Kinematika Manipulator Robot

, ST MT.ac.id

Jurusan Teknik Elektro

lhokseumawe

Pendahuluan
Robot Manipulators
Konfigurasi Robot
S
Spesifikasi
ifik i R
Robot
b t
Jumlah Axis, DOF(Degree Of Freedom)
Precision, Repeatability

Kinematics
Preliminary
World frame, joint frame, end-effector frame
Rotation Matrix, composite rotation matrix
Homogeneous Matrix
Direct kinematics
Denavit-Hartenberg Representation
Examples
Inverse kinematics

, ST MT Sekolah 2

1
 Manipulators
Lengan Robot , Robot industri
Rigid bodies (links) connected
by joints
Joints: revolute or prismatic
Drive: electric or hydraulic
End-effector (tool) mounted on
a flange or plate secured to
the wrist jjoint of robot

, ST MT Sekolah Tinggi 3

Manipulators
R(revolute) or P( prismatic)
Robot Configuration:

Cartesian: PPP Cylindrical: RPP Spherical: RRP

Hand coordinate:

SCARA: RRP n: normal vector; s: sliding vector;

Articulated: RRR (Selective Compliance a: approach vector, normal to the
ST MT Assembly robotic
tool mounting plate

2
Robots Cartesian
Cartesian robot
Arm moves in 3 linear
axes. X,y,z axes

x
y

, ST MT Sekolah 5

ST MT Sekolah 6

3
Manipulators
Motion Control Methods
P i t tto point
Point i t control
t l
a sequence of discrete points
spot welding, pick-and-place, loading & unloading
Continuous path control
follow a prescribed path, controlled-path motion
Spray painting,
painting Arc welding,
welding Gluing

Tugino, Sekolah T 7

Manipulators
Robot Specifications
Number of Axes
Major axes, (1-3) => Position the

wrist
Minor axes, (4-6) => Orient the

tool
Redundant, (7-n) => reaching

around obstacles, avoiding

undesirable configuration
Degree of Freedom (DOF)
Workspace
Payload (load capacity) Which one is more important?

MTPrecision v.s. Repebility

, ST

4
Kinematik
Forward kinematics
y0
x0 = l cos y1
x1
y0 = l sin
l

Inverse kinematics
x0
= cos ( x0 / l )
1

, ST MT S 9

, ST MT
Sekolah 10

5
 Tugino, ST MT Sekolah Tinggi Teknologi Nasional Yogyakarta 11

A Simple Example

Revolute and
Finding :
Prismatic Joints
y
Combined = arctan( )
x
More Specifically:
(x , y)
y arctan2() specifies that its in the
= arctan 2( ) first quadrant
x
Y

S
1 Finding S:
X

S = (x 2 + y 2 )

ST MT Sekolah 12

6
 Inverse Kinematics of a Two Link Manipulator

(x , y) Given: l1, l2 , x , y

2 Find: 1, 2
l2 Redundancy:
A unique solution to this problem
does not exist. Notice, that using the
givens two solutions are possible.
1
l1
Sometimes no solution is possible.

(x , y)

, ST MT 13
Sekolah

Preliminary
Robot Reference Frames
World
W ld frame
f
Joint frame
Tool frame z
y
z x
y T P

W
x
R

ST MT Sekolah 14

7
 Preliminary
Coordinate Transformation
Reference coordinate frame
OXYZ z
Body-attached frame Ouvw P
Point represented in OXYZ:
y
Pxyz = [ p x , p y , p z ]T w
r v
Pxyz = p x i x + p y jy + p z k z
Point represented in Ouvw:
x
r O, O
u
Puvw = pu i u + pv jv + pwk w
Two frames coincide ==> pu = p x pv = p y pw = p z
, ST MT Sekolah 15

Preliminary
Properties: Dot Product
3
Let x and y be arbitrary vectors in R and be
the angle from x to y , then
x y = x y cos
Properties of orthonormal coordinate frame
Mutually perpendicular
Unit vectors
v v
i j =0 v
v v | i |= 1
i k = 0 v
v v | j |= 1
k j =0 v
| k |= 1
, ST MT Sekolah 16

8
 Preliminary
Coordinate Transformation
Rotation only z
P
r
Pxyz = p x i x + p y jy + p z k z
r y
Puvw = pu i u + pv jv + pwk w w v
u
Pxyz = RPuvw x

How to relate the coordinate in these two frames?

, ST MT Sekolah 17

Preliminary
Basic Rotation
p x , p y,
and p zrepresent the projections of P
onto OX, OY, OZ axes, respectively

Since
P = pu i u + pv jv + pw k w
p x = i x P = i x i u pu + i x jv pv + i x k w pw
p y = jy P = jy i u pu + jy jv pv + jy k w pw

p z = k z P = k z i u pu + k z jv pv + k z k w pw
ST MT Sekolah 18

9
 p x = i x P = i x i u pu + i x jv pv + i x k w pw
p y = jy P = jy i u pu + jy jv pv + jy k w pw

p z = k z P = k z i u pu + k z jv pv + k z k w pw

px i x i u i x jv i x k w pu
p = j i j y jv jy k w pv
y y u
p z k z i u k z jv k z k w pw
ST MT 19

Preliminary
Basic Rotation Matrix
p x i x i u i x jv i x k w pu
p = j i j j j k p
y y u y v y w v
p z k z i u k z jv k z k w pw
Rotation about x-axis with

z
w
1 0 0 P v
Rot ( x, ) = 0 C S

0 S C y
u
ST MT x
Sekolah 20

10
 Preliminary
Is it True?
Rotation about x axis with
p x 1 0 0 pu
p = 0 cos
y sin pv z
p z 0 sin cos pw w
P v
p x = pu

p y = pv cos pw sin
y
p z = pv sin + pw cos u
, ST MT Sekolah x 21

Basic Rotation Matrices

Rotation about x-axis with
1 0 0
Rot ( x, ) = 0 C S
0 S C
Rotation about y-axis with
C 0 S
Rot ( y, ) = 0 1 0
S 0 C
R t ti about
Rotation b t z-axis
i with
ith

C S 0
Pxyz = RPuvw Rot ( z , ) = S C 0
0 0 1
, ST MT Sekolah 22

11
 Preliminary
Basic Rotation Matrix
ix iu i x jv ix k w
R = jy i u j y jv jy k w Pxyz = RPuvw
k z i u k z jv k z k w
Obtain the coordinate of from the coordinate of
Puvw
Pxyz Dot products are commutative!

pu i u i x i u jy i u k z px
p = j i Puvw = QPxyz
v v x jv jy jv k z p y
pw k w i x k w jy k w k z pz
Q = R 1 = RT
QR = R T R = R 1 R = I 3 <== 3X3 identity matrix
Tugino Sekolah T 23

Example 2
A point auvw = (4,3,2) is attached to a rotating frame, the frame
rotates 60 degree about the OZ axis of the reference frame.
Find the coordinates of the point relative to the reference
frame after the rotation.

a xyz = Rot ( z ,60)auvw

0.5 0.866 0 4 0.598
= 0.866 0.5 0 3 = 4.964
0 0 1 2 2

ST MT Sekolah 24

12
 Example 3
A point a xyz = (4,3,2)is the coordinate w.r.t. the
reference coordinate system, find the corresponding
point auvw w.r.t. the rotated OU
OU-V-W
V W coordinate
system if it has been rotated 60 degree about OZ
axis.

auvw = Rot ( z ,60)T a xyz

0.5 0.866 0 4 4.598
= 0.866 0.5 0 3 = 1.964
0 0 1 2 2

, ST MT Sekolah 25

Composite Rotation Matrix

A sequence of finite rotations
matrix multiplications
p do not commute
rules:
if rotating coordinate O-U-V-W is rotating about
principal axis of OXYZ frame, then Pre-multiply the
previous (resultant) rotation matrix with an
appropriate basic rotation matrix
if rotating coordinate OUVW is rotating about its
own principal axes, then post-multiply the previous
(resultant) rotation matrix with an appropriate basic
rotation matrix

ST MT Sekolah 26

13
 Example 4
Find the rotation matrix for the following
operations:
R = Rot ( y, ) I 3 Rot ( w, ) Rot (u, )
Rotation about OY axis
C 0 S C S 0 1 0 0
Rotation about OW axis = 0 1 0 S C 0 0 C S

Rotation about OU axis - S 0 C 0 0 1 0 S C
CC SS CSC CSS + SC
= S CC CS

Answer... SC SSC + CS CC SSS

Pre-multiply if rotate about the OXYZ axes

Post-multiply if rotate about the OUVW axes
27
ST MT Sekolah

Example 5
Translation along Z-axis with h:
1 0 0 0 x 1 0 0 0 pu pu
0 1 0 0 y 0 1 0 0 pv pv
Trans ( z , h) = = =
0 0 1 h z 0 0 1 h pw pw + h

0 0 0 1 1 0 0 0 1 1 1
z z P
P
w y
y
w v
v
x u
h O, O x
O,0 , ST MT
u 28

14
 Example 6
Rotation about the X-axis by

1 0 0 0 x 1 0 0 0 pu
0 C 0 y 0 C S
S 0 pv
Rot ( x, ) = =
0 S C 0 z 0 S C 0 pw

0 0 0 1 1 0 0 0 1 1
z
w
P v

y
u
ST MT 29

Homogeneous Transformation
Composite Homogeneous Transformation
Matrix
Rules:
Transformation (rotation/translation) w.r.t
(X,Y,Z) (OLD FRAME), using pre-
multiplication
Transformation (rotation/translation) w
w.r.t
rt
(U,V,W) (NEW FRAME), using post-
multiplication

ST MT Sekolah 30

15
 Example 7
Find the homogeneous transformation matrix (T) for
the following operation:
Rotation about OX axis
Translation of a along OX axis
Translation of d along OZ axis
Rotation of about OZ axis
T = Tz , Tz ,d Tx,aTx, I 44
Answer : C S 0 0 1 0 0 0 1 0 0 a 1 0 0 0
S C 0 0 0 1 0 0 0 1 0 0 0 C S 0
=
0 0 1 0 0 0 1 d 0 0 1 0 0 S C 0

0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1
ST MT Sekolah a 31

Homogeneous Representation
A frame in space (Geometric
Interpretation) P( px , p y , p z )
z
R P31 a (z)
F = 33 s(y)
0 1 n (X)
y
nx sx ax px
n sy ay p y
F = y
nz sz az pz x

0 0 0 1
Principal axis n w.r.t. the reference coordinate system
, ST MT Sekolah 32

16
 Homogeneous Transformation
Translation a
s
1 0 0 d x nx s x ax px n
z
0 1 0 d y n y s y p y
ay
a
Fnew = s
0 0 1 d z nz s z az pz
n
0 0 0 1 0 0 0 1
nx sx ax px + d x y
n s y a y p y + d y
= y
nz sz az pz + d z

0 0 0 1
Fnew = Trans (d x , d y , d z ) Fold
33

Orientation Representation
Euler Angles Representation ( , , )
Many different types
Description of Euler angle representations

Euler Angle I Euler Angle II Roll-Pitch-Yaw

Sequence about OZ axis about OZ axis about OX axis
of about OU axis about OV axis about OY axis
Rotations about OW axis about OW axis about OZ axis

T S 34

17
 Orientation Representation
Euler Angle I

cos sin 0 1 0 0

Rz = sin cos 0 , Ru ' = 0 cos sin ,
0 0 1 0 sin cos

cos sin 0

Rw'' = sin cos 0
0 0 1

, ST MT 35

Euler Angle I
Resultant eulerian rotation matrix:

R = Rz Ru ' Rw''
cos cos cos sin
sin sin
sin sin cos sin cos cos


sin cos sin sin cos sin
+ cos sin cos + cos cos cos


sin sin cos sin cos

36

18
 Euler Angle II, Animated

w'= z

w"'= w"

v"'
v' =v"
y
u"'
"'
u"
Note the opposite u'
(clockwise) sense of the x
third rotation, .
37

Euler Angle I, Animated

w'= z

w'"= w"
v'"
v"

v'
y

u'"

u' =u"
x
38

19
 Orientation Representation
Matrix with Euler Angle II

i sin
sin i i cos
sin
cos sin
+ cos cos cos sin cos cos


cos sin cos cos sin sin
+ sin cos cos sin cos cos

cos
cos sin sin sin

Quiz: How to get this matrix ?
39

Orientation Representation
Description of Roll Pitch Yaw
Z

Y

X

Quiz: How to get rotation matrix ?

40

20
 Terimakasih

z
y
z
y x
z
z x y
y x

41

Coordinate Transformations
position vector of P
in {B} is transformed
to position vector of P
in {A}

description of {B} as
seen from an observer
in {A}

Rotation of {B} with respect to {A}

Translation of the origin of {B} with respect to origin of {A}

42

21
 Coordinate Transformations
Two Special Cases
r = ARB B r P + Ar o '
A P

1 Translation
1. T l ti only
l
Axes of {B} and {A} are
parallel
A
RB = 1

2. Rotation onlyy
Origins of {B} and {A}
are coincident

r =0
A o'
43

Homogeneous Representation
Coordinate transformation from {B} to {A}
r = RB B r P + Ar o '
A P A

A r P A RB r Br P
A o'

=
1 013 1 1
Homogeneous
g transformation matrix
A RB r R33
A o'
P31
Rotation

TB = =
A matrix

013 1 0 1 Position
vector
Scaliling 44

22
 Homogeneous Transformation
Special cases
1. Translation

I r
A o'
A
TB = 33
013 1
2. Rotation

A RB 031
A
TB =
013 1
45

The Geometric Solution

(x , y) Using the Law of Cosines:
l2
2
c 2 = a 2 + b 2 2ab cos C
2 2
( x 2 + y 2 ) = l1 + l2 2l1l2 cos(180 2 )
cos(180 2 ) = cos( 2 )
2 2
l1 x 2 + y 2 l1 l2
cos( 2 ) =
2l1l2
1
x 2 + y 2 l12 l2 2
2 = arccos

2 l l
1 2
Using the Law of Cosines:
Redundant since 2 could be in the
sin B sin C
= first or fourth quadrant.
b c
sin 1 sin(180 2 ) sin( 2 )
= = Redundancy caused since 2 has two possible
l2 x +y
2 2
x 2 + y2 values

1 = 1 + l sin( ) y
1 = arcsin 2 2
+ arctan 2
y x 2 + y2 x
= arctan
2
x

23
The Algebraic Solution
l2 (x , y)
2

c 1 = cos 1
c 1 + 2 = cos( 2 + 1 )
l1 (1) x = l1 c 1 + l 2 c 1 + 2
(2) y = l1 s 1 + l 2 sin 1+ 2
1
(3) = 1 + 2

(1) 2 + (2) 2 = x 2 + y 2 =
( 2 2 2
) ( 2 2 2
= l1 c1 + l2 ((c1+ 2 ) 2 + 2l1l2 c1 ((c1+ 2 ) + l1 s1 + l2 ((sin 1+ 2 ) 2 + 2l1l2 s1 ((sin 1+ 2 ) )
= l1 + l2 + 2l1l2 (c1 (c1+ 2 ) + s1 (sin 1+ 2 ) )
2 2

2 2
= l1 + l2 + 2l1l2 c 2 Only Unknown Note:
x 2 + y 2 l1 l2 2 2
cos(a+b) = (cosa)(cosb)+ (sina)(sinb)
2 = arccos
2l1l2 sin(a+b) = (cosa)(sinb)+ (cosb)(sin
47 a)

x = l1 c 1 + l 2 c 1 + 2 Note:
= l1 c 1 + l 2 c 1 c 2 l 2 s 1 s 2 cos(a+b) = (cosa)(cosb)+ (sina)(sinb)
= c 1 ( l1 + l 2 c 2 ) s 1 ( l 2 s 2 ) sin(a+b) = (cosa)(sinb)+ (cosb)(sina)

y = l1 s 1 + l 2 sin 1+ 2 We know what 2 is from the previous

= l1 s 1 + l 2 s 1 c 2 + l 2 s 2 c 1 slide. We need to solve for 1 . Now
= c 1 ( l 2 s 2 ) + s 1 ( l1 + l 2 c 2 ) we have two equations and two
unknowns (sin 1 and cos 1 )
x + s1 ( l 2 s 2 )
c1 =
( l1 + l 2 c 2 )
x + s1 ( l 2 s 2 ) Substituting for c1 and simplifying
y= ( l 2 s 2 ) + s 1 ( l1 + l 2 c 2 ) many times
( l1 + l 2 c 2 )

=
1
( l1 + l 2 c 2 )
(
x l 2 s 2 + s 1 ( l1 2 + l 2 2 + 2 l1 l 2 c 2 ) ) Notice this is the law of cosines
and can be replaced by x2+ y2

y (l1 + l 2 c 2 ) x l 2 s 2 y ( l1 + l 2 c 2 ) x l 2 s 2
s1 = 1 = arcsin

2
x + y 2
x2+ y2
48

24
 Homogeneous Transformation
Composite Homogeneous Transformation Matrix
z1 z2 y2

y1 x2
z0 0
A1 1
A2
y0
x1

x0
?
i 1 Transformation matrix for
Ai adjacent coordinate frames

0
A2 = 0A1 1 A2 Chain product of successive
coordinate transformation matrices
49

Example 8
For the figure shown below, find the 4x4 homogeneous transformation matrices
and fori 1i=1, 0 5
A 2, 3, 4,A
i i

nx sx ax px 1 0 0 0
c
n
sy ay p y 0 A = 0 0 1 e + c
z3 F = y 1
0 1 0 a d
b y3 x3 nz sz az pz
z5 0 0 0 1
0 0 0 1
d

x5 0 1 0 b
y5 z4 0 0 1 a d
A2 =
e 1

x4 y4 z2 1 0 0 0
a

x2 0 0 0 1
z1 x1 y2
z0 0 1 0 b
y1 1 0 0 e + c
x0 y0
Can you find the answer by observation
0
A =
2
0 0 1 0
based on the geometric interpretation of
0 0 0 1 50
homogeneous tranformations ?

25
Orientation Representation
R P31
F = 33
0 1
Rotation matrix representation needs 9 elements to
completely describe the orientation of a rotating rigid
body.
Any easy way?

Euler Angles Representation

51

26