untu
14.04.5
L
TS
operating
system
is
used.
Robotics
System
T
oolbox
in
MA
TLAB
(R2018a)
is
applied
for
computer
programs.
Simulated
T
urtlebot
robot
model
is
tak
en
in
Gazebo
2.2.6
simulator
.
The
numerical
v
alues
of
the
v
ariables,
init
ialized
in
Algorithm
1,
are
presented
in
T
able
1.
T
able
1.
Description
of
v
ariables
and
initialized
v
alues
V
ariable
Description
Initialized
V
alue
R
max
Maximum
range
of
the
laser
scanner
Def
ault
g
a
counter
v
ariable
0
D
t
Threshold
distance
of
robot
and
obstacles
0.6
(meter)
c
a
counter
v
ariable
0
T
s
Starting
time
0
T
f
Finish
time
80
(seconds)
t
Threshold
dif
ference
between
of
tw
o
scans
0.0005
P
t
Threshold
distance
between
tw
o
positions
0.2
(meter)
v
c
F
orw
ard
linear
v
elocity
0.5
(meter/sec.)
!
c
Angular
v
elocity
0
:
6
(radian/sec.)
v
b
Backw
ard
linear
v
elocity
0
:
02
(meter/sec.)
v
Published
linear
v
elocity
-
!
Published
angular
v
elocity
-
U
X
coordinates
of
poses
when
obstacles
are
near
-
V
Y
coordinates
of
poses
when
obstacles
are
near
-
X
X
coordinates
of
all
poses
-
Y
Y
coordinates
of
all
poses
-
T
T
ime
at
scans
performed
-
S
Standard
de
viations
scan
range
v
ectors
-
The
robot
is
dri
v
en,
using
Algorithm
1,
in
the
Gazebo
world
as
gi
v
en
by
Figure
1.
In
the
be
ginning,
in
Figure
1,
the
T
urtlebot
is
located
at
the
centre
of
the
w
orld
and
heading
to
w
ards
positi
v
e
X
axis.
In
Figure
1,
coordinate
ax
es
and
their
measurement
mark
er
are
added
to
pro
vide
the
understanding
of
the
coordinate
po-
sitions
in
the
Gazebo
world
.
The
Gazebo
world
is
constructed
by
using
bric
k
boxes
of
the
dimensions
3x1x3
from
the
model
database
of
the
Gazebo
simulator
.
Purposely
,
The
Gazebo
wor
ld
is
constr
ucted
using
only
one
model
so
that
the
same
type
of
obstacle
encountered
each
time
to
increase
the
comple
xity
in
dif
ferentiating
the
obstacles
and
their
respecti
v
e
locations.
In
addition,
t
he
maze
structure
is
k
ept
closed
so
that
the
robot
can
be
forced
to
follo
w
a
repeated
path
in
a
loop.
The
Gazebo
world
and
the
simulated
T
urtlebot
can
be
brought
up
in
the
system
by
e
x
ecuting
the
follo
wing
command
in
a
terminal
of
Ub
untu
:
Robot
navigation
in
unknown
en
vir
onment
with
obstacle
r
eco
gnition
using
laser
sensor
(Neer
endr
a
K
umar)
Evaluation Warning : The document was created with Spire.PDF for Python.
1776
r
ISSN:
2088-8708
r
oslaunc
h
turtlebot_gazebo
turtlebot_world.launc
h
world_file:=file
name
-3
-2
-1
0
1
2
3
X (meters)
-2.5
-2
-1.5
-1
-0.5
0
0.5
1
1.5
2
2.5
3
Y (meters)
Figure
1.
Simulated
w
orld
in
Gazebo.
Robot
is
dri
v
en
at
a
constant
forw
ard
linear
v
elocity
and
zero
angular
v
elocity
until
an
y
closed
obstacle
is
encountered.
In
this
na
vig
ation
process,
if
an
y
obstacle
is
closed
enough
to
the
robot
then
a
backw
ard
linear
v
elocity
and
a
constant
non
zero
angular
v
elocity
is
pro
vided
to
the
robot.
Using
Algorithm
1,
the
robot
na
vig
ates
in
a
path
loop.
Figure
2
sho
ws
the
resultant
path
of
the
robot.
It
is
clear
from
Figure
2
that
robot
is
dri
v
en
in
a
na
vig
ation
loop
so
that
it
can
observ
e
the
same
obstacles
more
than
once.
During
the
na
vig
ation
process,
the
obstacles
distances
are
measured
by
laser
scan
ranges.
F
or
T
urtlebot
,
a
laser
scan
message
contains
a
v
ector
of
640
range
readings
of
the
scan
area
from
right
to
left.
If
robot
pose
is
same,
each
time,
the
laser
scan
of
an
obstacle
produces
same
v
ector
of
ranges.
T
o
find
the
similarity
between
tw
o
v
ectors
of
ranges,
standard
de
viations
of
the
tw
o
v
ectors
are
compared.
-2
-1.5
-1
-0.5
0
0.5
1
1.5
2
X (meters)
-1.5
-1
-0.5
0
0.5
1
1.5
Y (meters)
Robot navigation path
Robot positions
Figure
2.
Robot
path
generated
from
odometer
measurements.
Figure
3
presents
the
standard
de
viations
of
the
range
v
ectors
of
dif
ferent
scans
performed
during
the
na
vig
ation
of
the
robot.
Figure
3
depicts
that
there
are
periodic
spik
es
which
are
due
to
the
similarity
of
the
obstacles
and
repetiti
v
e
path
loop.
Further
,
the
pattern
of
the
graph
in
the
periods
is
found
v
ery
similar
to
each
other
because
of
the
similarity
of
the
obstacles.
Subsequently
,
it
can
be
inferred
that
the
standard
de
viations
of
the
similar
obstacles
are
observ
ed
close
to
each
other
.
Ho
we
v
er
,
in
addition
to
the
standard
de
viation,
the
odometer
measurements
,
for
the
robot
locations
during
the
na
vig
ation,
are
used
to
determine
the
location
of
the
obstacles.
Further
,
on
the
basis
of
the
standard
de
viations
and
odometer
measurements,
it
can
not
be
identified
that
ei
ther
the
obs
tacle
w
as
visited
by
the
robot
pre
vi
ously
or
the
obstacle
i
s
encountered
first
time
during
the
na
vig
ation.
F
or
this
reason,
there
is
the
need
for
additional
method
to
recognize
that
whether
an
y
particular
obstacle
is
found
first
time
or
it
is
repeated.
The
occurrence
number
(i.e.
first
time
encountered
or
repeated)
of
Int
J
Elec
&
Comp
Eng,
V
ol.
9,
No.
3,
June
2019
:
1773
–
1779
Evaluation Warning : The document was created with Spire.PDF for Python.
Int
J
Elec
&
Comp
Eng
ISSN:
2088-8708
r
1777
an
obstacle
can
be
described
on
the
basis
of
the
time
of
the
scan.
Ob
viously
,
the
standard
de
viations
and
robot
positions
of
tw
o
nearby
scans
at
v
ery
short
time
of
dif
ference
will
not
v
ary
lar
gely
.
On
the
other
hand,
if
the
standard
de
viations
of
the
distance
range
v
ectors
of
an
y
tw
o
scans
and
the
robot
positions
recei
v
ed
from
the
odometer
at
the
time
of
scans
are
v
ery
close
to
each
for
both
of
the
scans
and
the
time
dif
ference
between
the
tw
o
scans
is
lar
ge
than
a
threshold
v
alue
then
it
can
be
said
that
the
obstacle
is
same
and
the
occurrence
of
this
obstacle
is
the
repeated
one.
0
50
100
150
200
Scan number
1.7
1.75
1.8
1.85
1.9
1.95
2
2.05
Standard deviation (meters)
Standard deviation of scans
Figure
3.
Standard
de
viation
of
scan
ranges.
T
able
2
pre
sents
the
positions
of
the
robot
and
their
corresponding
times
with
respect
to
scan
range
matches.
It
can
be
observ
ed
from
the
T
able
2
that
using
standard
de
viat
ions
and
robot
positions
the
similarity
between
the
tw
o
scan
range
v
ectors
are
found
either
at
times
with
short
dif
ference
or
at
times
wh
e
re
the
time
interv
al
is
lar
ger
.
The
matches
at
short
and
lar
ge
time
interv
als
indicates
about
the
occurrence
time
of
the
obstacles.
F
or
e
xample,
the
first
ro
w
of
T
able
2
indicates
that
the
match
between
the
range
v
ectors
is
found
at
the
scan
times
10.23
(seconds)
and
10.46
(seconds).
Thus,
the
time
interv
al
(Time2
Time1)
between
these
scans
is
less
than
a
second.
K
eeping
in
mind
the
forw
ard
linear
v
elocity
of
the
robot
(i.e.
0.5
meter/seconds),
it
is
clear
t
hat
the
scans
were
performed
at
nearly
same
time
therefore
the
match
is
found
between
the
first
appearance
of
the
obstacle
and
this
is
not
the
case
of
repetition.
Similarly
,
the
last
ro
w
of
the
T
able
2
presents
the
time
interv
al
between
the
scans
as
34.57
(seconds).
Therefore,
on
the
basis
of
robot’
s
linear
and
angular
v
elocities
it
can
be
deduced
that
the
occurrence
of
the
obstacle
at
T
ime2
(at
last
ro
w)
is
repeated
in
na
vig
ation
path
loop.
T
able
2.
Laser
scan
range
v
ectors
matches
and
robot
positions
T
ime1
(Sec.)
T
ime2
(Sec.)
Scan
Numbers
Matched
Robot
Position
at
T
ime1
Robot
Position
at
T
ime2
10.23
10.46
35,
37
(1.59,
-1.11)
(1.59,
-1.11)
19.08
21.28
48,
63
(-1.66,
-0.99)
(-1.67,
-0.93)
21.73
21.84
64,
65
(-1.69,
-0.86)
(-1.70,
-0.81)
26.32
26.44
69,
70
(-1.64,
1.05)
(-1.64,
1.05)
28.68
28.8
85,
86
(-1.60,
1.05)
(-1.55,
1.06)
37.62
38.64
95,
103
(1.59,
1.01)
(1.57,
1.02)
37.17
39.66
92,
107
(1.59,
1.01)
(1.61,
0.89)
9.68
44.91
31,
113
(1.59,
-1.12)
(1.51,
-1.07)
9.68
45.13
31,
115
(1.59,
-1.12)
(1.51,
-1.08)
10.56
45.13
38,
115
(1.60,
-1.11)
(1.51,
-1.08)
Robot
navigation
in
unknown
en
vir
onment
with
obstacle
r
eco
gnition
using
laser
sensor
(Neer
endr
a
K
umar)
Evaluation Warning : The document was created with Spire.PDF for Python.
1778
r
ISSN:
2088-8708
4.
CONCLUSION
Using
the
standard
de
viation
of
laser
scan
range
v
ectors,
the
similarity
of
obstacles
is
e
xamined
in
unkno
wn
en
vironment.
In
addition
to
the
standard
de
viation,
the
robot
position
is
also
used
to
identify
whether
the
obstacle
position
is
same
as
of
some
earlier
position
during
the
na
vig
ation
path
loop.
It
is
e
vident
from
the
T
able
2
that
the
matches
between
scan
range
v
ectors
are
found
at
the
closed
positions
of
the
robot
in
each
round
of
the
na
vig
ation
loop.
T
ime
interv
al
between
the
scans
plays
an
important
role
in
the
detection
of
the
repetiti
v
e
occurrences
of
the
obstacles.
Using
the
standard
de
viation
of
laser
scan
range
v
ectors,
robot
positions
and
time
of
scan,
the
angular
v
elocity
can
be
re
v
ersed
to
break
the
path
loop.
A
CKNO
WLEDGEMENT
This
study
is
inspired
by
the
T
ekiContr
ol
research
project
of
J
ohn
von
Neumann
F
aculty
of
Informatics,
Ób
uda
Univer
sity
,
Budapest,
Hungary
.
REFERENCES
[1]
I.
Arv
anitakis,
A.
Tzes,
and
K.
Giannousakis,
“Mobile
robot
na
vig
ation
under
pose
uncertainty
in
un-
kno
wn
en
vironments,
”
IF
A
C-P
aper
sOnLine
,
v
ol.
50,
no.
1,
pp.
12
710
–
12
714,
2017.
[2]
F
.
Kamil,
T
.
S.
Hong,
W
.
Khaksar
,
M.
Y
.
Moghrabiah,
N.
Zulkifli,
and
S.
A.
Ahmad,
“Ne
w
robot
na
v-
ig
ation
algorithm
for
arbitrary
unkno
wn
dynamic
en
vironments
based
on
future
prediction
and
priority
beha
vior
,
”
Expert
Systems
with
Applications
,
v
ol.
86,
pp.
274
–
291,
2017.
[3]
H.
Mousazadeh,
H.
Jaf
arbiglu,
H.
Abdolmaleki,
E.
Omrani,
F
.
Monhaseri,
M.
reza
Abdollahzadeh,
A.
Mohammadi-Aghdam,
A.
Kiapei,
Y
.
Salmani-Zakaria,
and
A.
Makhsoos,
“De
v
eloping
a
na
vig
ation,
guidance
and
obstacle
a
v
oidance
algorithm
for
an
unmanned
surf
ace
v
ehicle
(usv)
by
algorithms
fusion,
”
Ocean
Engineering
,
v
ol.
159,
pp.
56
–
65,
2018.
[4]
L.
Zong,
J.
Luo,
M.
W
ang,
and
J.
Y
uan,
“Obstacl
e
a
v
oidance
handling
and
mix
ed
inte
ger
predicti
v
e
control
for
space
robots,
”
Advances
in
Space
Resear
c
h
,
v
ol.
61,
no.
8,
pp.
1997
–
2009,
2018.
[5]
Y
.
Zhao,
X.
Chai,
F
.
Gao,
and
C.
Qi,
“Obstacle
a
v
oidance
and
motion
planning
scheme
for
a
he
xapod
robot
octopus-iii,
”
Robotics
and
A
utonomous
Systems
,
v
ol.
103,
pp.
199
–
212,
2018.
[6]
A.
Narayan,
E.
T
uci,
F
.
Labrosse,
and
M.
H.
M.
Alkilabi,
“
A
dynamic
colour
perception
system
for
autonomous
robot
na
vig
ation
on
unmark
ed
roads,
”
Neur
ocomputing
,
v
ol.
275,
pp.
2251
–
2263,
2018.
[7]
G.
K
ertész,
S.
Szénási,
and
Z.
Vámossy
,
“Multi-directional
image
projections
with
fix
ed
resolution
for
object
matching,
”
Acta
P
olytec
hnica
Hungarica
,
v
ol.
15,
no.
2,
pp.
211
–
229,
2018.
[8]
L.
Nardi
and
C.
Stachniss,
“User
preferred
beha
viors
for
robot
na
vig
ation
e
xploiting
pre
vious
e
xperi-
ences,
”
Robotics
and
A
utonomous
Systems
,
v
ol.
97,
pp.
204
–
216,
2017.
[9]
H.
W
icaksono
and
C.
Sammut,
“T
ool
use
learning
for
a
real
robot,
”
International
J
ournal
of
Electrical
and
Computer
Engineering
,
v
ol.
8,
no.
2,
pp.
1230–1237,
2018.
[10]
A.
A.
Jaber
and
R.
Bick
er
,
“De
v
elopment
of
a
condition
monitoring
algorithm
for
industrial
robots
based
on
artificial
intelligence
and
si
gn
a
l
processing
techniques,
”
International
J
ournal
of
Electrical
and
Com-
puter
Engineering
,
v
ol.
8,
no.
2,
pp.
996–1009,
2018.
[11]
N.
K
umar
,
Z.
Vámossy
,
and
Z.
M.
Szabó-Resch,
“Robot
obstacle
a
v
oidance
using
b
umper
e
v
ent,
”
in
2016
IEEE
11th
International
Symposium
on
Applied
Computational
Intellig
ence
and
Infor
matics
(SA
CI)
,
May
2016,
pp.
485–490.
[12]
N.
K
umar
,
M.
T
akács,
and
Z.
Vámossy
,
“
Robot
na
vig
ation
in
unkno
wn
en
vironment
using
fuzzy
logic,
”
in
2017
IEEE
15th
International
Symposium
on
Applied
Mac
hine
Intellig
ence
and
Informatics
(SAMI)
,
Jan
2017,
pp.
279–284.
Int
J
Elec
&
Comp
Eng,
V
ol.
9,
No.
3,
June
2019
:
1773
–
1779
Evaluation Warning : The document was created with Spire.PDF for Python.
Int
J
Elec
&
Comp
Eng
ISSN:
2088-8708
r
1779
BIOGRAPHY
OF
A
UTHORS
Neer
endra
K
umar
is
Assistant
Professor
at
Department
of
Computer
Science
and
IT
,
Central
Uni
v
ersity
of
Jammu,
India.
His
teaching
interest
are
in
Artificial
i
ntelligence,
Programming
languages,
Algorithm
analysis
and
Robotics.
He
has
recei
v
ed
St
ipendium
Hung
aricum
Schol-
arship
a
w
ard
in
2015
for
the
duration
of
three
years.
He
is
doing
his
PhD
studies
at
John
v
on
Neumann
F
aculty
of
Informatics,
Ób
uda
Uni
v
ersity
,
Budapest,
Hung
ary
.
His
research
area
is
robot
na
vig
ation.
His
Scopus
and
ORCID
IDs
are
57190585849
and
0000-0002-6609-8974,
respecti
v
ely
.
The
detailed
information
can
be
accessed
on
the
follo
wing
URL:
http://cujammu.ac.in//5084/5084_media/nareendra.pdf
Zoltán
Vámossy
is
Associate
Professor
at
John
v
on
Neumann
F
aculty
of
Informatics,
Ób
uda
Uni
v
ersity
,
Budapest,
Hung
ary
.
He
has
completed
his
PhD
from
Budapest
Uni
v
ersity
of
T
echnology
and
Economics,
Hung
ary
in
2010.
The
area
of
his
PhD
studie
s
include
Na
vig
ating
mobile
robots
with
machine
vision
based
on
P
AL
optics.
His
current
research
interests
are
focused
on
mobile
robotics,
robot
na
vig
ati
on,
image
processing
and
computer
vision.
He
has
recei
v
ed
v
arious
a
w
ards
including
the
János
Neumann
A
w
ard
(János
Neumann
Computer
Science
Association)
in
2016.
F
or
Further
information
please
visit
the
URL:
http://uni-ob
uda.hu/en/node/701
Robot
navigation
in
unknown
en
vir
onment
with
obstacle
r
eco
gnition
using
laser
sensor
(Neer
endr
a
K
umar)
Evaluation Warning : The document was created with Spire.PDF for Python.