Vous êtes sur la page 1sur 26

# 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
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:

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
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

## 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

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?
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

1 0 0
Rot ( x, ) = 0 C S
0 S C
C 0 S
Rot ( y, ) = 0 1 0
S 0 C
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, )
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

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:
Translation of a along OX axis
Translation of d along 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

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}

## 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
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

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?

51

26