Design and Implementation of Coordinated Multipurpose Robotic System with RF and Light Communication Channels.

pillowfistsAI and Robotics

Nov 13, 2013 (3 years and 7 months ago)

59 views


Design and Implementation of Co
ordinated Multipurpose Robotic
System with RF and Ligh
t Communication Channels.

(A paper entirely bas
e
d on study, research a
nd experiments.)


Team Leader: Sakyasingha Mahabatra (SKR Engg Col
l
ege, Anna University)
Jakkamahes (SKR Engg Col
l
ege, Anna University)


Barath Sankararaman (Saveetha Engg College, Anna University)




Email: Barath1985@gmail.com
52/11,3
rd
Cross Street,
Balaji Nagar, Ekkattuthangal,
Chennai – 600097.
Ph: 9444408672

Abstract:


There are many engineering challenges invol
ved in traveling to the Moon and in
particular th
ere is strong
need in adv
a
nced
robo
tics. Robot will be
needed in space for in-
space ass
e
mbly and m
a
intenance surface in-space exploratio
n and hum
an assis
t
ance.

Robots m
u
st
be able to function in u
nknown unstructured and dynam
i
c environm
ents.
Mars is a cold, dry planet with extrem
e te
m
p
er
atures and a th
in atm
o
sphere.The terrain is
rough and often untraversable. Other m
a
terial
s, such as ore grade m
e
tals, petroleum

m
i
nerals, if they exist, need to be discovered.

We are in need for a m
u
l
tipurpose robotic m
odule that would explore Mars. W
e
are also
in need of multip
le robo
tic m
odules which would coordinate am
ong each other and
explore Mars.

All these robots have been fixed with local
sensors, so they can
only identify objects
closer to them. These robots
have to communicate between e
ach other during the task

This paper develops a control logic that
will enable m
u
ltipurpose robotic system
s to
navigate and coordinate with other robotic
m
odules. W
e
are currently working on various
communication and navigational set up. W
e
have
also com
e
up with an idea of Light
Communication in case of
m
u
ltiple robotic
sys
t
em
s. The idea behind th
is
paper is
experim
e
nted with three robotic m
o
dules. Th
ere m
i
ght som
e
abnorm
a
l conditions such as
breakdown of down of the autonomous robotic
modules. In that case, w
e
have built a
m
a
nually controlled robotic rescuer module
which helps the other robotic system
s to
recover from the abnorm
a
lities.








In other words this paper is based on five categories

!

Lead robotic System
(Multipu
r
pose)
!

RF Robotic Communication m
odule
!

Light chaser robot (Form
a
tion control)
!

Rescuer Robot
!

Mission Completion Robot

Lead Robot
(Multipurpose)(Tes
tin
g
Piece on Earth)

This paper proposes a new idea th
at will en
ab
le m
u
ltipurp
o
se m
obile robotic sys
t
em
s to
perf
orm
multip
le task
s in a do
m
e
stic envi
ronm
ent, such as i)monitor
i
ng a
local
environm
ent ii) detecting unusual situations su
ch as trespasses, bom
bs, fire, gas leakage
using a multip
le sen
s
or system
. iii) Ob
ject Recognition and Navigation Object
Recognition
and Navigation f
o
r m
o
bile robo
ts ha
s been one of
the m
a
jor problem
s in the
research of Artificial Intellig
ence. W
ithout this ability, the
possibilities for robots to carry
out useful tasks (as m
e
ntioned above) rem
a
in
lim
ited. The Robot has to have a better
efficiency in
understand
ing the environm
ent. W
e
have developed a sim
p
le m
u
ltipurp
o
se
robot with a highly reliable GaAs – I
R
Sens
or for
sensing and an
algorithm for navigation
along with reverse gear facility
.
W
e
have tested our m
u
ltipur
pose robot in perform
i
ng the
above m
e
ntioned tasks with satisfactory re
sults and are currently researching for

im
proving the reliability and scop
e
of
the robo
t in
rea
l
lif
e
situ
ations. The lead
ro
bot is
multipurpose in nature. The Lead
Robot (RF Module) will be
assisted by the follower
robot. The RF Module a PIC16c
77 with a RTD-AM transcei
ver m
odule from
Abacom
.

There were a lot of disadvant
ages in the RF module when were tying to establish a
communication between
the Lead Robotic Sys
t
em
and Follower Robotic System
such as
receiver end failures and bursty communi
cation problem
s
. So to overcome the
disadvantages of RF Comm
uni
cation, W
e
buil
t
a Light Chaser Robotic system
which
would take part in the form
ation approach.
W
e
could efficiently m
a
ke a for
m
ation among
the robotic system
s with light. The robot wi
ll turn towards the brig
htest light and m
ove
forward chasing or following it. It has two
m
o
tors (
l
ef
t & right) in o
r
der to m
a
ke the
turns; on top two light sensors (photodiodes) se
parated by a P
C
board in order to sim
u
late
"a nose". T
h
is nose is
particula
r
ly
im
porta
nt because it will provide
a shadow thus
preventing both sensors f
r
om
being illum
i
nated when a side light is pres
en
t (the m
o
tor on
the non-illum
i
nated sensor side will turn
on thus aim
i
ng the robot towards the light).
W
h
en the light is right in f
r
ont of the sens
ors, th
ere will be no shadow and both sensors
will be equ
a
lly illum
i
nated, i.
e.: b
o
th m
o
tors will be run
n
ing and th
e robot will m
ove
forward.

The m
a
in ro
bot boa
rd w
a
s an
old k
i
t; it had

two IR pairs of
s
e
nsors (em
itter-
rec
e
ptor
)
to

read a rotating cardboard circle painted bl
ack and white in two
circumferences. This
sequence of black and white sectors programm
e
d
the robot to m
a
ke tu
rns, m
ove forwa
r
d,
stop, etc. This part of the ci
rcuit was rem
oved a
nd a new circuit was designed and put in
place. The creatu
r
e has b
een equipped w
ith speech and voice detection circuits.


These robotic m
odules have been tested with va
rious experim
e
nts on ear
th. W
e
are trying to
develop a control logic using a to
ol called Petri Net Ke
rnel, which would help
us to analyse the
behavior of Coordinated Multipurpose robo
tic
system
s with RF and
Light Co
mmunicatio
n
channels.




Objectives of the Research:


"

Reliable co
ntrol log
i
c using C Interface.

"

Multi –Rob
ot Coordination w
i
th RF Modules

"

Experiments w
i
th t
w
o cheap and simple robots.

"

Visualiz
e and check th
e state o
f
ea
ch robot

"

Control Logic w
i
th short range sen
s
or information.

"

2D – Stimulation Using Stimulation Softw
ares.

"

Reverse Gear Facilities
.

"

Metal Detection, Gas Leakage Detection

"

Highly Sensitive Circuit Breaking

"

Active Gu
idance for the Blind.

"

Victim Detection during disasters.

"

Object Avoidance and Path Indentification

"

AC and DC Facilit
ies

"

Fire/Heat S
e
nsing Mechanism

"

Light Chasing – Coordination

"

Rescuer Robot


System Architechture: Lead Robot
:

Lead Robot
(Multipurpose)





It has a m
i
crocontroller AT89C51 and tw
o stepping m
o
tors. The AT89C51 is a low-
power, high-perform
ance CMOS 8-bit m
i
crocom
puter with 4K bytes of Flash
programm
a
ble and
erasable read on
ly m
e
m
o
ry (PEROM). The device is
m
a
nufactured
using Atm
e
l’s high-density nonvolatile m
e
m
o
ry
technology and is co
m
p
atible with the
industry-standard MCS-51 instruction set
and pinout. The on-chip Flash allows the

program
mem
o
ry to be reprogram
m
e
d in
-system
or by a conventional nonvolatile
m
e
mory programm
e
r. By com
b
ining a versa
t
ile 8-bit CPU with Flash o
n
a m
onolithic

chip, the A
t
m
e
l AT89C51 is a powerful
m
i
crocom
puter which provides a highly-
flexible and cost-effectiv
e solution to
m
a
ny e
m
bedded control applications.


The 555 monolith
ic tim
ing circuit is a highly

stable con
t
roller capab
le of producing

accurate time delays, or oscillation.
In
the tim
e delay m
o
de of operation, the tim
e is

precisely controlled by one external resistor
and
capacitor. F
o
r a stab
le o
p
eration as an

oscillato
r
, th
e free runnin
g
frequency
and the duty
cycle are bo
th accu
rately
contro
lled
with two ex
terna
l
r
e
sistors and

one capacito
r. The circ
uit m
a
y be triggered
and res
e
t o
n
f
a
lling
wavef
o
rm
s, and the
outp
u
t struc
t
ure
can source
or s
i
nk up
to

200m
A
.

GaAs-IR-Luminesz
enzdiode - GaAs Infrared
Emitter is used. Some of the feature
s

of this
IR Emitte
r
i) GaAs inf
r
ar
ed em
itting diode, f
a
bricated in a liquid phase epitax
y

process ii)
H
igh reliability
iii)High
pulse h
a
ndling capab
ility iv) Ava
ila
ble in
group
s. I
t

has been used for Photointerrupting.

The MC78XX/LM78XX/MC78XXA series of
th
ree
te
rminal pos
itiv
e regulator
s
are
available in the TO-220/D-PAK pac
k
age and
with several fixed output voltages, m
a
king
them
useful in a wide range of applications
. Each type em
ploys internal current lim
iting,
therm
a
l shut down and safe operating area pr
otection, m
a
king it essen
tially indestructible.
If adequate heat sinking is provided, they
can deliver over 1A
output current. Although
designed prim
arily as fixed voltage regulators,
these d
e
vices can be u
s
ed with ex
ternal
com
ponents to obtain adjustable voltages
and currents. The TSOP17.. – series are
m
i
niaturized
receiv
e
rs for infrared rem
o
te
control system
s. PIN diode and pream
p
lifier
are assem
b
led on lead fram
e, the epoxy package is designed as IR filter.T
h
e dem
odul
ated

output signal can directly be
decoded by a m
i
croprocessor.
TSOP17.. is the standard IR

rem
o
te control receiver series, s
upporting all m
a
jor transm
ission codes.

Ideally suite
d
f
o
r interf
acing betwee
n low-le
vel logic circuitry and m
u
lt
iple peripheral
power loads, the Series ULN20xxA/L high-
voltage, high-current
Darlington arrays
feature con
t
inuous load
curren
t
ratings to 500
mA for each of the seven drivers. At an
appropriate duty cycle depending on am
bient
temperature and num
ber
of drivers turned
ON si
m
u
ltaneously, typical power loads to
taling over 230 W (350 mA x 7,95 V) c
a
n be
controlled. Typical loads include relays,
solenoids, stepping m
o
tors, m
a
gnetic print
hammers, multip
lexed L
E
D and incandescen
t disp
lays, and heate
r
s. All devices f
e
a
t
ure

open-collector outputs with
integral clam
p diodes. The UL
N2003A/L and ULN2023A/L
have series input resistors selected for ope
ration directly with 5
V TTL or CMOS. These
devices will handle num
e
rous interface
needs — particularly those beyond the
capabilities of standard logic buffers.

The Specifications fo
r the Metal D
e
tector:


power supply ............................. 4.5 V;
DC consumption .......................15 mA;
Indication ...................................sound , LCD Dis
p
lay
Modes ........................................static or dynam
i
c;
Discrim
i
nation.............................ferro/non-ferro

Specificatio
n
s for Fire Sensor:


In this fire alarm
circuit, a therm
i
stor works as the heat sensor. W
h
en tem
p
erature
increases, its resis
t
ance decreases, and vice
vers
a. At nor
m
a
l tem
p
erature, the resis
t
an
ce
of
the therm
i
stor
(T
H1) is approximately
10 kilo-ohms,
which reduces to a few ohm
s

as the temperature increases beyond
100°C.
The circuit uses readily availab
l
e
com
ponents and can be easily cons
tructed on any general-purpose PCB

Specificatio
n
for the CCTV Camera:


W
i
reless code transaction can
transmit and Receive pict
ures through wall with high
resolution up to 100 m
e
teres.

Power Supply:
9 – 12 V
DC
Distance: Wireless Transm
ission
(100 Meteres)\


Basic Mech
anical Design:


Top View:

Fig 1:

Fig 1 shows the basic working m
odel.





Fig2: The design of the Low
er Portion:

The above block shows the basic mechanic
al design of the multipurpose robot
along with the basic components. The CCD Camera, Guide Cane and the
Squirting Components are detachable. They are to be attached during need.
The differentially steered robot
is sim
ilar to the differential
gears
used in autom
obiles
in
that both the wheels can have diffe
rent rate
s of
rotations, b
u
t unlike th
e dif
f
e
rentia
l
gearing
sys
t
em
, a dif
f
e
rentia
lly
ste
e
r
ed sys
t
em
will h
a
ve b
o
th th
e whe
e
ls power
ed.
Differential wheeled robots are
used extensively in robotics
,
since the
i
r motion is ea
sy
to program
and can be well controlled. The
dim
e
nsions of the Robot are 31 * 21 cm
s
.

Som
e
m
echanical cons
iderations su
ch as cast
er angles have been
taken in to account
while designing the robot. Provisions have b
een given for extension of the tasks done
by the robot. A cha
m
ber has been designed fo
r h
o
lding th
e b
a
tte
rie
s
. It runs on AC and
DC supplies.



Follow
er Robot 1:
System Architecture:
System Architechture Follow
er Robot 1(With RF Module) :


It has m
i
crochip pic 16c77
.
and two stepping m
o
tors. The robotic m
odule has a
transceiver module AM-RTD
from
abacom
for serial data comm
unication. The robot
can estim
a
te its m
oved distan
ce and
directi
on from
the initial position using encoders.
Note that only relative dist
ance and direction from a partic
ular point can be recognized





System Architechture Follow
er Robot: 2 (Lig
ht Chaser)







Our Light C
h
aser robot requires an
internal locomotor System
driv
en by two electric m
o
tors that
draws its en
ergy from
a rechargeable battery. It
is com
p
leted by three wheels out of which two
wheels are driven by th
e elec
tric motor and the
third one a
c
ts as a dif
f
e
rential whe
e
l tha
t
helps

the robo
t to turn 360
deg
r
ees. It consists of two
light sensitive cells cons
isting of light dependent
resistors that provide the electr
onic creature with direc
tional light sensing behavior. The two eyes
are fitted with LDRs placed at adequ
a
te dis
t
ance fr
om
each other. If the light im
pinges late
rally ,
then one of the two LD
R’s is subj
e
c
ted to less light than
th
e othe
r,
th
e result is th
at the m
o
tors

electronics circu
i
try ex
ecutes
a ch
ange of course which lasts unt
il the light im
pinges and from
the front and both LDR’s are subjected to sam
e
amount of light.
The Rescuer Robot:




The rescuer robot consists of
four DC m
o
tors operated by re
lays which act as a switch.
The voltage regulator I used to provide 6V
to the relay. The motors are supplied 12V

from
a DC Source. The robot is m
a
nually
operated using a remote (RF). The Robotic
module can push upto 10Kgs and lift upto 1Kg.

Functional Analysis:


Lead Robot
:



Obstacle Avoidance and Navigation:


We have used a IC 555 tim
er
(Astable Multivibrator) fo
r modulating the IR LED
to
produce IR at frequency of 38
KHz. W
h
en an obstacle is de
tected by any one of the IR
Sensors, the IR reflected from
the surface of the obstacle is
receiv
e
d by the IR rece
ive
r

(capacity of receiving IR
up to 38 KHz). It
gives a pulse to micr
ocontroller (AT89C51).
The algo
rith
m
has been written for the followi
ng behavior of the r
obo
t.
W
h
en it de
tects
an obstacle in the right, the left m
o
tor stops a
nd the right m
o
tor continues to rotate, so the

it turns tow
a
rds the left. W
h
en the obstacle
is detected in the left, the robot m
o
ves
towards the
righ
t us
ing
the sam
e
princ
i
pal (T
a
ngentia
l Tur
n
ing Prin
cip
l
e). In case of
obstacle in both the sides, the m
ove continue
s to move forwa
r
d. In case of an obstacle on
all the three sides, the robot
gets back (Reverse Gear F
acility) and sear
ches in all

directions and finds a path to m
ove.

Metal Detection and Ga
s Leakage Detection.


The goal of
our robo
t is f
o
r it to b
e
able to
detect land m
i
nes. Obviously, som
e
sort of

sensor m
u
st be inco
rpora
t
ed into the
system
th
at is ab
le to f
i
nd land
m
i
nes. W
e
chose to
initially u
s
e a simple m
e
tal detecto
r
as a
m
i
ne detec
ting device.



The m
e
tal detector used on our robots ut
ilizes a fairly sim
p
le de
sign. The m
a
in
com
ponent of the m
e
tal detector is the induc
tion coil. W
h
en a m
e
tal object com
e
s ne
at to
the coil, it'
s
inductance is a
ffected. This is turn alters
the oscillation frequency of the
circu
i
try. The alteration can be
detected, and act as a "st
op" comm
and for
the robot.
Heat detecto
r
s use a set
of tem
p
erature-sen
s
itive resis
t
ors

ca
lled the
r
m
i
stors tha
t
de
cre
a
se

in resis
t
ance as the temperatu
r
e rises. One th
erm
i
stor is sealed and protected from the
surrounding tem
p
erature while the other is
exposed. A sharp increa
se in tem
p
erature
reduces the resistance in the exposed therm
i
st
or, which allows a large current to activate
the detecto
r
'
s
alarm
.

This 230Vac m
a
ins powered gas leakage sen
s
or will detect propan
e
, butane (LP
G
) and
natural gas.
An onboar
d
buzzer an
d LED warn when gas
is present.
It uses a PIC
m
i
crocontro
lle
r and high
quality sens
or. It
f
o
rm
s a part of
the multiple sen
s
or system
.
Environment Monotoring:

The CCD photosites accom
p
lish their task
of sensing incom
i
ng light through the
photoelectric effect, which is a characteriza
tion of the action of certain m
a
terials to
release an electron when
hit with a photon of
light. The electrons em
itted within the CCD
are fenced within nonconductive
boundaries, so that they rem
a
in within the
area of the

photon strik
e
. As long as ligh
t
is allowe
d to
im
pinge on a photos
ite, electrons will
accum
u
late in that pix
e
l. W
h
en the source of
light is extin
guished (e.g
., the shutter is

closed),
sim
p
le electronic circuitry and a m
i
cr
oprocesso
r or com
puter are used to
un
load
the CCD array, count th
e electrons in each pixe
l, and process the resulting data in
to
an
im
age on a video m
oni
tor or other output m
e
dia. So victim
s can be identified during
disasters and the output m
e
dia is
m
a
de available to the rescuers.


Commerc
ia
l Applicat
io
ns and Marketab
ility o
f
Our Multipurpose Robot:



Due to its is multif
unctio
nal capab
ilities, it
is h
i
gh
ly usef
ul to the presen
t s
o
ciety.

Autom
a
tic fire detection system
s, when com
b
ined with other elem
ents of an em
ergency
response and evacuation plan, can signifi
cantly reduce property dam
a
ge, personal
injur
i
es, and
loss of
lif
e f
r
om
f
i
re in the wo
rkplace. Their m
a
in function is to quickly
identify a developing fire and alert buildi
ng occupants and em
ergency response personnel
before extensive dam
a
ge occurs. Autom
a
tic fire detection system
s do this by using
electronic sensors to detect th
e sm
oke
, heat, or flam
es from
a fire and providing an early
warning.

Application
s
:
Lead Robot
i
c System (Multipurpose)

Reverse Gear Facilities.
Metal Detection, Gas Leakage Detection
Highly Sens
itiv
e Circu
it Breaking
Environm
ent Monitoring.
Object Avoidance and P
a
th Identification
Heat/Flam
e
Detection
Alarm
i
ng System


Follower 1 :( RF Module)
Form
ation Approach using RF ID

Follower 2: (Light Chaser Module)
Form
ation Approach and Coordination using Light

Follower 3: (Rescue Robot)
Rescue Operations under abnorm
a
l conditions



Being an autonom
ous vehicle, the above ta
sks are im
possible without object avoidance
and navigations techniques. W
e
have develope
d an alogorithm to increase the efficiency
in understanding the environm
ent with sharp sensors. W
e
ha
ve tested out robot in
perform
i
ng the above m
e
ntioned task and the
results were satisfactory. W
e
are working
on m
echanical considerations such as diffe
rential steering, angle of cam
ber, etc…., we
are currently working on im
proving its
scope for real life situations.


Formation Control
:

In this pap
e
r, only in
accurate s
e
nsor info
rmation is av
ailab
l
e for coordinating
the
for
m
ation of m
u
ltiple m
obile robots.
Local senso
r
inform
ation m
eans that all the
robots have only local sensors that cannot
accurately m
easure absolute dis
t
an
ces and dire
ctions of objects. That is, all the sensors
have lim
itation on the range and uncertainty in
the values
. It is pos
sible to recog
n
ize
types of the detected object by using the fre
quency of the light as
robot ID. Each robot
has a unique fixed sensor frequency. If a robot
receives its own fr
equency, it recognizes
the object is static. If a robot receives a fr
equency of another robot
, it can recognize the
ID of the detected r
obot.
The surfaces of all robots are c
overed
withblack tapes to
prev
ent
reflection of light. Therefore,
a ro
bot canno
t receive its

own frequency reflected by
another robot.

Computational Modeling of Coordinated Robotic Systems:


A m
a
in goal of com
putational robotics is to
automatically synthesize robot motions to
achieve a given task. This course discusses ge
om
etric and alg
o
rithm
i
c iss
u
es that aris
e in
such an endeavour. Examples of a few sub-prob
lem
s
that need to be addressed are: how
can a robo
t plan its own collision
-
free m
o
tions?
How does it grasp a given object? How
does one account for uncertainty?
The course
employs a broad range of tools from

com
putational geom
etry, m
echanics and control
theory. The behavior is plotted with Petri
Net Models.



Behavior of Robotic Systems:



.

Conclusion:
Being an autonom
ous vehicle, the above
tasks are im
possible without object
avoidance and navigations techniques. W
e
ha
ve
developed an algorithm to increase the
efficiency in understanding the environm
ent with
sharp sensors. W
e
have tested out robot
in perform
i
ng the above m
e
ntioned task and
the results were satisfactory and we could
satisfy all our objectives to a greater extent
. We are working on considerations such as
differential steering, angle of
cam
ber, etc…., w
e
are currently working on im
proving its
scope f
o
r r
eal lif
e
situ
ations. The
m
u
lti - ro
bot coord
i
n
a
tion ha
s pr
oved to be quite
successful. We have overcom
e
the disadv
antages of RF Comm
unication with a new

concept of u
s
ing light to
es
tablish a communication between

m
u
ltiple ro
botic m
odul
es in

a dom
estic environm
ent. We have also veri
fied the perform
a
nce of the Rescuer Robot.


Note: W
e
c
onducted real tim
e
experim
e
nts
and have taken videos during our research.
Most of our objectives have been satisfied. I
have attached the photos of the Lead and the
other two follower robotic syst
em
s, if needed I shall send the videos. I’m
looking forward
to presen
t my ideas a
l
on
g with the w
o
rki
ng m
odel during the even
t. All the thre
e
working m
o
dels shall be available du
ring presentation at IIT – Bom
b
ay.

References
:


Form
at
i
on Ap
p
r
oac
h
fo
r
M
obi
l
e

R
o
b
o
t
s
wi
t
h
Inacc
urat
e

Se
n
s
or
I
n
f
o
r
m
atio
n
,
G
unh
ee
K
i
m
*
, D
o
o

Yo
ng

Lee*
*, an
d K
y
u
ngn
o Lee*
*


Victim
detection with Infrare
d
Cam
e
ra in a
"Rescue R
o
bot
"

Saeed M
o
ra
di,
Electrical Engi
neeri
n
g De
part
ment, Sha
r
if
Unive
r
sity of Te
chnology.

Val
u
e
-
B
a
sed

C
o
m
m
uni
cat
i
on Prese
r
vat
i
o
n fo
r
Mo
b
ile Robo
t
Team
s
Matth
ew Powers an
d

Tuc
k
er Balc
h

#

K
i
m
,
G
u
nh
ee
KI
ST


For
m
ati
o
n appr
o
a
ch

o
f
Mo
b
ile R
o
bo
t
s
., CMU Robo
tics.
#

[1
]
D. J. Stilwell an
d

J.
S. Bay
,
“To
w
ard th
e
d
e
v
e
l
o
p
m
en
t of a m
a
terial tran
sport system

u
s
ing

swarm
s
of ant
-
like robots,”
Proc.
of IEEE
In
t
e
rnational C
o
nf.
on R
obotics and
Autom
a
tion,
Min
n
e
ap
o
lis, Min
n
e
so
ta, U
S
A
,

May 2
-
6
,

pp.
7
66-
771
, 19
93
.
#

[2] T. Balch a
n
d R. C.
Arkin,
“Behavi
o
r-base
d
form
ation c
ont
rol
for m
u
ltirobot team
s,” IEEE
Tr
ans.
on
Robo
tics and

Au
tomatio
n
,
vo
l. 14. no
. 6.,
pp
.
926
-93
9
,
1
998
.
#

[3]
H. Yam
a
gu
chi
,


A
da
pt
i
v
e
fo
rm
ati
on
c
o
nt
rol
fo
r di
st
ri
b
u
t
e
d
a
u
t
o
nom
ous

m
obi
l
e

ro
bot
g
r
o
u
p
s,”
Pro
c
.
of
IE
EE
Int
e
r
n
at
i
o
nal
C
o
n
f
.
o
n
R
o
b
o
t
i
c
s an
d
A
u
t
o
m
a
t
i
on,
Al
b
u
que
r
que
,
New
M
e
x
i
co,
US
A,
A
p
r
.
20
-25
,
pp
.
23
00-
230
5, 1
9
9
7
.
#

[4]
L
.
E.
Pa
rke
r
, “
D
esig
nin
g
c
ont
rol la
ws
for
coope
r
ative a
g
ent t
eam
s,” Proc. of
IEEE

In
tern
ation
a
l C
o
nf. on
R
o
bo
tics and

Au
to
m
a
tio
n
,
Mi
n
n
e
apolis, Min
n
e
so
ta, USA, May 2-6
,

pp
.

5
82-
587
, 19
93
.
#

[5
] J. P. Desai,
V.
Ku
m
a
r, and J. P.
Ostro
w
sk
i,
“Con
tro
lling
form
at
io
n
s
o
f
m
u
l
tip
le
m
o
b
ile
robo
ts,” Pro
c
.
o
f
th
e
IEEE In
t
e
rn
ation
a
l C
o
nf.
on
Robo
tics
an
d Au
to
m
a
tio
n
,

Leuv
en
, Bel
g
iu
m
,

May 1
6
-
2
0
,
pp. 286
4-2
869
,
19
98
.
#

[6
]
I. Su
zuk
i
an
d M.
Yam
a
sh
ita, “Distribu
t
ed
an
on
ym
o
u
s

m
o
b
ile r
o
bo
ts :
fo
r
m
atio
n
of

geo
m
etr
i
c
p
a
tter
n
s,”
SIAM
Jou
r
n
a
l o
n
Co
m
p
u
tin
g
,

vol.
28
. no
. 4
.
, p
p
.
134
7-1
363
, 19
99
.
#

(7
) DES
A
I

A
N
D
VIJ
A
Y KU
M
A
R
#

IC
C
A
S
2
0
0
1

In
t
e
rnat
i
o
nal
C
o
nfe
r
ence
o
n
C
o
nt
r
o
l
,

Aut
o
m
a
ti
on a
n
d Sy
st
e
m
.

Proceedi
ngs
of the
IEEE
Internati
onal C
o
nfe
r
ence
on R
o
bot
i
cs and
Aut
o
mation,
Albuque
rque, NM, Apr.
2
1
-2
7, 1
997
, pp
. 12
83-
128
8


The
Guide Ca
ne A C
o
m
puterized T
r
avel
Ai
d fo
r t
h
e
Active
Gui
d
ance
of Bl
ind
Pede
strians

b
y
Joh
a
nn
Boren
s
tein
and
Iwan

Ulrich,
Un
i
v
ersity of Mich
ig
an
.

Design and Implementation of Co
ordinated Multipurpose Robotic
System with RF and Ligh
t Communication Channels.

Lead Robot
:






Follow
er Robot 2: RF Module and Follower Robot 3: Rescuer Module




Follow
er Robot: Light Chase 3 (Coordination
Model)




Experimental Verification:

Experiment 1: Performance of Multipurpose Robotic Module

Metal/Flux Detection
Flame/Heat Detecttion
Reverse Gear Facility,
Obstacle Avoi
dance a
nd Navi
gation.
Environm
ent Monitori
ng.


Emergency Alarm/Buzzing

Experim
e
nt 2:

Beha
vior of Robotic S
y
stem
s in a unstruct
u
red environm
ent.
Wall Following, Edge Detection …

Experim
e
nt 3:

Form
ation with R
F
C
o
m
m
unication Module.

Experim
e
nt 4:

Form
ation with Li
ght (360 De
gree)

Experim
e
nt 5:

Form
ation using light channels
when Lead Robo
t is static.

Experim
e
nt 6:

The performance of Light Chaser Robot
under various intensities of Light

Experim
e
nt 7:

The performance of the Rescue Robot