Intern
ati
o
n
a
l Jo
urn
a
l
o
f
R
o
botics
a
nd Au
tom
a
tion
(I
JR
A)
V
o
l.
3, N
o
. 3
,
Sep
t
em
b
e
r
2014
, pp
. 16
1
~
16
7
I
S
SN
: 208
9-4
8
5
6
1
61
Jo
urn
a
l
h
o
me
pa
ge
: h
ttp
://iaesjo
u
r
na
l.com/
o
n
lin
e/ind
e
x.ph
p
/
IJRA
The Novel of Six axes
Robotic
Arm for Industrial Applications
Ra
jendr
a Ap
a
r
na
thi*
, Ved Vy
as Dw
ivedi
*
*
* Depart
em
entof
El
ectr
i
c
a
l
Engin
eering
,
C
.
U.
S
h
ah Univers
i
t
y
,
W
a
dhwan Cit
y
,
Gujarat
,
Ind
i
a
** P
r
o-Vice
Cha
n
cel
lor,
C.
U. S
h
ah Univers
i
t
y
,
W
a
dhwancit
y
,
G
u
jara
t,
India
Article Info
A
B
STRAC
T
Article histo
r
y:
Received Oct 23, 2013
Rev
i
sed
May 29
, 20
14
Accepted
Jun 20, 2014
Automation tech
nolog
y
is widely
accept
ed and
rapidly
growing
techno
lo
g
y
in the f
i
eld
of core and man
y
other i
ndustries. A
n
y
one can
obser
ve that due
to
th
ese problems many
industries are turning
towards automaton. When
searching for pr
oblem of labor manpower
in mi
ddle-case industr
ies, we came
to know about
man
y
oth
e
r th
in
gs like produ
ction, speed of
manufactur
ing
and quality
of
the product ar
e necessar
y
in the curr
ent scen
ario. These
param
e
ters ar
e not being well
m
a
intained in
incorporat
e in
dustries with
manual manufacturing processes instead
of using automatic s
y
stem. Our
objective is to solve these
problems by
eff
i
cient
use
of different technolog
ies
for making an
industr
y
fu
lly
or
partia
lly
au
tomated
.
B
y
using
technolog
ies
we
can
tr
y
to
solve
or reduce the effects
of above problems.
Keyword:
ARM base
d cortex-xx se
ries
Aut
o
m
a
ti
on t
e
chn
o
l
o
gy
Indu
strial m
u
lt
i ax
is Robo
t
Pro
g
r
am
m
a
bl
e C
ont
r
o
l
l
e
r a
r
m
s
Copyright ©
201
4 Institut
e
o
f
Ad
vanced
Engin
eer
ing and S
c
i
e
nce.
All rights re
se
rve
d
.
Co
rresp
ond
i
ng
Autho
r
:
Raj
e
nd
ra Ap
arn
a
th
i
Depa
rtem
ent of Elect
ri
cal
E
n
gi
nee
r
i
n
g,
C. U.
S
h
a
h
U
n
i
v
ersity,
Wadhwancity, Guja
rat
State, India.
Em
a
il: raj
e
nd
raap
arn
a
th
i
@
live.co
m
1.
INTRODUCTION
Th
is Pro
j
ect is b
a
sed
o
n
an
Au
to
m
a
tio
n
Tech
no
log
y
. An
au
to
m
a
tio
n
o
r
au
to
m
a
tic
co
n
t
ro
l syste
m
is a
syste
m
wh
ich
is
m
a
d
e
u
p
o
f
d
i
fferen
t
con
t
ro
lling
and
p
hysical p
a
rts. Th
ese all th
e th
in
g
s
are
d
i
scu
s
sed
in
f
o
llow
i
ng
sectio
n
s
[1
]
Int
r
od
uct
i
o
n t
o
A
u
t
o
mat
i
o
n
Tech
nol
ogy
: -
Aut
o
m
a
ti
on
or
aut
o
m
a
t
i
c
cont
r
o
l
i
s
t
h
e us
e of
va
ri
o
u
s
cont
rol
syste
m
s
for operati
n
g equipm
en
t such as
m
achinery,
processes
in
fact
o
r
i
e
s. T
h
e bi
ggest
be
ne
fi
t
o
f
au
to
m
a
tio
n
is th
at it sav
e
s labo
r;
h
o
wev
e
r, it is also
u
s
ed
to
sav
e
en
erg
y
and
m
a
terials an
d
to
im
p
r
o
v
e
quality,
accuracy a
nd pre
c
ision. T
h
e term
automation, ins
p
ired
by the ea
r
lier word aut
o
m
a
tic (com
ing from
aut
o
m
a
t
on),
w
a
s n
o
t
wi
del
y
use
d
be
f
o
re
19
47
, w
h
e
n
Gene
ral
M
o
t
o
rs est
a
bl
i
s
he
d t
h
e a
u
t
o
m
a
ti
on
depa
rt
m
e
nt
.
It
was
du
ri
n
g
t
h
i
s
t
i
m
e
t
h
at
i
ndust
r
y
was
rap
i
dl
y
ado
p
t
i
n
g
f
eedbac
k
c
o
nt
ro
l
l
e
rs, w
h
i
c
h
w
e
re i
n
t
r
od
uce
d
i
n
t
h
e
19
3
0
s. A
u
t
o
m
a
t
i
on has
bee
n
achi
e
ve
d b
y
vari
o
u
s m
e
ans including
m
echanical
, h
ydr
au
lic, pn
eu
m
a
t
i
c,
electrical, and electronic a
n
d com
puter
s,
u
s
ually in
com
b
in
atio
n
.
Com
p
l
i
cated syste
m
s, such a
s
m
odern
facto
r
ies, airp
l
a
n
e
s and
sh
ip
s
typ
i
cally
use all these c
o
m
b
ined techniques
[2].
In
tro
d
u
c
tion
to Ro
bo
t:-
A ro
bot is a
m
ech
an
ical in
tell
ig
en
t ag
en
t wh
ich
can p
e
rfo
r
m
task
s o
n
its o
w
n
,
o
r
with
gu
id
ance. In
p
r
actice
a rob
o
t
is u
s
u
a
lly an
electro-mechanical m
a
chine
wh
i
c
h i
s
gui
ded
by
co
m
put
er
and
el
ect
ro
ni
c
pr
o
g
ram
m
i
ng.
R
o
b
o
t
s
ca
n
be
aut
o
nom
ous
o
r
sem
i
-aut
onom
ous
[
1
]
-
[
2
]
.
Fi
el
d of
Ro
b
o
t
i
c
s
[1]
:
-R
o
bot
i
c
s i
s
t
h
e bra
n
c
h
o
f
t
ech
nol
og
y
t
h
at
deal
s wi
t
h
t
h
e desi
gn
,
con
s
t
r
uct
i
o
n
,
o
p
e
ration
,
st
ructu
r
al
d
i
spo
s
itio
n, m
a
n
u
f
act
u
r
e and
ap
p
licatio
n of
robo
ts. R
o
bo
tics is related
to th
e sciences of
electronics
, e
n
ginee
r
ing, m
echanics
,
a
n
d software.
Robotic Arm
[1
]-
[2
]:-
A
rob
o
tic ar
m is a
r
o
b
o
tic m
a
n
i
p
u
l
ato
r
, usu
a
lly p
r
o
g
ramm
ab
le,
with
si
m
ilar
fu
nct
i
o
ns t
o
a hum
an arm
.
The l
i
nks
of s
u
c
h
a m
a
ni
pul
at
or are connected by joints
allowing
eith
er ro
t
a
tio
n
a
l
m
o
tion
(s
uch a
s
in a
n
articula
t
ed ro
bot
) or
translational (lin
ear)
dis
p
lacement. T
h
e li
nks
of the m
a
nipulator
can be c
onsi
d
e
r
ed t
o
f
o
rm
a
ki
nem
a
t
i
c
chain. T
h
e b
u
si
nes
s
end
of t
h
e
ki
nem
a
ti
c chai
n of t
h
e m
a
ni
pul
at
or i
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
089
-48
56
IJR
A
V
o
l
.
3, N
o
. 3,
Se
pt
em
ber 20
1
4
:
16
1 –
1
6
7
16
2
called the end
effectors a
nd it
is an
al
og
o
u
s t
o
t
h
e h
u
m
a
n h
a
nd
. The e
n
d e
ffect
o
r
s ca
n be
desi
g
n
ed t
o
pe
rf
orm
any
desi
re
d t
a
s
k
suc
h
as
wel
d
i
ng,
gri
ppi
ng
,
spi
n
ni
n
g
et
c.,
depe
n
d
i
n
g o
n
t
h
e ap
pl
i
cat
i
on.
For e
x
am
pl
e rob
o
t
arm
s
in
au
to
mo
tiv
e assem
b
ly
p
e
rform
a
v
a
riety o
f
ta
sks such as
wel
d
i
n
g an
d pa
rt
s rot
a
t
i
on an
d pl
ac
em
ent
d
u
ring
assem
b
lyMu
ltifu
n
c
tion
a
l Rob
o
t: The Mu
ltifun
ction
a
l
Rob
o
tic
Arm
is si
milar t
o
th
at
o
f
ro
bo
t
i
c arm
.
We
have a
d
de
d r
o
bot
i
c
ca
r a
t
base s
o
ou
r
pr
o
j
ect
can al
s
o
w
o
rk
as t
r
a
n
spo
r
t
a
t
i
o
n
de
vi
ce. A
n
d al
so
we are
addi
ng
an
ap
pl
i
cat
i
on o
f
c
o
nv
ey
er bel
t
.
2.
TERMS BEING
USE
D
IN
ROBOTICS
N
u
m
b
ers of
ax
es
– two a
x
es
are re
qui
red to reach a
n
y poin
t in a pla
n
e; three a
x
es are
requi
red t
o
reach any
point in s
p
ace. T
o
fully cont
rol
the orientation
of the
e
n
d of t
h
e
arm
(i.e. t
h
e
wrist) three m
o
re axes
(y
aw,
pi
t
c
h
,
a
n
d
r
o
l
l
)
a
r
e r
e
qui
red
.
S
o
m
e
desi
gns
(e.g. th
e SC
ARA
robo
t) trad
e li
mita
tio
n
s
in
m
o
t
i
o
n
possibilities for c
o
st, spee
d, a
n
d accuracy [2].
Deg
r
ees
o
f
f
r
e
e
dom
whi
c
h i
s
us
ual
l
y
t
h
e s
a
m
e
as t
h
e
nu
m
b
er of
axes
,
Wor
k
i
n
g e
n
v
e
l
ope
– t
h
e
regi
on
of s
p
ac
e a robot can
reach, Kinem
a
tics
– the actua
l arrangem
ent of ri
gid m
e
mbers a
n
d
joi
n
ts
in the
robo
t, wh
ich
d
e
term
in
es th
e robo
t's p
o
s
sib
l
e
m
o
tio
n
s
.
Classes o
f
rob
o
t
k
i
n
e
m
a
tic
s in
clud
e arti
cu
lated,
Car
t
esian
,
p
a
rallel an
d
SCA
R
A
.
Car
r
y
ing
cap
acity o
r
p
a
yload
–
h
o
w
m
u
ch w
e
igh
t
a
ro
bo
t can
lif
t
[
1
]-[2
].
Spee
d
– h
o
w
f
a
st
t
h
e rob
o
t
can p
o
si
t
i
on t
h
e end o
f
i
t
s
arm
.
Thi
s
m
a
y
be defi
ned i
n
t
e
rm
s of t
h
e
angular or line
a
r s
p
eed of eac
h a
x
is or as
a c
o
m
pound sp
ee
d i.e. the s
p
ee
d of the e
n
d
of t
h
e arm
whe
n
a
ll axes
are m
oving [2]
.
Acceleration
- how quickly
an
axis
can accelerate. Since this is a lim
i
ting factor a
robot
m
a
y not be
abl
e
t
o
reac
h i
t
s
speci
fi
ed m
a
xi
m
u
m
speed f
o
r
m
ovem
e
nt
s
ove
r a
sh
o
r
t
di
st
ance
or a
c
o
m
p
l
e
x pat
h
req
u
i
r
i
n
g
fre
que
nt
c
h
an
g
e
s o
f
di
rect
i
o
n
[2]
.
Accur
a
cy
–
how closely a robot ca
n reach a
comm
anded
position. Whe
n
the
ab
solute
position of the
robot is m
easured a
n
d com
p
a
r
ed t
o
th
e co
mman
d
e
d po
sitio
n
t
h
e erro
r is
a
m
easure of a
ccuracy.
Acc
u
racy can
be im
pro
v
e
d
w
i
t
h
ext
e
rnal
se
nsi
n
g f
o
r exam
pl
e a vi
si
on
sy
ste
m
or Infra-Red. See
robot calibration.
Accuracy
can
v
a
ry
with
sp
eed and
po
sitio
n with
i
n
th
e
work
i
n
g
en
v
e
l
o
p
e
and
with
paylo
a
d
(see
com
p
l
i
an
ce) [3
].
Rep
e
a
t
ab
ility -
h
o
w well th
e rob
o
t
will retu
rn
to
a progra
m
m
e
d
p
o
s
itio
n
.
Th
is is n
o
t
th
e sam
e
as
accuracy. It may be that
whe
n
tol
d
to
go t
o
a certain
X-Y-Z pos
ition that
it gets only to
within 1
mm
of t
h
at
position. This would
be its accuracy which
may be i
m
proved
by calibrati
on. But if that position is taught int
o
cont
roller m
e
m
o
ry and each tim
e
it is sent there it re
turns to within 0.1m
m
of
the taught position then the
rep
eatab
ility will b
e
with
i
n
0
.
1
m
m
[3
].
Accuracy and rep
eatab
ility are d
i
fferen
t measu
r
es. Repeatab
ility
is u
s
u
a
lly th
e
m
o
st i
m
p
o
r
tan
t
criterio
n
for a robo
t an
d
is si
milar to
th
e co
n
cep
t
o
f
'p
recis
i
on'
in
m
easure
m
ent - see
Acc
u
racy and prec
ision.
ISO 9283
sets
out a m
e
thod
whe
r
e
b
y bo
th
accuracy a
n
d repeatab
ility can be
m
easured. Typically a robot is
sent to a ta
ught position a num
b
er
of ti
m
e
s and the e
r
ror is m
easured
at each ret
u
rn
to the
position afte
r
v
i
sitin
g
4
o
t
h
e
r p
o
s
ition
s
. Repeatab
ility
is th
en
qu
an
tified
usin
g
th
e
stand
a
rd
d
e
v
i
ation
o
f
th
o
s
e sam
p
les in
all
three
dim
e
nsions. A typical robot ca
n,
of c
o
urse m
a
ke a
positional e
r
ror e
x
ceedi
n
g that and that c
ould be
a
p
r
ob
lem
fo
r the pro
cess. M
o
reo
v
e
r t
h
e rep
e
atab
ility is d
i
ffe
ren
t
i
n
d
i
fferent p
a
rts
of th
e wo
rk
ing
env
e
lope and
also cha
n
ges with spee
d and payload.
ISO
9283 specifies that accuracy an
d re
peatability shoul
d
be m
e
a
s
ure
d
at m
a
x
i
m
u
m
s
p
eed an
d at max
i
m
u
m
p
a
ylo
a
d
.
B
u
t th
is resu
lts in p
e
ssimistic v
a
lu
es
whereas t
h
e robo
t cou
l
d
be m
u
ch m
o
re accurate a
nd
re
peatable at light loads a
n
d speeds. Re
peatabilit
y in an
industrial process is also
subject to the accuracy
of the end effector, for exam
ple a
gri
ppe
r, an
d even to the design
of the '
f
ingers'
that
m
a
t
c
h t
h
e
gri
p
per
t
o
t
h
e
o
b
je
ct
bei
n
g
g
r
asp
e
d.
Fo
r e
x
am
ple if a
robot
picks a
sc
rew by its h
ead th
e
screw
coul
d
be at
a
r
a
nd
om
angl
e.
A s
u
b
s
eq
ue
nt
at
t
e
m
p
t
t
o
i
n
se
rt the sc
rew int
o
a
hole could
easily fail. The
s
e and
sim
i
l
a
r scenari
o
s ca
n
be i
m
prove
d
wi
t
h
'
l
ead
-i
ns'
e.g
.
by
m
a
ki
n
g
t
h
e
ent
r
an
ce t
o
t
h
e
h
o
l
e
t
a
pere
d
[2]
,
[
4
]
.
Motion control
–
fo
r so
m
e
a
p
p
lication
s
, such
as sim
p
le
pick-and-place
asse
m
b
ly, the robot nee
d
merely retu
rn
rep
eatab
ility to
a li
m
i
ted
nu
mb
er
of pr
e-taugh
t po
sitio
ns. Fo
r m
o
re sop
h
i
sticated
ap
p
licatio
n
s
,
suc
h
as wel
d
i
n
g an
d fi
ni
s
h
i
n
g (s
pray
pai
n
t
i
ng
), m
o
t
i
on
m
u
st
be co
nt
i
n
u
ousl
y
co
nt
r
o
l
l
e
d t
o
fol
l
o
w a pat
h
i
n
space, with controlled
orie
nt
ation a
n
d vel
o
ci
ty [5].
Power sourc
e
–
s
o
m
e
robots
use electric m
o
tors,
ot
hers
use
hy
dra
u
l
i
c
act
uat
o
rs.
Th
e f
o
rm
er are
faster, the latter are
stronge
r
a
n
d adva
nt
age
o
us i
n
a
ppl
i
cat
i
o
ns s
u
c
h
as
sp
ra
y
pai
n
t
i
n
g,
w
h
ere a s
p
a
r
k
co
u
l
d set
of
f an e
x
pl
osi
o
n;
ho
we
ver, l
o
w i
n
t
e
rnal
ai
r-
press
u
ri
zat
i
on
of t
h
e arm
can
pre
v
e
n
t
i
ngre
s
s
of fl
am
m
a
bl
e
vap
o
r
s
as well as
othe
r c
ontam
inants [6].
Drive
– som
e
r
o
bots conne
ct electric
m
o
tors to the jo
ints via
gears; ot
hers
co
nn
ect th
e
mo
tor to
the
j
o
i
n
t d
i
rectly (d
irect d
r
i
v
e). Using
g
e
ars resu
lts in
m
easu
r
ab
le 'b
ack
lash
' wh
ich
is free
m
o
v
e
m
e
n
t
in
a
n
ax
is.
Sm
al
l
e
r ro
bot
arm
s
freq
u
ent
l
y
em
pl
oy
hi
g
h
s
p
eed
, l
o
w t
o
r
q
ue DC
m
o
t
o
rs
, w
h
i
c
h ge
neral
l
y
re
qui
re
hi
g
h
geari
n
g
rat
i
o
s;
t
h
i
s
has
t
h
e
di
s
a
dva
nt
age
o
f
b
ackl
a
sh
. I
n
suc
h
cases
t
h
e
ha
r
m
oni
c dri
v
e
i
s
oft
e
n
used
[
6
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
RA I
S
SN
:
208
9-4
8
5
6
The N
o
vel
of
Si
x axes
Ro
b
o
t
i
c
Arm f
o
r
In
d
u
st
ri
al
Ap
pl
i
c
at
i
o
ns (
R
aj
en
dra
A
par
n
a
t
h
i
)
16
3
Co
m
p
lian
ce
- th
is is a m
easu
r
e o
f
th
e am
o
unt in
ang
l
e or
d
i
stan
ce th
at a
rob
o
t
ax
is
will
mo
v
e
wh
en
a
force is a
p
plied to it. Because of
com
p
lianc
e when a robot
goes to a
po
sition ca
rrying its m
a
xi
m
u
m
payload it
will b
e
at a
positio
n
sligh
tly lo
wer th
an
wh
en
it is carryin
g no
p
a
ylo
a
d. C
o
m
p
lian
ce can also
b
e
respon
sib
l
e
for ove
rs
hoot whe
n
ca
rryin
g high payloads
i
n
whic
h
case
a
cceler
ation would nee
d
to
be reduce
d.
3.
PROP
OSE
D
AR
CHITE
C
T
URE OF PR
O
J
ECT
Block Di
agram
:- Th
e
Indu
strial
m
u
lti ax
is Ro
bo
t Con
s
ist
o
f
Man
i
pu
lato
r
Or m
ech
an
ical Stru
cture
and
A
n
ot
he
r S
y
st
em
i
s
C
ont
r
o
l
l
e
r [
7
]
.
Fi
gu
re
1.
Pr
o
p
o
se
d B
l
oc
k
Di
agram
Th
e b
l
o
c
k
d
i
ag
ram
o
f
ind
u
s
trial
m
u
lti ax
i
s
robo
t th
at is sh
own
in
abo
v
e
figu
re t
h
at co
n
s
ist of
manipulator
with end e
ffecte
r
, c
o
ntroller as
comm
ande
r, powe
r
s
u
pply and
fee
d
back mechanism
or sens
or
feedb
a
ck
[7
]. Wh
en
th
ese
wh
o
l
e system
s c
o
m
b
in
e to
g
e
ther th
en
it wo
rks as au
to
m
a
tic
in
du
strial
m
u
lti ax
is
ro
b
o
t
w
h
i
c
h
i
s
neede
d
i
n
m
a
ny
i
n
d
u
st
ri
es t
o
do
su
ch
ki
nd
o
f
pr
ocess.
Mechanic
a
l Part:-
T
h
e
robot'
s m
a
nipulative
arm
is the
m
e
c
h
ani
cal
unit.
T
h
is m
echanical unit is also
com
p
ri
sed
of
a fa
bri
cat
ed
st
ruct
ural
fram
e
wi
t
h
p
r
o
v
i
s
i
o
ns
fo
r s
u
pp
ort
i
ng m
echani
c
a
l
l
i
nkage
an
d
joi
n
t
s
,
gui
des, act
uat
o
rs (l
i
n
ear
or r
o
t
a
ry
), co
nt
r
o
l
v
a
l
v
es, an
d sens
ors
.
The p
h
y
s
i
cal
dim
e
nsi
ons
, desi
g
n
, an
d w
e
i
ght
-
carrying
ab
ility
d
e
p
e
nd
on
ap
plicatio
n
req
u
i
remen
t
s [3
].
Mechanic
a
l Consider
ation for
robots
[
3
]
:
Wo
r
k
En
vel
o
p
:
The set
of p
o
i
n
t
s
R
e
prese
n
t
i
ng t
h
e m
a
xim
u
m
ext
e
nt
or rea
c
h o
f
t
h
e r
o
b
o
t
arm
or w
o
r
k
i
n
g t
ool
in
all d
i
rection
.
Paylo
a
d
:
Th
e ab
ility to
carry,
co
n
tinuo
usly an
d satisfactorily g
i
v
e
n
m
a
x
i
mu
m
weig
h
t
at a g
i
v
e
n
sp
eed
.
Velo
city: Th
e
max
i
m
u
m
sp
eed
at
wh
ich
th
e
tip
of a
robo
t is cap
ab
le
o
f
m
o
v
i
n
g
at fu
ll exten
s
ion
,
exp
r
essed
i
n
inches
or m
illimeter per
second.
C
y
cl
e:
Tim
e
i
t
t
a
kes f
o
r t
h
e
ro
b
o
t
t
o
com
p
l
e
t
e
one cy
cl
e of
pi
cki
ng
u
p
a gi
ve
n
ob
jec
t
at
a gi
ven
h
e
i
ght
,
m
o
v
i
n
g
it to a
g
i
v
e
n
d
i
stan
ce
lo
wering
it, releasin
g it, and
retu
rn
ing
to th
e
startin
g
po
in
t.
Accuracy: A Robot’s
Ability t
o
position t
h
e end effecter at a
specified poi
nt
in
space upon
receivi
ng.
Rep
eatab
ility:
Th
e ab
ility o
f
a ro
bo
t to return con
s
isten
c
y to a prev
iou
s
ly hav
i
ng
attain
ed
th
at po
sitio
n.
Reso
lu
tion
:
The sm
al
lest in
cre
m
en
tal ch
ange in
p
o
sitio
n
t
h
at it
m
a
k
e
o
r
its con
t
ro
l system
can
m
easu
r
e.
The m
a
ni
pul
at
or
, whi
c
h i
s
t
h
e rob
o
t
’
s, co
ns
i
s
t
s
of
seg
m
en
ts j
o
in
ted
to
g
e
th
er with
ax
es cap
ab
le of
m
o
ti
on i
n
vari
ous
di
rect
i
o
n a
l
l
o
wi
n
g
t
h
e r
o
bot
t
o
per
f
o
rm
w
o
r
k
.
T
h
e e
n
d e
ffect
or
w
h
i
c
h i
s
a
g
r
i
p
pe
r
t
ool
,
a
special de
vices
, or
fixture atta
ched to t
h
e
robo
t’s arm
,
actu
a
lly p
e
rfo
r
m
s
th
e wo
rk
.
Man
i
pu
lato
r is am
ech
an
ical u
n
it th
at
prov
i
d
es m
o
tio
n
si
milar to
th
at of a
hum
an arm
.
Its
prim
ary
fun
c
tion
is to
prov
id
e t
h
e sp
ecific
m
o
tio
n
that will en
ab
le
t
h
e to
o
ling
at t
h
e end
of t
h
e arm
to
d
o
th
e
req
u
i
red
wo
rk
.
A r
o
bot
m
ovem
e
nt
can
be di
vi
de
d i
n
t
o
t
w
o
ge
ner
a
l
cat
ego
r
i
e
s:
Arm
and
b
ody
(s
ho
ul
de
r a
n
d
el
bo
w
)
m
o
ti
on a
nd
w
r
i
s
t
m
o
t
i
ons. T
h
e I
n
di
vi
d
u
al
j
o
i
n
t
m
o
t
i
ons
a
ssociated
with these categ
ori
e
s are refe
rre
d to as
degree
of free
dom
.
Each a
x
is
is equa
l to one
degree
of free
dom
.
Typically
industrial robots are e
q
uipped with
4
– 6
d
e
g
r
ees
o
f
freedo
m
. Th
e po
i
n
ts th
at
man
i
p
u
l
ator
b
e
n
d
s
, slid
es, or
ro
tates are called
jo
i
n
ts or positio
n
axes. M
a
ni
pul
at
or i
s
carri
e
d
out
u
s
i
n
g m
e
chani
cal
de
vi
ce
s, suc
h
as l
i
n
k
a
ge, gea
r
s, act
uat
o
rs, an
d fee
dbac
k
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
089
-48
56
IJR
A
V
o
l
.
3, N
o
. 3,
Se
pt
em
ber 20
1
4
:
16
1 –
1
6
7
16
4
devices
. Position a
x
es are c
a
lled as worl
d coordi
nates,
is id
en
tified
as Bein
g
fix
e
d
lo
catio
n
wit
h
in
th
e
m
a
nipulato
r
th
at serve
s
as
A
b
solute
fram
e
of
refe
re
nce [
8
]
.
Th
e x-ax
is trav
el
m
o
v
e
s th
e
m
a
n
i
p
u
l
ato
r
i
n
an
in-and
-out Mo
tio
n
.
The y-ax
is
m
o
tio
n cau
ses th
e
m
a
ni
pul
at
o
r
t
o
m
ove si
de-t
o-
si
de. T
h
e z a
x
i
s
m
o
t
i
on caus
e
s t
h
e m
a
ni
pul
at
or t
o
m
ove i
n
a
nd
u
p
a
n
d
-
Do
w
n
m
o
t
i
o
n
.
Th
e
mech
an
ical d
e
sig
n
of
a
r
obot
m
a
n
i
p
u
l
ator
r
e
lates d
i
r
ectly. To its wo
rk
en
v
e
l
o
p
e
and
m
o
t
i
o
n
Characteristics
.
End effecte
r
is the device that is
m
echani
cal
l
y
opene
d an
d cl
ose
d
. Act
a
s
t
h
e t
ool
- m
ount
i
n
g
pl
at
e. De
pen
d
i
ng
o
n
t
h
e t
y
pe
of
op
erat
i
o
n, c
o
n
v
e
n
t
i
onal
e
n
d Ef
fect
o
r
s are
equi
ppe
d
wi
t
h
vari
ous
de
vi
ces an
d
to
o
l
Attach
m
e
n
t
s, as fo
llows:
Fi
gu
re
2.
De
gr
ees o
f
free
dom
G
r
i
p
p
e
r
s
,
h
ooks, scoop
s, electr
o
m
a
g
n
e
ts,
v
a
cu
u
m
cu
p
s
, and A
d
h
e
siv
e
f
i
nger
s
fo
r
m
a
ter
i
al h
a
nd
lin
g,
Sp
ray g
u
n
for p
a
in
ting
.
Attach
m
e
n
t
s fo
r sp
ot an
d
arc
weld
i
n
g
and
arc cu
ttin
g
,
Power t
o
ols su
ch
as
d
r
ills, nut
dri
v
ers
,
and burrs. Special de
vices an
d fi
xt
u
r
es fo
r m
achi
n
i
ng a
nd assem
b
l
y
. M
easuri
n
g i
n
st
rum
e
nt
s, su
ch as
di
al
i
ndi
cat
ors
,
de
pt
h
ga
uge
s [
9
]
.
Electronics C
o
ntrol: -
T
h
e
co
nt
r
o
l
l
e
r i
s
a
c
o
m
m
uni
cat
i
on an
d
i
n
form
ation processing
Device t
h
at
initiates, ter
m
i
n
ates and c
o
ordinates the m
o
tions and seque
nces of a robot. It accepts
necessary inputs to the
robo
t and
pro
v
id
es th
e
Ou
t
p
ut d
r
iv
e
si
gn
als to
a con
t
ro
llin
g
m
o
to
r or act
u
a
tor to
C
o
rresp
ond
with
th
e robo
t
m
ovem
e
nt
s and o
u
t
s
i
d
e
wo
rl
d. S
h
ow
n i
n
fi
gu
re 3
bl
oc
k
d
i
agram
i
l
l
u
st
rat
e
s t
h
e m
a
ny
diffe
rent
p
a
rt
s o
f
ro
bot
C
ont
r
o
l
l
e
r.
Th
e heart
of t
h
e
cont
rol
l
e
r i
s
t
h
e com
put
er
an
d
its so
lid
-state Me
m
o
ry. Th
e inpu
t and o
u
t
p
u
t
sect
i
on
of
a c
o
nt
r
o
l
sy
st
em
must
pr
ovi
de a c
o
m
m
uni
cat
i
on
i
n
t
e
rface
bet
w
een t
h
e
r
o
bot
c
ont
rol
l
e
r C
o
m
put
e
r
an
d fo
llow
i
ng
p
a
r
t
s:
Feedback se
ns
ors
Pr
odu
ction
senso
r
s
Produ
ction
m
a
ch
in
e t
o
o
l
s
Teaching de
vice
Program
storage de
vices
Har
d
c
o
py
de
v
i
ces
Fi
gu
re
3.
El
ect
ro
ni
cs C
ont
r
o
l
B
l
ock
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
RA I
S
SN
:
208
9-4
8
5
6
The N
o
vel
of
Si
x axes
Ro
b
o
t
i
c
Arm f
o
r
In
d
u
st
ri
al
Ap
pl
i
c
at
i
o
ns (
R
aj
en
dra
A
par
n
a
t
h
i
)
16
5
The com
put
er
cont
rol
s
t
h
e m
o
t
i
on
of t
h
e
ro
bot
arm
by
M
e
ans o
f
d
r
i
v
e si
gnal
s
t
h
at
pa
ss
t
h
ro
u
gh t
h
e
dri
v
e Inte
rface
to the actuators on
the arm
.
Robots are often classified
un
de
r the thre
e
m
a
jor Categ
o
ries,
according to the type of control sy
ste
m
used, Non servo – ope
n loop s
y
ste
m
. Servo
– closed loop syste
m
,
Serv
o-con
t
ro
lled
– clo
s
ed
loo
p
system
s wi
th
con
tinu
o
u
s
l
y
Co
n
t
ro
lled
path
. Eith
er aux
iliary co
m
p
u
t
ers
or
e
m
bedde
d m
i
c
r
oprocess
o
rs are used
for prac
tically all cont
ro
l o
f
indu
str
i
al
r
o
b
o
t
s tod
a
y.
Th
ese perf
or
m
all o
f
the require
d
c
o
m
putational functions as
well as inte
rface with and control associate
d
sens
ors,
gri
ppe
rs,
t
ool
i
n
g, an
d
ot
her ass
o
ci
at
ed
peri
phe
ral
eq
ui
pm
ent
.
The control syste
m
p
e
rform
s
the necessary sequencing
and m
e
m
o
ry
funct
i
o
ns f
o
r o
n
-l
i
n
e se
nsi
n
g,
branc
h
i
n
g
,
an
d i
n
t
e
g
r
at
i
on
o
f
ot
he
r eq
ui
pm
ent
.
Pr
og
ram
m
i
ng o
f
th
e co
n
t
ro
llers can
b
e
don
e o
n
-lin
e or at re
m
o
te o
f
f-lin
e
cont
r
o
l
st
at
i
ons wi
t
h
el
ect
r
oni
c dat
a
t
r
a
n
sfer
o
f
p
r
og
ram
s
b
y
cassette, flopp
y d
i
sc,
or teleph
on
e m
o
d
e
m
.
Self-d
iagno
stic cap
ab
ility for tro
u
b
l
esh
o
o
tin
g and
main
ten
a
n
ce
greatly redu
ces ro
bo
t system do
wn
tim
e. So
m
e
ro
bo
t con
t
ro
llers
h
a
v
e
su
fficien
t
cap
a
city, in
term
s o
f
co
mp
u
t
ation
a
l ab
il
ity,
m
e
m
o
ry
cap
acity, and
in
pu
t-o
u
t
p
u
t
cap
a
b
ility to
serv
e also
as
syste
m
cont
rol
l
e
rs
an
d
ha
ndl
e
m
a
ny
ot
he
r m
achi
n
e
s
an
d
p
r
oces
se
s. P
r
o
g
r
am
m
i
ng
of
r
o
bot
c
o
n
t
rol
l
e
rs a
n
d sy
st
em
s
has
n
o
t
bee
n
s
t
anda
rdi
z
e
d
by
t
h
e
ro
b
o
t
i
c
s i
n
d
u
st
ry
;
t
h
e
r
e
f
ore
,
t
h
e
m
a
nuf
act
urer
s
use t
h
ei
r o
w
n
pr
op
ri
et
ary
pr
o
g
ram
m
i
ng
l
a
ng
ua
ges w
h
i
c
h req
u
i
r
e speci
al
t
r
ai
ni
n
g
of
p
e
rso
n
n
el
.
Co
n
t
ro
l Un
it:-
Th
e con
t
ro
ller is a co
mm
u
n
icatio
n
an
d
in
fo
rm
atio
n
p
r
o
cessing
Device th
at in
itia
tes, ter
m
in
ates an
d
coo
r
di
nat
e
s t
h
e
m
o
t
i
ons an
d s
e
que
nces
of a
ro
b
o
t
i
n
sh
ow
n
fi
gu
re.
4
. It
ac
cept
s
necessa
r
y
i
nput
s t
o
t
h
e
ro
bot
an
d
p
r
ov
id
es th
e Ou
tpu
t
d
r
i
v
e sign
als to
a co
n
t
ro
llin
g
m
o
to
r o
r
actuato
r to
Corresp
ond
with
th
e rob
o
t
m
o
v
e
m
e
n
t
s an
d
ou
tsid
e
wo
rl
d
.
As sh
own
i
n
b
e
ll
o
w
figu
re co
n
t
ro
lling
of th
e R
o
bo
tic Arm
will b
e
do
n
e
. It is
sho
w
n t
h
at
se
r
vo m
o
t
o
rs
are
cont
rol
l
e
d
wi
t
h
P
W
M
si
gnal
s
and t
h
e c
ont
rol
l
i
ng f
o
r
t
h
e
de
gree at
whi
c
h i
t
wi
l
l
rot
a
t
e
o
r
wo
rk
can b
e
pe
rf
orm
e
d. T
h
e
r
e are
v
a
ri
et
i
e
s of c
o
nt
rol
l
e
rs a
r
e a
v
ai
l
a
bl
e fo
r c
ont
r
o
l
l
i
ng
of
r
o
b
o
t
i
c
arm
and
Se
rv
o m
o
tors
. T
h
ere
are
two
m
a
jor ty
pe
s o
f
c
ont
rollers
[
2
]
,
[
8
]
,
[9]
:
Mechanical C
o
ntrollers
Electro
n
i
cs (Di
g
ital) Con
t
ro
llers
Fi
gu
re
4.
C
o
nt
r
o
l
l
i
ng B
l
ock
D
i
agram
Mechanic
a
l C
ontr
o
llers:-
M
echanical c
o
ntrollers
are
controlle
r that c
a
n
be treate
d
as a
n
alog
co
n
t
ro
llers. They work
o
n
mech
an
ism
s
an
d v
a
riou
s prin
cip
a
ls. Du
e t
o
th
is a
po
ssi
b
ility o
f
g
e
tting
h
i
gh
accuracy is
very low. He
nce, Mecha
n
ica
l
controlle
rs
are
not
prefe
r
red now a
da
y. As
com
p
ared t
o
Mech
an
ical con
t
ro
llers, Electron
ics (Dig
ital) con
t
ro
llers
are very fast and
due
t
o
digital
processi
ng of
inform
ation they are
widely
pre
f
erre
d.
Als
o
due
to slow proces
sing
s
p
eed
a
n
d data accuracy
Mec
h
anical
cont
rol
l
e
rs ar
e not
bei
ng
pre
f
e
rre
d an
d t
h
ey
are av
oi
de
d d
u
e
t
o
t
h
ese reaso
n
s. T
h
e ne
xt
i
s
El
ect
roni
cs
(Di
g
i
t
a
l
)
Co
n
t
ro
ller
s
.A
f
i
gu
r
e
I
llu
str
a
t
e
d
b
e
l
o
w show
s th
e
b
a
sic
desig
n
and
ar
chitectu
r
e of
proto
t
yp
e.
A
bov
e f
i
gu
r
e
sho
w
s t
h
e
bas
i
c pr
ot
ot
y
p
e
f
o
r
ro
b
o
t
arm
,
whi
c
h c
onsi
s
t
of
fi
ve
de
gree
s as nam
e
d A
1
,
A2
,
A3
, A
4
,
A
5
resp
ectiv
ely. Each
ax
is
h
a
s
sev
e
ral lim
its fo
r m
o
v
e
m
e
n
t
i
n
p
a
rt
i
c
ul
ar
d
i
rect
i
on l
i
ke
f
o
r
w
ar
d
an
d
re
verse
direction.
A ta
ble s
h
own as
below i
ndicates
the total
degre
e
of
free
dom
for each axis.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
089
-48
56
IJR
A
V
o
l
.
3, N
o
. 3,
Se
pt
em
ber 20
1
4
:
16
1 –
1
6
7
16
6
Axis
Degree*
A1
180
A2
120
A3
120
A4
90
A5
90
*i
n
d
i
cat
es t
o
t
a
l
de
gree
o
f
free
dom
.
B
y
expl
o
r
i
n
g o
u
r
desi
g
n
an
d
m
echani
cal
wo
rk
we ha
ve st
a
r
t
e
d w
o
rk
fo
r
devel
opm
ent
o
f
arm
body
.
At
fi
rst
we
p
u
r
c
hase
d a
pol
y
c
a
rb
o
n
at
e sheet
fr
om
l
o
cal
m
a
rket
.
As
per t
h
e
gui
dance
we
m
a
de
one
d
r
awi
n
g
on
pape
r o
f
di
f
f
er
ent
part
. B
e
f
o
r
e
pre
p
ari
n
g a h
a
rd
ware
desi
g
n
m
odel
for r
o
b
o
t
i
c
arm
we have pr
epa
r
ed ea
ch pa
rt
wi
t
h
har
d
pa
pe
r, whi
c
h
a
r
e of
t
h
e
sa
m
e
size that of
robotic a
r
m
’
s parts.
Electro
n
i
cs (
D
ig
ita
l)
Co
n
t
ro
llers:-
C
o
nt
r
o
l
l
e
rs a
r
e t
h
e
heart
o
f
a
n
y
o
f
t
h
e s
y
st
em
. C
ont
r
o
l
l
e
rs p
r
ovi
de
br
ai
n t
o
t
h
e sy
st
em
. A cont
rol
l
e
r
d
o
e
v
ery
t
a
s
k
a
nd
pr
ocesses
t
h
e i
n
f
o
rm
at
i
on’s i
n
si
de i
t
.
Acc
o
r
d
i
n
g t
o
t
h
e
obt
ai
ne
d
i
n
f
o
rm
at
i
on t
h
ey
pr
ovi
de c
o
nt
r
o
l
si
g
n
al
s t
o
t
h
e
sy
st
em
.
An
d acc
or
di
n
g
t
o
t
h
ese
si
g
n
a
l
s
sy
st
em
respo
n
d
s.
Hence
,
C
o
nt
ro
l
l
e
r i
n
any
of t
h
e sy
st
em
wor
k
s l
i
k
e a brai
n
do i
n
hum
an bo
dy
. It
co
nt
r
o
l
s
alm
o
st
every
t
hi
ng
.
Acco
r
d
i
n
g t
o
o
u
r
pr
o
j
ect
re
qu
i
r
em
ent
s
, we h
a
ve g
o
n
e t
h
ro
u
gh m
a
ny
co
nt
r
o
l
l
e
rs s
o
m
e
of whi
c
h are
di
sc
usse
d
bel
o
w:
8
051
Fam
ily
PIC fam
i
ly
AVR fam
i
ly
ARM base
d cortex-xx se
ries
All th
e ab
ov
e
men
tio
n
e
d
contro
llers are
from
m
a
rk
et
’s l
eadi
n
g m
a
nufac
t
u
rer
s
o
f
m
i
crochi
ps an
d
cont
rol
/
p
roc
e
ss
i
ng IC
s.
App
lica
tio
n:-
N
o
w
a
d
a
y, ev
er
y indu
str
y
w
a
n
t
s t
o
im
p
l
e
m
en
t tech
no
lo
g
i
cal so
lu
tion, bu
t du
e t
o
h
i
gh
er rates man
y
o
f
th
em
are no
t ab
le to p
u
rch
a
se it.
So
, th
is
p
r
od
uct is an
in
itiat
i
v
e
to
wards low co
st
so
lu
tion
s
i
n
au
to
m
a
t
i
o
n
tech
no
log
y
.
We
will try to
d
e
sign
i
n
du
strial
robo
t
su
ch
a
way th
at it h
a
s,
Hi
g
h
er
pr
od
uc
t
i
o
n
High
er Qu
ality.
No
La
bo
r
pr
o
b
l
em
Higher Speed
Man
u
facturing flex
i
b
ility
4.
CO
NCL
USI
O
N
A research
an
d d
e
v
e
l
o
p
m
en
t p
r
oj
ect in
robotic asse
m
b
ly a
u
to
m
a
tio
n
wh
i
c
h
m
e
rits sp
eci
al
m
e
n
tio
n
in
t
h
i
s
researc
h
p
a
per st
an
ds f
o
r
adapt
a
bl
e p
r
o
g
ram
m
abl
e
assem
b
l
y
sy
st
em
. t
h
e anal
y
s
i
s
m
e
t
hods i
n
cl
u
d
e t
h
e
pay
b
ac
k m
e
t
h
od
, ret
u
rn
on i
nve
st
m
e
nt
m
e
tho
d
s, a
n
d eq
ui
val
e
nt
u
n
i
f
orm
ann
u
al
cost
m
e
t
h
o
d
. R
o
b
o
t
and t
h
e
pr
o
g
ram
m
abl
e
aut
o
m
a
t
i
on pr
o
j
ect
s pre
s
ent
ce
rt
ai
n u
n
i
q
ue p
r
obl
em
i
n
t
h
e econ
om
i
c
s just
i
f
i
cat
i
on o
f
i
n
d
u
st
ri
al
ap
p
lication
s
REFERE
NC
ES
[1]
Mikell P. Groover,
Mitchel W
i
ss, “
Industrial Robotics
,
Techn
o
logy
, Programming, and Applications
”,
Ta
ta
Mcgraw Hill Ed
ucation Private
Lim
ited
,
Ind
i
a. I
S
BN(10) 1-25-900621-2.
[2]
S.
P.
Pa
rke
r
(e
d),
McGraw-Hill
E
n
cyc
lopedia
of
E
l
ec
tronics and
Computers
, McGraw-Hill, New Y
o
rk, 1984
.
[3]
R.R. Schr
eib
e
r,
“Robot Vision:
An
Ey
e to the
Future”,
Robotics Today
, June 183
.pp53-57
[4]
M.P Groover and E.W. Zimmers, Jr,
CAD/C
A
M: Computer-Aided Design a
nd manufacturing
, P
r
entic
e-ha
l
l
,
Englewoodcliffs
, NJ,1984, Chap
.10
[5]
Lee, J.K
., Stiehl, W.D., Tos
can
o
,
R.L.,
and Br
eazeal, C. 2009.
S
e
mi-Autonomous
Robot Ava
t
ar a
s
a Medium
for
Family Communication
and
Edu
c
ation
. Proceedings of Advanced
Robotics. 1925-
1949
[6]
Behavior
al Turing Test using Two-axis Actuato
r
s,
“Behavioral
Turing Test usin
g Two-axis Actuators”,
2012 IEEE
RO-MAN: The 2
1
st IEEE International S
y
mposiu
m on Robot and
Human Interactive Communica
tion
. September 9
-
13, 2012
. Paris
,
France.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
RA I
S
SN
:
208
9-4
8
5
6
The N
o
vel
of
Si
x axes
Ro
b
o
t
i
c
Arm f
o
r
In
d
u
st
ri
al
Ap
pl
i
c
at
i
o
ns (
R
aj
en
dra
A
par
n
a
t
h
i
)
16
7
[7]
Kozima, H., N
a
kagawa, C
.,
and
Yano, H., 2004
.
Can a robot empathize with p
e
o
p
le? Artificial Li
fe and Robo
tics,
Springer Japan
,
8(1), 83–88
.
[8]
Rajendr Ap
arnatthi, Ved
V
y
as Dwivedi,
“
E
le
ctr
i
cal Mach
ine and
Dr
ive (
I
ntr
oduce Advan
c
e
Contr
o
l)
”
Engineerin
g
acad
em
ic Publ
ic
ation-USA, ISB
N
:580-0-098-10107-9, ©
2013
BIOGRAP
HI
ES OF
AUTH
ORS
Assist. Prof. Rajendr
a Aparnathi. Receiv
ed
his B.E (Electrical Engin
eer
ing)
degree from
Bhavnager Univ
ersity
, qu
alif
ied
GATE-2009, an
d M.
E. (Industrial Electroni
cs) fr
om the Facu
lty
of Technolog
y
and Engineering
,
Maharaja Saya
jir
ao University
of Baroda. P.hd* (Pursuing)
CUS
h
ah Univers
i
t
y
, W
a
dhwanci
t
y
, S
u
rendran
ag
ar,
Gujar
a
t: IND
I
A The m
a
jor fi
elds of inter
e
st
are Industrial Automation and Power S
y
stems. He
joined C U Shah College of
Technolog
y
an
d
Engineering, C
U Shah University
, Gujar
a
t-Ind
i
a and Gujarat
Technolog
y
Univ
ersity
, Gujar
a
t,
India and as
an
As
s
i
s
t
ant P
r
ofess
o
r. Now he is
t
h
e tutor of grad
uate studen
t
s and Post graduate
s
t
udents
m
a
jorin
g
in P
o
wer E
l
ec
tronics
and
Driv
es
. In r
ecen
t
ye
a
r
s
his
res
ear
ch i
n
teres
t
s
fo
cus
on the fie
l
d of re
newable
energ
y
,
es
peci
all
y
on th
e invert
er t
echno
log
y
. He als
o
w
o
rked on R&D
projec
t with
com
p
anies
of
repu
te
in th
e fi
eld
of
el
ectr
i
ca
l
ele
c
tron
i
c
s
engin
eer
ing.
Dr.
Ved V
y
as Dwivedi,
Professor– Gujarat Tech
nological Univer
sity
, Ahmedabad
;
Pro-VC in C
U Shah University
,
wadhwancity,
Gujarat- INDIA;
is
a P
h
. D
.,
M
.
E.
, B.E
.
(a
ll E
.
C. Eng
i
ne
ering
)
has subm
itted his Post-doctorat
e
report; is a re
co
gnized Ph. D. g
u
ide for 06 cand
i
dat
e
s in R. K.
Univ. Rajko
t
(G
ujarat), Pacific
Univ. Udaipur
,
J.J.T. Univ. Jhunjhunu; 02 Ph. D. theses and 36
M Tech
dissert
at
ions subm
itted,
and no. of p
a
per
s
published ~
12
5, no. of
exper
t
t
a
lks deliver
ed
~ 47 in internation
a
l confer
en
ces, workshops,
STTP…; completed 03 research projects /
consultan
c
y
(
Govt. and nongovernment organizations
). His fields of research inter
e
st are
energ
y
-sensor-wireless-optical-radar-sate
llite-RF technolog
ies and
s
y
stems.
Evaluation Warning : The document was created with Spire.PDF for Python.