Robot Kinematics

1
 Review of Fundamentals
What is a robot?
By general agreement a robot is:
 A programmable machine that imitates the actions or appearance of
an intelligent creature–usually a human.
To qualify as a robot, a machine must be able to:
1) Sensing and perception: get information from its surroundings
2) Carry out different tasks: Locomotion or manipulation, do something
physical–such as move or manipulate objects
3) Re-programmable: can do different things
4) Function autonomously and/or interact with human beings
Why use robots?
–To Perform 4A tasks in 4D environments
4A: Automation, Augmentation, Assistance, Autonomous
4D: Dangerous, Dirty, Dull, Difficult

2
 Manipulators
Robot arms, industrial robot
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
joint of robot

3
 Manipulators
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
Assembly Robot Arm)
tool mounting plate
4
Application of Manipulators
Motion Control Methods
Point to point control
 a sequence of discrete points
 spot welding, pick-and-place, loading &
unloading
Continuous path control
 follow a prescribed path, controlled-path motion
 Spray painting, Arc welding, Gluing

5
 Specifications
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?
 Precision v.s. Repeatability
6
 Robot kinematics
Kinematics –study of motion with out regard
to the force generating it
In robots
Forward kinematics
 Deals with determination of position and orientation
of end effector as a function of joint positions
Inverse kinematics
 Deals with determination of joint positions for a
given end effector position and orientation

7
 Robot Kinematics
Forward Kinematics:
to determine where the robot’s hand is?
(If all joint variables are known)
Inverse Kinematics:
to calculate what each joint variable is?
(If we desire that the hand be
located at a particular point)

8
 Forward kinematics

Description of objects in space

Needs six parameters
 3 for rotation or orientation
 3 for position
For 3D space manipulation, 6DOF robot is required
Most industrial robots are 5 or 6DOF
To describe position and orientation of a link in
space, a coordinate frame is attached to the link

9
 Kinematics Model
Forward (direct) Kinematics
q  (q1 , q2 ,  qn ) z
Joint Position and Orientation
variables of the end-effector
Direct Kinematics

Inverse Kinematics x

Y  ( x, y, z,  , , )
Inverse Kinematics

10
A point P in space :
3 coordinates relative to a reference frame

^ ^ ^
P  a x i  by j  c z k

A Vector P in space :
3 coordinates of its tail and of its head
__ ^ ^ ^
P  a x i  by j  c z k
x
__  y
P 
z
 
 w 11
 Frame of Reference
Representation of a Frame at the
Origin of a Fixed-Reference Frame
 nx ox a x 
F  n y o y a y 
 nz oz a z 

Each Unit Vector is mutually perpendicular. :

normal, orientation, approach vector

 nx ox ax Px 
n oy ay Py 
F  y

 nz oz az Pz 
 
0 0 0 1

12
Representation of a Rigid Body

An object can be represented in space by attaching a

frame to it and representing the frame in space.

 nx ox ax Px 
n oy ay Py 
Fobject   y

 nz oz az Pz 
 
0 0 0 1

13
Homogeneous Transformation Matrices
A transformation matrices must be in square form for the
following reasons:

• It is much easier to calculate the inverse of square

matrices.
• To multiply two matrices, their dimensions must match.

 nx ox ax Px 
n oy ay Py 
F  y

 nz oz az Pz 
 
0 0 0 1

14
 Transformations
A transformation is defined as making a movement in space.
• A pure translation.
• A pure rotation about an axis.
• A combination of translation or rotations.

1 0 0 dx 
0 1 0 d y 
T 
0 0 1 dz 
 
0 0 0 1

15
 Pure Rotation about an Axis
Assumption : The frame is at the origin of the reference frame
and parallel to it.

16
 Inverse of Transformation Matrices
Inverse of a matrix calculation steps :
• Calculate the determinant of the matrix.
• Transpose the matrix.
• Replace each element of the transposed matrix by its own
minor(adjoint matrix).
• Divide the converted matrix by the determinant.

The Universe, robot, hand, part, and end effecter frames.

17
 Forward Kinematics
Forward Kinematics Analysis:
• Calculating the position and orientation of the hand of
the robot.
• If all robot joint variables are known, one can
calculate where the robot is at any instant.

1 0 0 Px 
0 1 0 Py 
R
TP  Tcart 
0 0 1 Pz 
 
0 0 0 1
18
 Cylindrical Coordinate
2 Linear translations and 1 rotation
• translation of r along the x-axis
• rotation of  about the z-axis
• translation of l along the z-axis

TP  Tcyl (r, , l )  Trans(0,0, l )Rot(z, )Trans(r,0,0)

R

C  S 0 rC 
 S C 0 rS 
R
TP  Tcyl   
 0 0 1 l 
 
 0 0 0 1 
Cylindrical Coordinates.

19
 Spherical Coordinates
2 Linear translations and 1 rotation
• translation of r along the z-axis
• rotation of  about the y-axis
• rotation of  along the z-axis

TP  Tsph (r ,  , l )  Rot(z,  )Rot( y,  )Trans(0,0,  )

R

C  C  S S  C rS  C 
 C  S C S  S rS  S 
TP  Tsph
R
 
  S 0 C rC 
 
 0 0 0 1 

Spherical Coordinates.

20
 Forward Kinematics
Forward kinematics
z

Given joint variables

q  (q1 , q2 , q3 , q4 , q5 , q6 ,  qn )
y

x
Y  ( x, y, z, O, A, T )
End-effector position and orientation, -Formula?

21
 Inverse Kinematics
Inverse kinematics
z
End effector position
and orientation
( x, y, z, O, A, T )
y

q  (q1 , q2 , q3 , q4 , q5 , q6 ,  qn ) x

Joint variables -Formula?

22
 Example
Forward kinematics y0
x0  l cos y1
x1
y0  l sin 
l

Inverse kinematics 
x0
  cos ( x0 / l )
1

23
24
 Preliminary

Robot Reference Frames

 World frame
 Joint frame z
y
 Tool frame
z x
y T P

W
x
R

25
 Continued…
Coordinate Transformation
Reference coordinate frame OXYZ
Body-attached frame O’uvw z

Point represented in OXYZ:

P

Pxyz  [ p x , p y , p z ] T
y
 w
Pxyz  p x i x  p y jy  p z k z v
Point represented in O’uvw: x
 O, O’
u
Puvw  pu i u  pv jv  pw k w
Two frames coincide ==> pu  p x pv  p y pw  p z
26
 Properties: Dot Product
Let x and y be arbitrary vectors in R3 and  be the
angle from x to y , then

x  y  x y cos 

Properties of orthonormal coordinate frame

Mutually perpendicular Unit vectors
  
i  j 0 | i | 1
  
i k  0 | j | 1
  
k j 0 | k | 1
27
 Coordinate Transformation
Rotation only
z
P

Pxyz  p x i x  p y jy  p z k z
 y
Puvw  pu i u  pv jv  pw k w w v

Pxyz  RPuvw u
x

How to relate the coordinate in these two frames?

28
 Basic Rotation
Basic Rotation
p x, p y, and pz represent the projections of P onto
OX, OY, OZ axes, respectively
Since
P  pu i u  pv jv  pw k w

px  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

pz  k z  P  k z  i u pu  k z  jv pv  k z  k w pw

29
 Rotation Matrix
 px   i x  i u i x  jv i x  k w   pu 
 p    j i jy  jv  
jy  k w   pv  
 y  y u
 p z  k z  i u k z  jv k z  k w   pw 

z
Rotation about x-axis with  w
P v
1 0 0 

Rot ( x, )  0 C 
 S 
 y
0 S C  u
x

30
 Rotation about x
Rotation about x axis with 
 p x  1 0 0   pu 
 p   0 cos   
 sin    pv 
 y  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
x

31
 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 

Rotation about z-axis with 

C  S 0
Pxyz  RPuvw Rot ( z, )   S C 0
 0 0 1
32
 Rotation Matrix
Basic Rotation Matrix
 i x  i u i x  jv i x  k w 
R   jy  i u jy  jv jy  k w  Pxyz  RPuvw
k z  i u k z  jv k z  k w 
Obtain the coordinate of Puvw from the coordinate of Pxyz
Dot products are commutative!

 pu   i u  i x i u  jy i u  k z   px  Puvw  QPxyz
 p    j i jv  j y

jv  k z   p y 
 v  v x
 pw  k w  i x k w  jy k w  k z   p z  1
QR R T

QR  R T R  R 1 R  I 3 <== 3X3 identity matrix

33
Denavit-Hartenberg
Representation

34
 Inverse Homogeneous Matrix?
The homogeneous transformation matrix is represented as:
 R33 P31  Position of the origin of OUVW
T   coordinate frame w.r.t. OXYZ frame

 0 1  Orientation of OUVW coordinate

frame w.r.t. OXYZ frame

The inverse of the transformation matrix is:

1  R T
 RT P  Position of the origin of OXYZ
T  
reference frame w.r.t. OUVW frame

0 1  Inverse of the rotation submatrix is

equivalent to its transpose

1  R T
 R T P   R P   R T R 0
T T       I 44
0 1   0 1   0 1
35
 Forward and Inverse Kinematics
Equations for Orientation
Once the position of the robot is determined, the next step will be
the frame appropriately in order to achieve a desired orientation
without changing its position.

The appropriate sequence of rotations depends on the design of

the wrist of the robot, and the way the joints are assembled
together.

The following are the common configurations:

Roll, Pitch, Yaw (RPY) angles
Euler angles
Articulated joints

36
 Roll, Pitch, Yaw(RPY) Angles
The movements relating to RPY rotations are relative to the
current moving axes.
Roll: Rotation of  a about a -axis (z-axis of the moving frame)
Pitch: Rotation of  o about o -axis (y-axis of the moving frame)
Yaw: Rotation of  n about n -axis (x-axis of the moving frame)

37
 Orientation Representation
Description of Roll Pitch Yaw
Z

 Y

X

How to get rotation matrix ?

38
 Rotation Matrix of YPR
How to get the resultant rotation matrix for YPR?
T  Rz , R y , Rx ,
C  S 0 0   C 0 S 0 1 0 0 0
 S C 0 0  0 1 0 0 0 C  S 0
 
0 0 1 0    S 0 C 0 0 S C 0
    
0 0 0 1  0 0 0 1 0 0 0 1
Z


 Y

X 39
 Euler Angles
• Rotation of  abouta -axis (z-axis of the moving frame)
followed by
• Rotation of  about o -axis (y-axis of the moving frame)
followed by
• Rotation of  about a -axis (z-axis of the moving frame).

40
 Euler Angle
w'= z

w'"= w"
 v'"
v"

v'
y

u'"

u' =u"
x

41
 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 

Rotation about z-axis with 

C  S 0
Pxyz  RPuvw Rot ( z, )   S C 0
 0 0 1
42
 Orientation Representation
Euler Angle
 cos  sin  0 1 0 0 
   
Rz   sin  cos 0 , Ru '   0 cos  sin  ,
 1   0 sin  cos 
 0 0 
 cos  sin  0
 
Rw''   sin  cos 0
 0 1 
 0

43
 Euler Angle
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 
 
44
 Forward And Inverse Kinematics
Equations For Orientation
The matrix representing the final location and orientation of the
robot is a combination of the preceding equations, depending on
which coordinates are used.
 Assumption : Robot is made of a Cartesian and an RPY set of
joints. R
TH  Tcart ( Px , Py , Pz )  RPY (a , o , n )

 Assumption : Robot is made of a Spherical Coordinate and an

Euler angle. R
TH  Tsph (r ,  ,  )  Euler( ,  , )

Another Combination can be possible…

Denavit-Hartenberg Representation

45
 Denavit-hartenberg Representation of
Forward Kinematic Equations of
Robot
Denavit-Hartenberg Representation :
@ Simple way of modeling robot links and
joints for any robot configuration,
regardless of its sequence or complexity.
@ Transformations in any coordinates
is possible.
@ Any possible combinations of joints
and links and all-revolute articulated
robots can be represented.

46
 Denavit-hartenberg Representation
Procedures
Start point:
Assign joint number n to the first
z
shown joint. y
z
Assign a local reference frame y x
z z
for each and every joint before or x y
after these joints. y x
Y-axis does not used in D-H
representation. x

47
Denavit Hartenberg Representation

48
 Procedures For Assigning A Local
Reference Frame To Each Joint
‫ ٭‬All joints are represented by a z-axis. (right-hand rule for
rotational joint, linear movement for prismatic joint)
‫ ٭‬The common normal is one line mutually perpendicular to any
two skew lines.
‫ ٭‬Parallel z-axes joints make a infinite number of common
normal.
‫ ٭‬Intersecting z-axes of two successive joints make no common
normal between them(Length is 0.).

49
 Symbol Terminologies

 : A rotation about the z-axis.

d : The distance on the z-axis.
a : The length of each common normal (Joint offset).
 : The angle between two successive z-axes (Joint twist)

 Only  and d are joint variables.

50
 Link and Joint Parameters
Joint angle  i : the angle of rotation from the Xi-1 axis to the Xi
axis about the Zi-1 axis. It is the joint variable if joint i is
rotary.

Joint distance d i : the distance from the origin of the (i-1)

coordinate system to the intersection of the Zi-1 axis and the
Xi axis along the Zi-1 axis. It is the joint variable if joint i is
prismatic.

Link length a i : the distance from the intersection of the Zi-1 axis
and the Xi axis to the origin of the ith coordinate system along
the Xi axis.

Link twist angle  i: the angle of rotation from the Zi-1 axis to the
Zi axis about the Xi axis.

51
52
 Transformation between i-1 and I
Procedure
Four successive elementary transformations are required to relate
the i-th coordinate frame to the (i-1)-th coordinate frame:
• Rotate about the Z i-1 axis an angle of i to align the X i-1
axis with the X i axis.
• Translate along the Z i-1 axis a distance of di, to bring Xi-1
and Xi axes into coincidence.
• Translate along the Xi axis a distance of ai to bring the two

origins Oi-1 and Oi as well as the X axis into coincidence.

• Rotate about the Xi axis an angle of αi ( in the right-handed

sense), to bring the two coordinates into coincidence.

53
 Transformation between i-1 and i
D-H transformation matrix for adjacent coordinate
frames, i and i-1.
The position and orientation of the i-th frame coordinate can be
expressed in the (i-1)th frame by the following homogeneous
transformation matrix:
Source coordinate

Ti i 1  T ( zi 1 , d i ) R( zi 1 , i )T ( xi , ai ) R( xi , i )
C i  C i S i S i S i ai C i 
Reference  S C  i C i  S i C i ai S i 
Coordinate  i
 0 S i C i di 
 
 0 0 0 1 
54
 The Fundamental Problem With
D-h Representation
Defect of D-H presentation : D-H cannot represent any
motion about the y-axis, because all motions are about the x-
and z-axis. The Parameters Table For The
Stanford Arm

#  d a 
1 1 0 0 -90

2 2 d1 0 90

3 0 d1 0 0

4 4 0 0 -90

5 5 0 0 90

6 6 0 0 0
55
Link Frame Assignments

56
 Solution
 c1  s1 0 a0  c1  s1 0 0
 s c c1c 0  s 0  s 0 d1   s1 c1 0 0
0
T   1 0

1
 s1s 0 c1s 0 c 0 c 0 d1   0 0 1 0
   
 0 0 0 1  0 0 0 1
c 2  s 2 0 L1  c3  s3 0 L2 
 s c 2 0 0   s c3 0 0 
1
T   2
3T 
2  3
2
 0 0 1 0  0 0 1 0
   
 0 0 0 1  0 0 0 1
c123  s123 0.0 l1c1  l2 c12 
s c 0.0 l s  l s 
B
T  0
T   123 123 1 1 2 12 
.
W 3
 0.0 0.0 1.0 0.0 
 
 0 0 0 1  57
 The Inverse Kinematic Solution of
Robot
Determine the value of each joint to place the arm at a
desired position and orientation.
R
TH  A1 A2 A3 A4 A5 A6

C1 (C234C5C6  S 234 S 6 ) C1 (C234C5C6  S 234C6 ) 

 S S C C1 (C234 S5 )  S1C5 C1 (C234 a4  C23a3  C2 a2 )
 1 5 6  S1S5C6 
S1 (C234C5C6  S 234 S 6 ) S1 (C234C5C6  S 234C6 )
 S1 (C234 S5 )  C1C5 S1 (C234 a4  C23a3  C2 a2 ) 
  C1S5C6  C1S5C6 
 S 234C5C6  C234 S 6  S 234C5C6  C234C6 S 234 S5 S 234 a4  S 23a3  S 2 a2 
 
 0 0 0 1 
 nx ox ax px 
n oy ay p y 

y

 nz oz az pz 
 
0 0 0 1 58
 Continued…
 nx ox axpx 
 p y 
1  n y oy ay
A1   A11[ RHS ]  A2 A3 A4 A5 A6
 nz oz azpz 
 
0 0 0 1
C1 S1 0 0  nx ox a x p x 
0 0 1 0 n y o y a y p y 
   A2 A3 A4 A5 A6
 S1  C1 0 0  nz oz a z p z 
   
0 0 0 1  0 0 0 1 

 py  (C3a3  a2 )( p z  S 234 a4 )  S3a3 ( p xC1  p y S1  C234 a4 )

1  tan 1    2  tan 1
 x
p (C3a3  a2 )( p xC1  p y S1  C234 a4 )  S3a3 ( Pz  S 234 a4 )

 S3   4   234  2 3
 3  tan 1  
 C3 

C234 (C1a x  S1a y )  S 234 a z

 5  tan 1
S1a x  C1a y

59
 Kinematic Equations
Forward Kinematics q  (q1 , q2 ,  qn )
 Given joint variables
 End-effector position & orientation Y  ( x, y, z,  , , )
Homogeneous matrix T0n
 specifies the location of the ith coordinate frame w.r.t. the
base coordinate system
 chain product of successive coordinate transformation
matrices of Ti i 1
Position
T  T T T
0
n 1 2
0 1
n
n 1 vector

Orientation  R0n P0n  n s a P0n 

matrix   
0 1  0 0 0 1 
60
 Kinematics Equations
Other representations
reference from, tool frame

Treftool  Bref
0
T0n H ntool
Yaw-Pitch-Roll representation for orientation

T  Rz , R y , Rx ,
C  S 0 0   C 0 S 0 1 0 0 0
 S C 0 0  0 1 0 0 0 C  S 0
 
0 0 1 0    S 0 C 0 0 S C 0
    
0 0 0 1  0 0 0 1 0 0 0 1

61
 Representing Forward Kinematics
Yaw-Pitch-Roll representation for orientation

CC CSS  SC CSC  SS px 

 SC SSS  CC SSC  CS p y 
T0n  
  S CS CC pz 
 
 0 0 0 1
 nx sx ax px    sin 1 ( nz )
n p y  az
T0n   y
sy ay
  cos (
1
)
 nz sz az pz  cos
  nx
0 0 0 1   cos (
1
)
cos
Solution is inconsistent and ill-conditioned!!
62
 Yaw-Pitch-Roll Representation
T  Rz , R y , Rx ,

C  S 0 0  C 0 S 0 1 0 0 0
 S C 0 0  0 1 0 0 0 C  S 0
  
0 0 1 0   S 0 C 0 0 S C 0
     
0 0 0 1  0 0 0 1 0 0 0 1

 nx sx ax 0
n sy ay 0
 y
 nz sz az 0
 
0 0 0 1

63
 Yaw-Pitch-Roll Representation
1
Rz , T  R y , Rx ,
 C S 0 0  nx sx ax 0
 S 0 n
 C 0
 y sy ay 0
 0 0 1 0  nz sz az 0
   
 0 0 0 1 0 0 0 1
(Equation A)

 C 0 S 0 1 0 0 0
 0 0 C 0
1 0 0   S

 S 0 C 0 0 S C 0
   
 0 0 0 1 0 0 0 1
64
 Yaw-Pitch-Roll Representation
Compare LHS and RHS of Equation A, we have:

 sin   nx  cos   n y  0   a tan 2(n y , nx )

cos  nx  sin   n y  cos


 nz   sin    a tan 2( nz , cos   nz  sin   n y )

  sin   s x  cos  s y  cos


 sin   a x  cos  a y   sin 

  a tan 2(sin   a x  cos   a y , sin   s x  cos   s y )

65
 Kinematic Model

Steps to derive kinematics model:

 Assign D-H coordinates frames
 Find link parameters

 Transformation matrices of adjacent joints

 Calculate Kinematics Matrix

66
Assignment 1

67
Assignment 2

68
 Composite transformation
The total transformation will be, product of the
transformation matrices
i 1
Ti  Tz i Tz di Tx a i Tz i 

C  SC SS aC

 S CC  CS aS 
i 1
Ti   
0 S C d 
 
 0 0 0 1 

69
 Complete transformation matrix of a
Robot
The complete transformation matrix of a robot
gives, the location and orientation of the end
effector with respect to the base.
It is found by multiplication of transformation
matrix of adjacent frames.
0
Tn  0 T1 1 T2 2 T3 ...n 1 Tn

Above equation is known as Kinematic model of

robot.
70
 Example – 2DOF planar robot
For the following two DOF robot, solve the
forward kinematics

x  L1 cos(1 )  L 2 cos(1   2 )
y  L1 sin( 1 )  L 2 sin( 1   2 )

71
 Example
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 

72
 Example
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-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 

73
 Composite Rotation Matrix
A sequence of finite rotations
matrix multiplications 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
74
 Example
Find the rotation matrix for the following
operations: Rotation  about OY axis
Rotation  about OW axis
Rotation  about OU axis
Answer ...
R  Rot ( y,  ) I 3 Rot ( w, ) Rot (u,  )
 C 0 S  C  S 0 1 0 0 
  0 1 0   S C 0 0 C  S 
- S 0 C   0 0 1 0 S C 
 CC SS  CSC CSS  SC 
  S CC  CS 

 SC SSC  CS CC  SSS 

75
 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}

76
 Coordinate Transformations
Two Special Cases
r  RB B r P  Ar o'
A P A

1. Translation only
Axes of {B} and {A} are
parallel A
RB  1
2. Rotation only
Origins of {B} and {A}
are coincident
r 0
A o'

77
 Homogeneous Representation
• Coordinate transformation from {B} to {A}
r  ARB B r P  Ar o'
A P

 A r P   A RB 
A o'
r  r 
B P

   
 1   013 1  1 
• Homogeneous transformation matrix

 RB
A
r   R33
A o'
P31  Rotation
matrix
TB  
A
  Position
 013 1   0 1  vector
Scaling
78
 Homogeneous Transformation
Special cases
1. Translation
 I 33 r 
A o'
A
TB   
013 1 

2. Rotation
 A
RB 031 
A
TB   
 013 1 

79
 Example
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 O, O’
u x
h
O, O’
u
80
 Example
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  p w 
      
0 0 0 1  1  0 0 0 1  1 
z
w
P v

y
u
x
81
 Transformation Rule
Composite Homogeneous Transformation
Matrix Rules:
Transformation (rotation/translation) w.r.t
(X,Y,Z) (OLD FRAME), using pre-
multiplication
Transformation (rotation/translation) w.r.t
(U,V,W) (NEW FRAME), using post-
multiplication

82
 Example
Find the homogeneous transformation matrix (T) for the
following operations:
Rotation  about OX axis
Translation of a along OX axis
T  Rx , I 44
Translation of d along OZ axis
Rotation of  about OZ axis
T  Tz , Tz ,d Tx ,aTx , I 44
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

83
 Homogeneous Representation
A frame in space (Geometric
Interpretation) P( px , p y , pz )
z
 R33 P31  a (z’)
F  
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

84
 Homogeneous Transformation
Translation
a
1 0 0 d x   nx s x ax px  s
0 1 0 d y  n y s y ay p y 
Fnew   z a n
0 0 1 d z   nz s z az pz  s
   
0 0 0 1 0 0 0 1
sx ax px  d x 
n
 nx
n s y a y p y  d y 
y
 y
 nz sz az pz  d z 
 
0 0 0 1 

Fnew  Trans (d x , d y , d z )  Fold

85
 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 0A11 A2 Chain product of successive coordinate
transformation matrices

86
Forward kinematics: Find coordinate of yellow dot with
respect to base frame
Y3

Y4
3

Y2 2 X3 3
2 X2
X4

H = Rz(1 ) * Tx1(l1) * Rz(2 ) * Tx2(l2) * Rz(3 ) * Tx3(l3)

Y0
1 This takes you from the X0Y0 frame to the X4Y4 frame.

1 The position of the yellow dot relative to the X4Y4 frame

X0
is (0,0).

Inverse kinematics is given (x4, y4), find the angles y3,y2 and y1
87
 Articulated Robot
Using DH algorithm
Link a  d  q C S
1 0 90 0 1 1 0 1
2 L2 0 0 2 2 1 0
3 L3 0 0 3 3 1 0

C1 0 S1 0 C1
C 2  S2 0 L 2C 2 
S 0  C1 0 S C2 0 L 2S2 
0
T1   1 1
T2   2
0 1 0 0 0 0 1 0 
   
0 0 0 1 0 0 0 1 

C 3  S3 0 L 3C 3  C1C 23  C1S23 S1 C1 L 3C 23  L 2 C 2 
S 0 L 3S3  0 S C
C3  S1S23  C1 S1 L 3C 23  L 2 C 2 
2
T3   3 T3   1 23
0 0 1 0   S23 C 23 0 L3S23  L 2S2 
   
0 0 0 1   0 0 0 1 
88
 Articulated Manipulator
Let the end effector position and orientation be
 0.35 0.866 0.35 0.11 
 0.6 0.5  0.6  0.18
T 
 0.7 0 0.7 0.21 
 
 0 0 0 1 

Find the three joint angles

Solutions
Step 1) apply guide line a, b, c and d
C1C 23  C1S23 S1 C1 L3C 23  L 2 C 2   r11 r12 r13 r14 
S C  S S  C S L C  L C   r r22 r23 r24 
0
T3  1 23 1 23 1 1 3 23 2 2   21

 S23 C 23 0 L3S23  L 2S2  r31 r32 r33 r34 
   
 0 0 0 1  0 0 0 1
89
 Articulated joint Cont’d
S1 r24  0.18
 
C1 r14 0.11

Which gives 1=-58.57 degree

For the other two angles, use guideline e
0
T1 T2 T3  T
1 2

T1 T2  T T3
0 1
 
2 1

Use elements, (1,4), (2,4) and (3,4) which are functions

of 1 and 2 only L C C  L r  r
2 2 2 3 11 14

L 2S1C 2  L3r21  r24

L 2S2  L3r31  r34

90
 Articulated joint Cont’d
Squaring the first two equations and adding them and
using the trigonometric identity

L 2C2    L3r11  r14 2   L3r21  r24 2

Dividing the third equation by this one

S2  L3r31  r34

C2   L3r11  r14 2   L3r21  r24 2

91
 Articulated joint Cont’d

The third angle is obtained by dividing (3,1) by (3,2)

and equating to right side

C1C 23  C1S23 S1 C1 L3C 23  L 2 C 2   r11 r12 r13 r14 

S C  S1S23  C1 S1 L 3C 23  L 2 C 2   r21 r22 r23 r24 
0
T3   1 23 
 S23 C 23 0 L3S23  L 2S2  r31 r32 r33 r34 
   
 0 0 0 1  0 0 0 1

S23 r31
This gives  and hence
C 23 r32

2+3= atan(r31,r32)
92