Introduction to ROBOTICS

Kinematics of Robot Manipulator

Jizhong Xiao
Department of Electrical Engineering
City College of New York
jxiao@ccny.cuny.edu

The City College of New York 1

 Outline
• Review

• Robot Manipulators
– Robot Configuration
– Robot Specification
• Number of Axes, DOF
• 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

The City College of New York 2

 Review
• 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?
–Perform 4A tasks in 4D environments
4A: Automation, Augmentation, Assistance, Autonomous
4D: Dangerous, Dirty, Dull, Difficult

The City College of New York 3

 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

The City College of New York 4

 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
The City College of New York 5
 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

The City College of New York 6

 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)
– Precision v.s. Repeatability Which one is more important?

The City College of New York 7

 What is Kinematics
• Forward kinematics
z

Given joint variables

q = (q1 , q2 , q3 , q4 , q5 , q6 , L qn )
y

x
Y = ( x, y, z , O, A, T )

End-effector position and orientation, -Formula?

The City College of New York 8
 What is Kinematics
• Inverse kinematics
End effector position z
and orientation
( x, y, z , O, A, T )
y

q = (q1 , q2 , q3 , q4 , q5 , q6 , L qn ) x

Joint variables -Formula?

The City College of New York 9

 Example 1

Forward kinematics
y0
x0 = l cos θ y1
x1
y0 = l sin θ
l

Inverse kinematics θ
−1
x0
θ = cos ( x0 / l )

The City College of New York 10

 Preliminary
• Robot Reference Frames
– World frame
– Joint frame
z
– Tool frame y
z x
y T P

W
x
R

The City College of New York 11

 Preliminary
• Coordinate Transformation
– Reference coordinate frame
OXYZ z
– Body-attached frame O’uvw P
Point represented in OXYZ:
T y
Pxyz = [ p x , p y , p z ] w
r v
Pxyz = p x i x + p y jy + p z k z
Point represented in O’uvw:
x
r O, O’
u
Puvw = pu i u + pv jv + pw k w
Two frames coincide ==> pu = p x pv = p y pw = p z
The City College of New York 12
 Preliminary
Properties: Dot Product
3
Let x and y be arbitrary vectors in R and θ be
the angle from x to y , then
x ⋅ y = x y cos θ
Properties of orthonormal coordinate frame
• Mutually perpendicular • Unit vectors
v v v
i ⋅ j =0 | i |= 1
v v v
i ⋅k = 0 | j |= 1
v v v
k⋅ j =0 | k |= 1
The City College of New York 13
 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 + pw k w w v
u
Pxyz = RPuvw x

How to relate the coordinate in these two frames?

The City College of New York 14

 Preliminary
• Basic Rotation
– p x , p y, and p z represent 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

The City College of New York 15

 Preliminary
• Basic Rotation Matrix
 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 
– Rotation about x-axis with θ
z
w
1 0 0  P v
Rot ( x,θ ) = 0 Cθ − Sθ 
θ
0 Sθ Cθ  y
u
x
The City College of New York 16
 Preliminary
• Is it True?
– Rotation about x axis withθ
 p x  1 0 0   pu 
 p  = 0 cos θ  
− sin θ   pv  
 y  z
 p z  0 sin θ cos θ   pw  w
P v
p x = pu
θ
p y = pv cos θ − pw sin θ
y
p z = pv sin θ + pw cos θ u
x
The City College of New York 17
 Basic Rotation Matrices
– Rotation about x-axis with θ
1 0 0 
Rot ( x,θ ) = 0 Cθ − Sθ 
0 Sθ Cθ 
– Rotation about y-axis with θ
 Cθ 0 Sθ 
Rot ( y, θ ) =  0 1 0 
− Sθ 0 Cθ 

– Rotation about z-axis with θ

Cθ − Sθ 0
Pxyz = RPuvw Rot ( z ,θ ) =  Sθ Cθ 0
 0 0 1
The City College of New York 18
 Preliminary
• Basic Rotation Matrix

ix
iu
 ⋅ 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!
iu
ix

 pu   ⋅ i u ⋅ jy i u ⋅ k z   px 
 p  =  j ⋅i jv ⋅ jy

jv ⋅ k z   p y  Puvw = QPxyz
 v  v x  
 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

The City College of New York 19

 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 

The City College of New York 20

 Example 3
• A point axyz = (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 

The City College of New York 21

 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

The City College of New York 22

 Example 4
• Find the rotation matrix for the following
operations:
R = Rot ( y, φ ) I 3 Rot ( w, θ ) Rot (u, α )
Rotation φ about OY axis
 Cφ 0 Sφ  Cθ − Sθ 0 1 0 0 
Rotation θ about OW axis =  0 1 0   Sθ Cθ 0 0 Cα − Sα 
Rotation α about OU axis - Sφ 0 Cφ   0 0 1 0 Sα Cα 
 CφCθ SφSα − CφSθCα CφSθSα + SφCα 
=  Sθ CθCα − CθSα 

Answer ... − SφCθ SφSθCα + CφSα CφCα − SφSθSα 

Pre-multiply if rotate about the OXYZ axes

Post-multiply if rotate about the OUVW axes
The City College of New York 23
 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}

The City College of New York 24
 Coordinate Transformations
• Two Special Cases
A P A B P A o'
r = RB r + r
1. Translation only
– Axes of {B} and {A} are
parallel A
RB = 1

2. Rotation only
– Origins of {B} and {A}
are coincident
A o'
r =0
The City College of New York 25
Homogeneous Representation
• Coordinate transformation from {B} to {A}
r = RB B r P + Ar o '
A P A

 A r P   A RB A o'
r   r 
B P

 =  
 1   01×3 1  1 
• Homogeneous transformation matrix

A  RB
A
r   R3×3
A o'
P3×1 
Rotation
matrix
TB =  =  Position
 01×3 1   0 1  vector
Scaling
The City College of New York 26
Homogeneous Transformation
• Special cases
1. Translation

A  I 3×3 r 
A o'
TB =  
01×3 1 

2. Rotation

A  A
RB 03×1 
TB =  
 01×3 1 

The City College of New York 27

 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, O’
u
The City College of New York 28
 Example 6
• Rotation about the X-axis by
1 0 0 0  x  1 0 0 0  pu 
0 C θ − Sθ 0  y   0 Cθ − Sθ 0  pv 
Rot ( x,θ ) =   =
0 Sθ Cθ 0  z  0 Sθ Cθ 0  pw 
      
0 0 0 1  1  0 0 0 1  1 
z
w
P v

y
u
x
The City College of New York 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.r.t
(U,V,W) (NEW FRAME), using post-
multiplication

The City College of New York 30

 Example 7
• Find the homogeneous transformation matrix
(T) for the following operations:
Rotation α about OX axis T = Rx ,α I 4×4
Translation of a along OX axis
Translation of d along OZ axis
Rotation of θ about OZ axis
T = Tz ,θ Tz ,d Tx ,aTx ,α I 4×4
Answer : Cθ − Sθ 0 0  1 0 0 0  1 0 0 a  1 0 0 0
 Sθ Cθ 0 0 0 1 0 0  0 1 0 0  0 Cα − Sα 0
=
 0 0 1 0  0 0 1 d  0 0 1 0  0 Sα Cα 0
    
 0 0 0 1  0 0 0 1  0 0 0 1  0 0 0 1

The City College of New York 31

 Homogeneous Representation
• A frame in space (Geometric
Interpretation) P( p x , p y , p z )
z a (z’)
 R3×3 P3×1  s(y’)
F = 
 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

The City College of New York 32

 Homogeneous Transformation
• Translation a
s
1 0 0 d x   nx s x ax px 
z n
0 1 0 d y  n y s y ay p y 
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
The City College of New York 33
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 Chain product of successive

A2 = 0A1 1 A2 coordinate transformation matrices

The City College of New York 34

 Example 8
• For the figure shown below, find the 4x4 homogeneous transformation
matrices i −1 Ai and 0 Ai for i=1, 2, 3, 4, 5
 n s a p  − 1 0 0 0 
c x
n s a
x x x
  0 0 −1 e + c 
 y y y py  0 A =  
z3 F= 1
 0 −1 0 a − d 
b y3 x  nz s z a z p z   
z5
3
d    0 0 0 1 
0 0 0 1 
x5 0 − 1 0 b 
y5 z4 0 0 − 1 a − d 
A2 =  
e 1

a x4 y4 z2 1 0 0 0 
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
A2 = 
0 0 1 0 
based on the geometric interpretation of  
homogeneous transformation matrix? 0 0 0 1 
The City College of New York 35
 Orientation Representation
 R3×3 P3×1 
F = 
 0 1 
• Rotation matrix representation needs 9
elements to completely describe the
orientation of a rotating rigid body.
• Any easy way?

Euler Angles Representation

The City College of New York 36

 Orientation Representation
• Euler Angles Representation ( φ , θ, ψ )
– Many different types
– Description of Euler angle representations

Euler Angle I Euler Angle II Roll-Pitch-Yaw

Sequence φ about OZ axis φ about OZ axis ψ about OX axis
of θ about OU axis θ about OV axis θ about OY axis
Rotations ψ about OW axis ψ about OW axis φ about OZ axis

The City College of New York 37

 Euler Angle I, Animated

w'= z

w'"= w"
φ v'"
v"
ϕ
v'
y

u'"
θ
u' =u"
x

The City College of New York 38

 Orientation Representation
• Euler Angle I
 cos φ − sin φ 0 1 0 0 
   
Rzφ =  sin φ cos φ 0 , Ru 'θ =  0 cosθ − sin θ ,
 0 0 1   0 sin θ cos θ 
 
 cos ϕ − sin ϕ 0
 
Rw''ϕ =  sin ϕ cos ϕ 0
 0 0 1 


The City College of New York 39

 Euler Angle I
Resultant eulerian rotation matrix:

R = Rzφ Ru 'θ Rw''ϕ

 cos φ cos ϕ − cos φ sin ϕ 
 sin ϕ sin θ 
 − sin φ sin ϕ cosθ − sin φ cos ϕ cosθ 
 
 
 sin φ cos ϕ − sin φ sin ϕ − cos φ sin θ 
 + cos φ sin ϕ cosθ + cos φ cos ϕ cosθ 
 
 
 sin ϕ sin θ cos ϕ sin θ cosθ 
 
The City College of New York 40
 Euler Angle II, Animated

w'= z

w"'= w" φ

ϕ v"'
θ v' =v"
y
u"'
u"
Note the opposite u'
(clockwise) sense of the x
third rotation, φ.
The City College of New York 41
 Orientation Representation
• Matrix with Euler Angle II
 − sin φ sin ϕ − sin φ cos ϕ 
 cos φ sin θ 
 + cos φ cos ϕ cosθ − sin φ cos ϕ cosθ 
 
 
 cos φ sin ϕ cos φ cos ϕ sin ϕ sin θ 
 + sin φ cos ϕ cosθ − sin φ cos ϕ cosθ 
 
 cosθ 
 − cos ϕ sin θ sin ϕ sin θ
 
Quiz: How to get this matrix ?
The City College of New York 42
 Orientation Representation
• Description of Roll Pitch Yaw
Z

ϕ Y
θ
X

Quiz: How to get rotation matrix ?

The City College of New York 43

 Description of Roll Pitch Yaw
Z
X- Roll: sağa-sola hareket

Y- Pitch: ileri-geri hareket Y

Z- Yaw: bulunduğu konumda sağa-sola döndürme
X

Throttle: deviri arttırmaya ve azaltmaya yarar

The City College of New York 44