Inter
national
J
our
nal
of
Electrical
and
Computer
Engineering
(IJECE)
V
ol.
10,
No.
2,
April
2020,
pp.
2173
2181
ISSN:
2088-8708,
DOI:
10.11591/ijece.v10i2.pp2173-2181
r
2173
Swarm
r
obotics
:
Design
and
implementation
Ashraf
Ab
uelhaija,
A
yham
J
ebr
ein,
T
arik
Baldawi
Department
of
Electrical
Engineering,
Applied
Science
Pri
v
ate
Uni
v
ersity
,
Amman,
Jordan
Article
Inf
o
Article
history:
Recei
v
ed
Mei
17,
2018
Re
vised
Oct
18,
2019
Accepted
No
v
3,
2019
K
eyw
ords:
Artificial
intelligence
(AI)
Infrared
(IR)
array
Sw
arm
robotics
Whisk
ers
circuit
ABSTRA
CT
This
project
presents
a
sw
arming
and
herding
beha
viour
using
simple
robots.
The
main
goal
is
to
demonstrate
the
applicability
of
artificial
intelligence
(AI)
in
simple
robotics
that
can
then
be
scaled
to
industrial
and
consumer
mark
ets
to
further
the
ability
of
automation.
AI
can
be
achie
v
ed
in
man
y
dif
ferent
w
ays;
this
paper
e
xplores
the
possible
platforms
on
which
to
b
uild
a
simple
AI
robots
from
consumer
grade
microcontrollers.
Emphasis
on
simplicity
is
the
main
focus
of
this
paper
.
Cheap
and
8
bit
microcontrollers
were
used
as
the
brain
of
each
robot
in
a
decentralized
sw
arm
en
vironment
were
each
robot
is
autonomous
b
ut
still
a
part
of
the
whole.
These
simple
robots
don’
t
communicate
directly
with
ea
ch
other
.
The
y
will
utilize
simple
IR
sensors
to
sense
each
other
and
simple
limit
switches
to
sense
other
obstacles
in
their
en
vironment.
Their
main
objecti
v
e
is
to
assemble
at
certain
location
after
initial
start
from
random
locations,
and
after
con
v
er
ging
the
y
w
ould
mo
v
e
as
a
single
unit
without
collisions.
Using
readily
a
v
ailable
microcon-
trollers
and
simple
circuit
design,
semi-consistent
sw
arming
beha
viour
w
as
achie
v
ed.
These
robots
don’
t
follo
w
a
s
et
path
b
ut
will
react
dynamically
to
dif
ferent
scenarios,
guided
by
their
simple
AI
algorithm.
Copyright
c
2020
Insitute
of
Advanced
Engineeering
and
Science
.
All
rights
r
eserved.
Corresponding
A
uthor:
Ashraf
Ab
uelhaija,
Applied
Science
Pri
v
ate
Uni
v
ersity
,
Al
Arab
st.
21,
Amman,
Jordan.
T
el:
065609999
Email:
a
ab
ualhijaa@asu.edu.jo
1.
INTR
ODUCTION
Automation
is
an
important
part
of
most
industries
b
ut
on
the
other
hand
normal
automation
has
some
shortcomings.
F
or
e
xamples,
robot
in
manuf
acturing
industry
can
only
do
what
its
code
tells
it
to
do,
the
cruise
control
on
a
car
can
only
speed
up
or
speed
do
wn
the
car
,
and
equipment
in
hospitals
can
only
monitor
patients
and
alert
the
doctors
in
case
of
anomalies
.
In
all
pre
vious
e
xamples
the
de
vices
cannot
mak
e
decisions
or
change
their
beha
viour
without
the
input
of
a
human
operator
.
Sw
arm
robotics
ha
v
e
been
studied
in
the
conte
xt
of
producing
dif
ferent
collecti
v
e
beha
viors
to
solv
e
tasks
such
as:
aggre
g
ation
[1],
pattern
formation
[2],
self-assembly
and
morphogenesis
[3],
object
clustering,
assembling
and
construction
[4],
collecti
v
e
search
and
e
xploration
[5,
6],
coordinated
motion
[7],
collecti
v
e
transportation
[8,
9],
self-deplo
yment
[10],
foraging
[11]
and
others.
The
objecti
v
e
of
this
w
ork
is
focused
on
ho
w
the
field
of
artificial
intelligence
can
be
used
to
transfer
automation
into
the
ne
xt
le
v
el
of
scientific
adv
ancements.
Self
dri
ving
cars,
robots
that
can
perform
sur
gery
,
robots
in
customer
service
that
can
understand
the
intricacies
of
human
speech
and
respond
accordingly
,
these
are
all
adv
ancements
that
are
happening
currently
.
One
of
the
most
important
applications
that
is
implemented
using
artificial
intelligence
is
what
is
kno
wn
as
sw
arm
robotics.
Sw
arm
robotics
is
a
field
of
robotics
that
deals
with
multi
-robot
systems
where
a
l
ar
ge
number
of
simple
robots
coordinate
t
o
display
col
lecti
v
e
beha
viour
when
interacting
with
each
other
or
the
en
vironment.
The
field
of
sw
arm
robotics
puts
emphasis
on
number
,
J
ournal
homepage:
http://ijece
.iaescor
e
.com/inde
x.php/IJECE
Evaluation Warning : The document was created with Spire.PDF for Python.
2174
r
ISSN:
2088-8708
simplicity
and
scalability
of
the
robots.
And
another
k
e
y
component
is
the
collecti
v
e
intelligence
of
the
sw
arm
where
the
indi
vidual
is
simple
b
ut
the
collecti
v
e
can
display
comple
x
beha
viour
,
that
tak
e
inspiration
from
insects
such
as
ants.
The
ability
to
communicate
is
paramount
to
achi
e
v
e
decentralization,
and
to
insure
constant
feedback
between
indi
viduals.
This
paper
simply
demonstrates
ho
w
to
b
uild
and
program
se
v
eral
robotics
in
order
to
obtain
sw
arm
robotics
with
the
follo
wing
functionality:
sensing
other
robots
in
the
vicinity
and
na
vig
ating
an
area
with
se
v
eral
other
robots
present.
The
suggested
solution
is
to
use
un
algorithm
that
induces
sw
arming
or
herding
beha
viour
consistently
.
These
robots
will
interact
with
their
en
vironment
using
se
v
eral
e
xternal
sensors,
micro-
switches,
infrared
sensors
and
other
hardw
are
pieces.
These
robot
will
react
to
these
e
xternal
de
vices
depending
on
the
algorithms
used
in
their
main
control
unit.
Man
y
solutions
ha
v
e
been
suggested
by
researchers
to
achie
v
e
this
goal.
Micael
S.Couceiro
and
his
group
[12]
g
a
v
e
surv
e
y
on
multi-robot
search
inspired
on
sw
arm
intelligence.
Fi
v
e
state-of-the-art
sw
arm
robotic
algorithms
are
described
and
compared.
Simulat
ed
e
xperiments
of
a
mapping
task
are
carried
out
to
compare
the
fi
v
e
algorithms.
The
three
best
performing
algorithms
are
deeply
compared
using
14
e-pucks
on
a
source
localization
problem.
The
Robot
ic
Darwinian
P
article
Sw
arm
Optimization
(RDPSO)
algorithm
depicts
an
impro
v
ed
con
v
er
gence.
M.
Rubenstein
[13]
and
his
team
in
their
paper”
A
Lo
w
Cost
Scalable
Robot
System
for
Collecti
v
e
Beha
viors.
“
presented
Kilobot,
a
lo
w-cost
robot
designed
to
mak
e
testing
collecti
v
e
algorithms
on
hundreds
or
thousands
of
robots
accessible
to
robotics
researchers.
T
o
enable
the
possibility
of
lar
ge
Kilobot
collecti
v
es
where
the
number
of
robots
is
an
order
of
magnitude
lar
ger
than
the
lar
gest
that
e
xist
today
,
each
robot
is
made
with
only
$
14
w
orth
of
parts
and
tak
es
5
minutes
to
assemble.
Furthermore,
the
robot
design
allo
ws
a
single
user
to
easily
operate
a
lar
ge
Kilobot
collecti
v
e,
such
as
programming,
po
wering
on,
and
char
ging
all
robots,
which
w
ould
be
dif
ficult
or
impossible
to
do
with
man
y
e
xisting
robotic
systems.
M.
Rubenstein
et
al.
[14]
created
a
lar
ge
sw
arm
of
programmed
robots
that
can
form
c
ollaborations
using
only
local
information.
The
robots
could
communicate
only
with
nearby
members,
within
about
three
times
their
diameter
.
The
y
were
abl
e
to
assemble
into
comple
x
preprogrammed
shapes.
If
the
robots
formation
hit
snags
when
the
y
b
umped
into
one
another
or
because
of
an
outlier
,
additional
algorithms
guided
them
to
rectify
their
collecti
v
e
mo
v
ements.
M.
Senanayak
e
and
his
team
[15]
re
vie
wed
the
seminal
w
orks
that
addressed
this
problem
i
n
the
area
of
sw
arm
robotics,
which
is
the
application
of
sw
arm
intelligence
principles
to
the
control
of
multi-robot
systems.
Rob
ustness,
scalability
and
fle
xibility
,
as
well
as
distrib
uted
sensing,
mak
e
sw
arm
robotic
systems
well
suited
for
the
problem
of
tar
get
search
and
tracking
in
real-w
orld
applications.The
y
classify
their
w
ork
according
to
the
v
ariations
and
aspects
of
the
search
and
tracking
problems
the
y
addressed.
Micael
S.Couceiro
and
his
group
[16]
mentioned
the
e
xtension
of
the
P
article
Sw
arm
Opt
imiza-
tion
to
multi-robot
applications
which
has
been
pre
viously
proposed
and
denoted
as
Robotic
Darwinian
PSO
(RDPSO).
His
w
ork
contrib
utes
with
a
further
e
xtension
of
the
RDPSO,
thus
inte
grating
tw
o
research
aspects:
(i)
an
autonomous,
realistic
and
f
ault-tolerant
initial
deplo
yment
st
rate
gy
denoted
as
Extended
Spi-
ral
of
Theodorus
(EST),
and
(ii)
a
f
ault-tolerant
distrib
uted
search
to
pre
v
ent
communication
netw
ork
splits.
The
e
xploring
agents,
denoted
as
scouts,
are
autonomously
deplo
yed
using
supporting
agents,
denoted
as
rangers.
Experiment
al
results
with
15
ph
ysical
scouts
and
3
ph
ys
ical
rangers
sho
w
that
the
algorithm
con
v
er
ges
to
the
optimal
soluti
on
f
aster
and
more
accurately
using
the
EST
approach
o
v
er
the
random
de-
plo
yment
strate
gy
.
Also,
a
more
f
ault-tolerant
strate
gy
clearly
influences
the
time
needed
t
o
con
v
er
ge
to
the
final
solution,
b
ut
is
less
susceptible
to
robot
f
ailures.
M.
K
ubo
et
al.
[17]
e
xplained
ho
w
to
achie
v
e
a
highly
scalable
tar
get
enclosure
model
about
the
number
of
tar
get
to
enclose,
the
y
introduce
sw
arm
bas
ed
task
assignment
capability
to
T
akayama’
s
enclosure
model.
The
original
model
discussed
only
single
tar
get
en
vironment
b
ut
it
is
well
suited
for
applying
to
the
en
vironments
with
multiple
tar
gets.The
y
sho
w
ho
w
the
robots
can
enclose
the
tar
gets
without
predefined
position
assignment
by
analytic
discussion
based
on
switched
systems
and
a
series
of
computer
simulations.
As
a
consequence
of
this
property
,
the
proposed
robots
can
change
their
tar
get
according
to
the
criterion
about
robot
density
while
the
y
enclose
multiple
tar
gets.
B.
Y
ang
and
his
team
[18]
proposed
a
decentralized
control
algorithm
of
sw
arm
robot
for
tar
get
search
and
trapping
inspired
by
bacteria
chemotaxis.
First,
a
local
coordinate
system
is
established
according
to
the
initial
positions
of
the
robots
in
the
tar
get
area.
Then
the
tar
get
area
is
di
vided
into
V
oronoi
cells.
After
the
initialization,
sw
arm
robots
start
performing
tar
get
search
and
trapping
missions
dri
v
en
by
the
proposed
bacteria
Int
J
Elec
&
Comp
Eng,
V
ol.
10,
No.
2,
April
2020
:
2173
–
2181
Evaluation Warning : The document was created with Spire.PDF for Python.
Int
J
Elec
&
Comp
Eng
ISSN:
2088-8708
r
2175
chemotaxis
algorithm
under
the
guidance
of
the
gradient
informat
ion
defined
by
the
tar
get.
Simulation
results
demonstrate
the
ef
fecti
v
eness
of
the
algorithm
and
its
rob
ustness
to
une
xpected
robot
f
ailure.
F
or
more
details
about
sw
arm
robotics
design
and
technology
,
refer
to
the
follo
wing
references
[19-23]
2.
DESIGN
INTEGRA
TION
2.1.
Hard
war
e
The
hardw
are
choices
will
f
a
v
or
simple
and
cheap
parts
for
better
scalability
while
sacrificing
the
ability
of
the
indi
vidual
robot
of
doing
comple
x
beha
vior
.
An
8-bit
micro-controller
from
Atmel
w
as
chosen
as
the
brains
of
each
robot,
this
8pin
chip
is
simple
to
program
and
has
enough
computing
po
wer
to
achie
v
e
the
desired
design.
A
duel
H-bridge
motor
dri
v
er
chip
from
T
e
xas
Instruments
is
used
to
interf
ace
and
control
tw
o
DC
motors
for
mo
v
ement.
A
combination
of
and
and
not
g
ate
array
chips
will
act
as
I/O
e
xtenders
so
as
to
free
pin
on
the
chip
for
other
uses.
A
combination
of
se
v
eral
IR
transmitter
diodes,
a
single
IR
recei
v
er
diode
and
tw
o
limit
switches
are
used
for
sensing
and
interacting
with
the
en
vironment.
The
flo
wchart
is
sho
wn
in
Figure
1.
Start
Start
turning
Reading
IR sensor
Stop
turning
Y
es
Is
sensor value
within
limits?
Apply
algorithm to
sensor value
Product of
algorithm
(algorithm value)
Move
according to
algorithm value
W
ere
the whiskers
triggered?
No
Move
until end
and stop
Which
side?
T
urn left
180 degree
T
urn right
180 degree
Y
es
Right
Left
Move forward
2 cm
Figure
1.
Hardw
are
flo
wchart
On
first
po
wered
an
initializing
phase
will
commence,
in
this
stage
all
the
needed
modules
inside
the
microchip
will
be
acti
v
ated
and
in
case
of
the
ADC
module,
it
wi
ll
perform
se
v
eral
reading
that
will
be
discarded.
This
is
done
to
insure
accurate
and
f
ast
reading
during
operation.
After
the
initialization
phase
the
robot
will
start
to
operate.
It
starts
by
continuously
turning
in
place
and
reading
v
alues
from
the
IR
sensor
.
This
method
is
needed
because
of
the
IR
array
shape
which
lea
v
es
blind
g
aps
that
are
eliminated
by
contin-
uously
turning
the
array
.
If
a
spik
e
in
the
v
oltage
reading
is
sensed
this
will
indicate
that
another
robot
is
in
Swarm
r
obotics
:
Design
and
implementation
(Ashr
af
Ab
uelhaija)
Evaluation Warning : The document was created with Spire.PDF for Python.
2176
r
ISSN:
2088-8708
line
of
site.
This
v
alue
is
compared
to
kno
wn
limits.
If
the
v
alue
is
belo
w
the
lo
wer
limit
it
will
be
treated
as
noise
and
the
robot
will
continue
turning
and
searching
for
a
v
alid
v
alue.
If
the
v
alue
is
abo
v
e
the
higher
limit
it
will
be
treated
as
an
object
that
is
close
enough.
If
the
v
alue
f
alls
within
the
limit
the
robot
will
tak
e
this
v
alue,
run
it
through
the
algorithm,
then
the
outcome
will
translate
into
a
timed
interv
al
of
forw
ard
mo
v
ement.
The
relationship
between
the
algorithm
output
and
distance
between
the
robots
is
represented
by
a
simple
parabolic
function,
were
the
longest
distance
mo
v
ed
forw
ard
is
when
the
object
is
belie
v
ed
to
be
in
the
middle
of
the
kno
wn
limits
while
the
robot
will
not
mo
v
e
a
great
deal
if
the
other
robots
are
too
f
ar
or
too
close.
This
beha
vior
is
repeated
indefinitely
.
After
deciding
ho
w
the
robots
will
beha
v
e
we
e
xamine
the
main
control
unit
to
decide
ho
w
to
achie
v
e
the
w
anted
design
using
it.
The
A
Ttin
y85
MCU
with
it
s
fi
v
e
bidirectional
ports
is
used
to
control
outside
peripherals
such
as
motors
and
motor
dri
v
er
.
Because
of
ho
w
it
is
designed
the
A
T
in
y85
cannot
supply
enough
current
to
run
a
motor
directly
,
we
use
a
motor
dri
v
er
to
interf
ace
the
MCU
with
the
motors.
The
chosen
dri
v
er
is
the
DR
V8833
on
a
preb
uilt
board
from
Pololu.
This
dri
v
er
interf
aces
tw
o
bidirectional
DC
motors
to
the
MCU
as
seen
in
figure
2.
The
scheme
of
control
Unit
is
depicted
in
Figure
3.
Figure
2.
Dual
H-Bridge
connection
diagram
Figure
3.
Control
unit
schematic
Each
robot
needs
a
w
ay
to
sense
other
robots
around
him
and
react
to
them.
There
are
a
couple
of
w
ays
to
achie
v
e
this,
through
ultrasonic
sensors,
limit
switches
or
infrared
sensors.
F
or
this
project
a
combination
of
switches
and
infrared
sensors
w
as
chosen.
The
A
Ttin
y85
has
an
ADC
module
that
can
read
changes
in
v
oltage
on
some
of
the
I/O
pins
we
will
use
this
ability
to
read
the
v
oltage
from
an
infrared
recei
v
er
to
determine
the
distance
between
tw
o
robots.
Each
robot
will
need
to
broadcast
infrared
signals
in
all
directions
so
other
robots
can
read
these
v
alues
and
determine
the
distance.
one
recei
v
er
is
enough
for
each
robot
b
ut
because
these
diodes
are
v
ery
directional
each
robot
needs
a
number
of
transmitter
diode
so
as
to
transmit
in
all
directions.
A
simple
solution
is
to
create
an
array
of
diodes
transmitting
in
a
circle.
T
esting
sho
wed
that
such
a
solution
will
need
an
array
of
man
y
diodes
to
mitig
ate
the
problem
of
signal
directi
vity
so
as
another
solution
is
to
ha
v
e
each
robot
turn
in
continuous
manner
so
as
to
create
a
light
house
ef
fect,
where
at
some
point
a
recei
v
er
will
be
in
line
of
sight
of
a
transmitter
.
F
or
this
project
an
array
of
eight
diodes
arranged
in
a
he
xagon
w
as
chosen,
as
seen
in
figure
4.
Int
J
Elec
&
Comp
Eng,
V
ol.
10,
No.
2,
April
2020
:
2173
–
2181
Evaluation Warning : The document was created with Spire.PDF for Python.
Int
J
Elec
&
Comp
Eng
ISSN:
2088-8708
r
2177
Figure
4.
IR
array
This
IR
pair
system
is
only
useful
when
sensing
other
robots
it
can’
t
sense
obstacles.
F
or
this
pr
o
j
ect
micro
switches
were
used.
By
connecting
a
micro
switch
to
a
long
whisk
er
-lik
e
met
al
rod
it
can
sense
objects
in
front
of
the
robot.
Using
tw
o
switches
the
robot
can
discern
the
location
of
the
obstacle
in
relation
to
mo
v
ement
direction.
A
micro
switch
has
three
pins
which
are;
common,
connected
and
not
connected.
The
pins
flip
when
the
switch
is
dispersed.
This
system
need
to
w
ork
with
only
tw
o
pins.
One
pin
will
alert
the
MCU
to
an
obstacle
and
the
other
pin
will
decide
the
location
of
the
obs
tacle.
The
location
mechanism
is
achie
v
ed
by
using
a
v
oltage
di
vider
,
when
a
switch
is
pressed
the
v
alue
of
the
di
vider
will
change,
this
v
alue
is
then
read
by
the
MCU
and
compared
ag
ainst
a
kno
wn
v
alue
that
will
gi
v
e
which
switch
w
as
pressed.
The
chosen
design
can
be
seen
in
Figure
5.
Figure
5.
whisk
ers
circuit
design
In
order
to
lessen
the
noise
from
digital
part
of
the
circuit
we
isolate
the
analog
and
digital
lines
to
dif
ferent
sides
of
the
MCU.
In
the
real
circuit
decoupling
capacitors
were
mounted
for
each
indi
vidual
IC
to
help
with
po
wer
supply
stability
.
The
final
circuit
w
as
b
uilt
in
four
independent
PCB
boards
that
are
connected
using
headers
to
mak
e
the
circuit
more
compact
as
seen
in
Figure
6.
Figure
6.
Schematic
of
full
circuit
Swarm
r
obotics
:
Design
and
implementation
(Ashr
af
Ab
uelhaija)
Evaluation Warning : The document was created with Spire.PDF for Python.
2178
r
ISSN:
2088-8708
T
o
po
wer
each
robot
tw
o
rechar
geable
4.2V
lithium-ions
are
used;
one
for
the
motor
circuit
and
the
other
for
the
control
circuit.
The
chassis
for
the
robot
is
made
from
ple
xiglass
in
the
shape
of
a
non-uniform
he
xagon.
This
shape
w
as
chosen
for
its
ease
of
manuf
acturing.
Three
robots
prototype
are
sho
wn
in
Figure
7.
Figure
7.
Prototype
2.2.
Softwar
e
The
code
for
the
robot
w
as
written
in
A
VR
assembly
using
Atmel
Studio.
The
code
w
as
brok
en
do
wn
into
small
subroutines
then
the
y
were
connected
together
after
testing
each
one.
The
algorithm
w
as
decided
upon
after
testing
ho
w
the
v
alues
read
from
the
infrared
recei
v
er
correlate
with
distance
as
seen
in
the
hardw
are
section.
The
equation
is
a
simple
second
order
polynomial;
this
equation
will
mak
e
the
robots
get
closer
to
each
other
.
Once
the
y
are
suf
ficiently
close,
the
algorithm
will
pre
v
ent
them
from
straying
too
f
ar
from
each
other
.
This
ef
fect
can
be
seen
in
Figure
8.
Figure
8.
algorithm
graph
The
x-axis
in
graph
abo
v
e
represents
the
v
oltage
being
read
from
the
infrared
sensor
.
The
MCU
will
con
v
ert
the
v
oltage
into
a
10
bit
digital
v
alue
using
the
follo
wing
equation:
D
ig
ital
v
al
ue
=
V
in
1024
V
cc
(1)
where
V
in
is
the
v
alue
of
v
oltage
on
pin
and
V
cc
is
the
v
oltage
supply
to
MCU.
The
lo
west
and
highest
limits
represent
the
distance
robot
will
respond
to
respecti
v
ely
.
The
y-axis
represents
the
output
of
the
algorithm
which
is
the
follo
wing
equation:
Y
=
output
=
(
x
+
58)(
x
968)
2048
(2)
The
output
will
be
represented
as
an
8
bit
v
alue
ranging
from
0
to
101,
which
will
be
subsequently
con
v
erted
into
mo
v
ement
time
ranging
from
0
to
1.65478
s
econds.
The
motor
speed
is
around
15cm/s
making
the
maximum
mo
v
ement
range
24.82
cm.
Int
J
Elec
&
Comp
Eng,
V
ol.
10,
No.
2,
April
2020
:
2173
–
2181
Evaluation Warning : The document was created with Spire.PDF for Python.
Int
J
Elec
&
Comp
Eng
ISSN:
2088-8708
r
2179
3.
RESUL
TS,
DISCUSSION
AND
COMP
ARISON
Sw
arming
and
herding
beha
viour
using
AI
robotics
w
as
successfully
implemented
and
de
v
eloped
with
simple
and
accessible
hardw
are
and
softw
are
tools.
The
possibilities
of
utilizing
small
and
simple
robots
to
achie
v
e
compl
e
x
collecti
v
e
beha
viour
w
as
demonstrated
as
well.
F
our
robots
were
b
uilt
according
to
the
specifications
chosen,
and
softw
are
w
as
de
v
eloped
to
achie
v
e
the
d
e
sired
beha
viour
.
T
esting
this
sw
arm
robot
has
sho
wn
that
the
hardw
are
and
softw
are
beha
v
es
as
e
xpected
in
controlled
cases,
the
sw
arm
succeeded
in
attacking
and
reaching
the
desired
goal.
A
brief
comparison
of
pre
vious
w
orks
is
introduced
in
the
follo
wing:
(a)
Raf
ael
Mathiasde
and
his
team
[24]
proposed
a
simple
yet
ef
ficient
distrib
uted
control
algorithm
to
implement
dynamic
task
allocation
in
a
robotic
sw
arm.
In
this
algorithm,
each
robot
that
inte
grates
the
sw
arm
runs
the
algorithm
whene
v
er
it
senses
a
change
in
the
en
vironment.
The
algorithm
w
as
imple-
mented
and
e
xtensi
v
ely
tested
in
dif
ferent
size
sw
arms
of
robots.
The
corresponding
performance
and
ef
fecti
v
eness
are
promising.
The
sw
arm
robot
used
is
Elisa-III.
(b)
Luneque
Silv
a
Juniora
and
Nadia
Nedjah
[25]
presented
t
he
W
a
v
e
Sw
arm
as
a
general
strate
gy
to
manage
the
se
qu
e
nce
of
subtasks
that
compose
the
collecti
v
e
na
vig
ation,
which
is
an
important
task
in
sw
arm
robotics.
The
proposed
strate
gy
is
based
mai
nly
on
the
e
x
ecution
of
w
a
v
e
algorithms.
The
sw
arm
is
vie
wed
as
a
distrib
uted
system,
wherein
the
communication
is
achie
v
ed
by
message
passing
among
robot’
s
neighborhood.
Message
propag
ation
delimits
the
start
and
end
of
each
subtask.
Simulations
are
performed
to
demonstrate
that
controlled
na
vig
ation
of
robot
sw
arms/clus
ters
is
achie
v
ed
with
three
subtasks,
which
are
recruitment,
alignment
and
mo
v
ement.
(c)
I
˜
naki
Na
v
arro
and
Fernando
Mat
´
ıa
[26]
g
a
v
e
an
o
v
ervie
w
of
sw
arm
robotics,
describing
its
main
proper
-
ties
and
characteristics
and
comparing
it
to
general
multi-robotic
systems.
A
re
vie
w
of
dif
ferent
research
w
orks
and
e
xperimental
results,
together
with
a
disc
u
s
sion
of
the
future
sw
arm
robotics
in
real
w
orld
applications
ha
v
e
been
e
xplained
in
details.
(d)
M.Bakhshipoura,
M.Jabbari
Ghadib
and
F
.Namdaria
[27]
proposed
a
no
v
el
heuristic
algorithm
to
solv
e
continuous
non-linear
optimization
problems.
The
presented
algorithm
is
a
collecti
v
e
global
search
inspired
by
the
sw
arm
artificial
intelligent
of
coordinated
robots.
Cooperati
v
e
recognition
and
sensing
by
a
sw
arm
of
mobile
robots
ha
v
e
been
fundamental
inspirations
for
de
v
elopment
of
Sw
arm
Robotics
Search
and
Rescue
(SRSR).
Sw
arm
robotics
is
an
approach
with
the
aim
of
coordinating
multi-robot
systems
which
consist
of
numbers
of
mostly
uniform
simple
ph
ysical
robots.
The
ultimate
aim
is
to
emer
ge
an
eligible
cooperati
v
e
beha
vior
either
from
interactions
of
autonomous
robots
with
the
en
viron-
ment
or
their
mutual
interactions
between
each
other
.
In
this
algorithm,
robots
which
represent
initial
solutions
in
SRSR
terminology
ha
v
e
a
sense
of
en
vironment
to
detect
victim
in
a
search
and
rescue
mission
at
a
disaster
site.
(e)
Seeja
G
and
his
team
[28]
studied
the
progress
in
the
research
of
nature
inspired
sw
arm
robotics.
In
this
w
ork
an
artificial
intelligence
aided
coordination
approach
is
used
for
the
self-or
g
anization
and
decen-
tralization
of
multiple
robots.
Being
a
promising
centralized
approach
with
f
ault
tolerance,
redundanc
y
and
scalability
potentials,
the
y
can
e
v
en
w
ork
when
it
is
technically
infeasible
to
set
up
the
infrastructure
required
to
control
the
robots
in
a
centralized
w
ay
.
But
the
design
of
indi
vidual
robot
le
v
el
practice
to
achie
v
e
a
desired
collecti
v
e
beha
vior
is
really
dif
ficult
as
it
is
hard
to
predict
the
simultaneous
interac-
tions
between
lar
ge
numbers
of
indi
vidual
robots.
In
order
to
e
xplore
the
possibilities
to
mak
e
a
better
progress
in
this
technology
,
the
e
xisting
modelling,
analysis
methods
and
the
challenges
has
to
be
studied
first.
F
ollo
wed
by
this,
a
s
tudy
on
sw
arm
communication
and
the
hardw
are
units
including
sensors
and
actuators
w
as
done.
(f)
Amrit
Saggu,
P
alla
vi
Y
ada
v
,
Monika
Roopak
[29]
said
that
the
inherent
intelligence
of
sw
arms
has
inspired
man
y
social
and
political
philosophers,
in
that
the
collecti
v
e
mo
v
ements
of
an
aggre
g
ate
often
deri
v
e
from
independent
decision
making
on
the
part
of
a
single
indi
vidual.
A
CKNO
WLEDGEMENT
This
w
ork
has
been
done
at
the
Applied
Science
Pri
v
ate
Uni
v
ersity
,
Amman,
JORD
AN,
F
aculty
of
Engineering,
department
of
Electrical
Engineering.
The
author
w
ould
lik
e
to
thank
this
uni
v
ersity
for
their
strong
support
to
this
w
ork
.
Swarm
r
obotics
:
Design
and
implementation
(Ashr
af
Ab
uelhaija)
Evaluation Warning : The document was created with Spire.PDF for Python.
2180
r
ISSN:
2088-8708
REFERENCES
[1]
M.
Dorigo,
V
.
T
rianni,
E.
S
¸
ahin,
R.
Groß,
T
.
H.
Labella,
G.
Baldassarre,
S.
Nolfi,
J.-L.
Deneubour
g,
F
.
Mondada,
D.
Floreano,
et
al.,
”Ev
olving
self-or
g
anizing
beha
viors
for
a
sw
arm-bot,
”
Autonomous
Robots
,
v
ol.
17,
no.
2-3,
pp.
223–245,
2004.
[2]
V
.
T
rianni,
E.
T
uci,
C.
Ampatzis,
and
M.
Dorigo,
”Ev
olutionary
sw
arm
robotics:
A
theoretical
and
methodological
itinerary
from
indi
vidual
neuro-controllers
to
collecti
v
e
beha
viours,
”
The
hori
zons
of
e
v
o-
lutionary
robotics
,
v
ol.
153,
2014.
[3]
G.
Baele,
N.
Bredeche,
E.
Haasdijk,
S.
Maere,
N.
Michiels,
Y
.
V
an
de
Peer
,
T
.
Schmickl,
C.
Schw
arzer
,
and
R.
Thenius,
”Open-ended
on-board
e
v
olutionary
robotics
for
robot
sw
arms,
”
in
2009
IEEE
Congress
on
Ev
olutionary
Computation
,
IEEE,
2009,
pp.
1123–1130.
[4]
G.
S.
Nitschk
e,
M.
C.
Schut,
and
A.
Eiben,
”Ev
olving
beha
vioral
specialization
in
robot
teams
to
solv
e
a
collecti
v
e
construction
task,
”
Sw
arm
and
Ev
olutionary
Computation
,
v
ol.
2,
pp.
25–38,
2012.
[5]
G.
K.
V
enayag
amoorth
y
,
L.
L.
Grant,
and
S.
Doctor
,
”Collecti
v
e
robotic
search
using
h
ybrid
techniques:
Fuzzy
logic
and
sw
arm
intelligence
inspired
by
nature,
”
Engineering
Applications
of
Artificial
Intelli-
gence
,
v
ol.
22,
no.
3,
pp.
431–441,
2009.
[6]
J.
M.
Hereford
and
M.
A.
Siebold,
”Bio-inspired
search
s
trate
gies
for
robot
sw
arms,
”
in
Sw
arm
Robotics
from
Biology
to
Robotics
,
IntechOpen,
2010.
[7]
E.
Ferrante,
A.
E.
T
ur
gut,
C.
Huepe,
A.
Stranieri,
C.
Pinciroli,
and
M.
Dorigo,
”Self-or
g
anized
flocking
with
a
mobile
robot
sw
arm:
a
no
v
el
motion
control
method,
”
Adapti
v
e
Beha
vior
,
v
ol.
20,
no.
6,
pp.
460–477,
2012.
[8]
D.
Zhang,
G.
Xie,
J.
Y
u,
and
L.
W
ang,
”Adapti
v
e
task
assignment
for
multiple
mobile
robots
via
sw
arm
intelligence
approach,
”
Robotics
and
Autonomous
Systems
,
v
ol.
55,
no.
7,
pp.
572–588,
2007.
[9]
L.
Chaimo
wicz,
M.
F
.
Campos,
and
V
.
K
umar
,
”Dynamic
role
assignment
for
cooperati
v
e
robots,
”
in
Proceedings
2002
IEEE
International
Conference
on
Robotics
and
Automation
(Cat.
No.
02CH37292)
,
v
ol.
1.
IEEE,
2002,
pp.
293–298.
[10]
M.
S.
Couceiro,
C.
M.
Figueiredo,
D.
Portug
al,
R.
P
.
Rocha,
and
N.
M.
Ferreira,
”Initial
deplo
yment
of
a
robotic
team-a
hierarchical
approach
under
communication
constraints
v
erified
on
lo
w-cost
platforms,
”
2012
IEEE/RSJ
International
Conference
on
Intelligent
Robots
and
Systems
,
IEEE,
2012,
pp.
4614–
4619.
[11]
A.
Campo,
´
A.
Guti
´
errez,
S.
Nouyan,
C.
Pinciroli,
V
.
Longchamp,
S.
Garnier
,
and
M.
Dorigo,
”Artificial
pheromone
for
path
selection
by
a
foraging
sw
arm
of
robots,
”
Biological
c
ybernetics
,
v
ol.
103,
no.
5,
pp.
339–352,
2010.
[12]
M.
S.
Couceiro,
P
.
A.
V
ar
g
as,
R.
P
.
Rocha,
and
N.
M.
Ferreira,
”Benchmark
of
sw
arm
robotics
distrib
uted
techniques
in
a
search
task,
”
Robotics
and
Autonomous
Systems
,
v
ol.
62,
no.
2,
pp.
200–213,
2014.
[13]
M.
Rubenstein,
C.
Ahler
,
and
R.
Nagpal,
”Kilobot:
A
lo
w
cost
scalable
robot
system
for
collecti
v
e
be-
ha
viors,
”
Robotics
and
Automation
(ICRA),
2012
IEEE
International
Conference
on
,
IEEE,
2012,
pp.
3293–3298.
[14]
M.
Rubenstein,
A.
Cornej
o
,
and
R.
Nagpal,
”Programmable
self-assembly
in
a
thousand-robot
sw
arm,
”
Science
,
v
ol.
345,
no.
6198,
pp.
795–799,
2014.
[15]
M.
Senanayak
e,
I.
Senthooran,
J.
C.
Barca,
H.
Chung,
J.
Kamruzzaman,
and
M.
Murshed,
”Search
and
tracking
algorithms
for
sw
arms
of
robots:
A
surv
e
y
,
”
Robotics
and
Autonomous
Systems
,
v
ol.
75,
pp.
422–434,
2016.
[16]
M.
S.
Couceiro,
C.
M.
Figueiredo,
R.
P
.
Rocha,
and
N.
M.
Ferreira,
”Darwinian
sw
arm
e
xploration
un-
der
communication
constraints:
Initial
deplo
yment
and
f
ault-tolerance
assessment,
”
Robotics
and
Au-
tonomous
Systems
,
v
ol.
62,
no.
4,
pp.
528–544,
2014.
[17]
M.
K
ubo,
H.
Sato,
T
.
Y
oshimura
,
A.
Y
amaguchi,
and
T
.
T
anaka,
”Multiple
tar
gets
enclosure
by
robotic
sw
arm,
”
Robotics
and
Autonomous
Systems
,
v
ol.
62,
no.
9,
pp.
1294–1304,
2014.
[18]
B.
Y
ang,
Y
.
Ding,
Y
.
Jin,
and
K.
Hao,
”Self-or
g
anized
sw
arm
robot
for
tar
get
search
and
trapping
inspired
by
bacterial
chemotaxis,
”
Robotics
and
Autonomous
Systems
,
v
ol.
72,
pp.
83–92,
2015.
[19]
S.-J.
Chung,
A.
A.
P
aranjape,
P
.
Dames
,
S.
Shen,
and
V
.
K
umar
,
”A
surv
e
y
on
aerial
sw
arm
robotics,
”
IEEE
T
ransactions
on
Robotics
,
v
ol.
34,
no.
4,
pp.
837–855,
2018.
[20]
M.
Bonani,
V
.
Longchamp,
S.
Magnenat,
P
.
R
´
etornaz,
D.
Burnier
,
G.
Roulet,
F
.
V
aussard,
H.
Bleuler
,
and
F
.
Mondada,
”The
marXbot,
a
miniature
mobile
robot
opening
ne
w
perspecti
v
es
for
the
collecti
v
e-robotic
research,
”
in
2010
IEEE/RSJ
International
Conference
on
Intelligent
Robots
and
Systems
,
IEEE,
2010,
Int
J
Elec
&
Comp
Eng,
V
ol.
10,
No.
2,
April
2020
:
2173
–
2181
Evaluation Warning : The document was created with Spire.PDF for Python.
Int
J
Elec
&
Comp
Eng
ISSN:
2088-8708
r
2181
pp.
4187–4193.
[21]
M.
Brambilla,
E.
Ferrante,
M.
Birattari,
and
M.
Dorigo,
”Sw
arm
robotics:
a
re
vie
w
from
the
sw
arm
engineering
perspecti
v
e,
”
Sw
arm
Intelligence
,
v
ol.
7,
no.
1,
pp.
1–41,
2013.
[22]
G.
Caprari
and
R.
Sie
gw
art,
”Mobile
micro-robots
ready
to
use:
Alice,
”
in
2005
IEEE/RSJ
international
conference
on
intelligent
robots
and
systems
,
IEEE,
2005,
pp.
3295–3300.
[23]
M.
Dorigo,
E.
T
uci,
R.
Groß,
V
.
T
rianni,
T
.
H.
Labella,
S.
Nouyan,
C.
Ampatzis,
J.-L.
Deneubour
g,
G.
Baldassarre,
S.
Nolfi,
et
al.,
”The
sw
arm-bot
s
project,
”
in
International
W
orkshop
on
Sw
arm
Robotics
,
Springer
,
2004,
pp.
31–44.
[24]
R.
M.
de
Mendonc
¸
a,
N.
Nedjah,
and
L.
de
Macedo
Mourelle,
”Ef
ficient
distrib
uted
algorithm
of
dynamic
task
assignment
for
sw
arm
robotics,
”
Neurocomputing
,
v
ol.
172,
pp.
345–355,
2016.
[25]
L.
S.
Junior
and
N.
Nedjah,
”Ef
ficient
strate
gy
for
collecti
v
e
na
vig
ation
control
in
sw
arm
robotics,
”
Pro-
cedia
Computer
Science
,
v
ol.
80,
pp.
814–823,
2016.
[26]
I.
Na
v
arro
and
F
.
Mat
´
ıa,
”An
introduction
to
sw
arm
robotics,
”
Isrn
robotics
,
v
ol.
2013,
2012.
[27]
M.
Bakhshipour
,
M.
J.
Ghadi,
and
F
.
Namdari,
”Sw
arm
robotics
search
&
rescue:
A
no
v
el
artificial
intelligence-inspired
optimization
approach,
”
Applied
Soft
Computing
,
v
ol.
57,
pp.
708–726,
2017.
[28]
G.
Seeja,
A.
Selv
akumar
,
et
al.,
”A
Surv
e
y
on
Sw
arm
Robotic
Modeling,
Analysis
and
Hardw
are
Archi-
tecture,
”
Procedia
Computer
Science
,
v
ol.
133,
pp.
478–485,
2018.
[29]
A.
Saggu,
P
.
Y
ada
v
,
and
M.
Roopak,
”Applications
of
sw
arm
intelligence,
”
2013.
Swarm
r
obotics
:
Design
and
implementation
(Ashr
af
Ab
uelhaija)
Evaluation Warning : The document was created with Spire.PDF for Python.