Professional Documents
Culture Documents
Chapter 6 - 6 Robot Technology
Chapter 6 - 6 Robot Technology
DEPARTMENT
Chapter 6 :
ROBOT TECHNOLOGY
3.12 Robot Kinematics and Dynamics
3.13 Robot Arm Dynamics
2
3
4
1
Prismatic Joint
1 DOF (linear) (Variables - d)
Spherical Joint
3 DOF ( Variables - 1, 2, 3)
We are interested in two kinematics topics
Forward Kinematics (angles to position)
What you are given: The length of each link
The angle of each joint
A b x
A B A B cos θ b
θ y
B
Matrix Representation:
a x b x
A B a xb x a yb y
a y b y
Unit Vector
Vector in the direction of a chosen vector but whose
magnitude is 1.
B B
uB
B uB
Quick Matrix Review
Matrix Multiplication:
Matrix Addition:
a b e f a e b f
c d g
h c g d h
Basic Transformations
Moving Between Coordinate Frames
Translation Along the X-Axis
Y O
(VN,VO)
XY NO VO
V V
P
N
X V
N
Px
Y O
XY NO VO
V V
P
N
X VN
XY PX V N NO
V O P V
V
Translation along the X-Axis and Y-Axis O
NO VO
V
N
VXY VN
PXY
XY Px
X P
PY
PX V N
V XY P V NO O
P
Y V
Using Basis Vectors
Basis vectors are unit vectors that point along a coordinate axis
O
V N V cosθ V NO cosθ V NO n
NO
V NO O NO NO NO
V V sinθ V cos(90 θ) V o
Y
Rotation (around the Z-Axis)
Z X
Y
O
VO
V N
VY
VN
X
VX
V X V N
V XY Y V NO O
V V
x
Y Unit vector along X-Axis
O
V
Can be considered with respect to
the XY coordinates or NO coordinates
V VO
N V XY
V NO
VY
VN
X
VX
X XY NO NO
V V cos α V cos α V x
X N O (Substituting for VNO using the N and O
V (V n V o) x components of the vector)
V X V N (x n ) V O (x o )
V N (cos θ ) V O (cos( θ 90))
V N (cos θ ) V O (sin θ )
Similarly….
VY V NO
sin α V NO
cos(90 α) V NO
y
V Y (V N n V O o ) y
V Y V N (y n ) V O (y o )
V N (cos(90 θ)) V O (cos θ )
V N (sin θ ) V O (cos θ )
So….
X N
V V (cos θ ) V (sin θ )O V X
V XY Y
V Y V N (sin θ ) V O (cos θ ) V
Written in Matrix Form
VXY
X1
P
Translation along P followed by rotation by
X0
V X Px cos θ sin θ V N
V XY Y sin θ
V Py cos θ V O
(Note : Px, Py are relative to the original coordinate frame. Translation
followed by rotation is different than rotation followed by translation.)
In other words, knowing the coordinates of a point (VN,VO) in
some coordinate frame (NO) you can find the position of that
point relative to your original coordinate frame (X0Y0).
HOMOGENEOUS REPRESENTATION
Putting it all into a Matrix
V X cos θ sin θ Px V N
V Y sin θ cos θ Py V O Simplifying into a matrix form
1 0 0 1 1
cos θ sin θ Px
H sin θ cos θ Py Homogenous Matrix for a Translation in
XY plane, followed by a Rotation around
0 0 1 the z-axis
Rotation Matrices in 3D – OK, lets return from
homogenous representation
cos θ sin θ 0
Rotation around the
R z sin θ cos θ 0 Z-Axis
0 0 1
cos θ 0 sin θ
R y 0 1 0 Rotation around the
sin θ 0 cos θ Y-Axis
1 0 0
R z 0 cos θ sin θ Rotation around the
0 sin θ cos θ X-Axis
Homogeneous Matrices in 3D
H is a 4x4 matrix that can describe a translation, rotation, or both in one matrix
O
1 0 0 Px
Y
0 1 0 Py
P
N H
0 0 1 Pz
X A
Translation without 0 0 0 1
Z rotation
Y
n x ox ax 0
n oy ay 0
H y
O
N
n z oz az 0
X 0 0 0 1
Rotation part:
Z Rotation without Could be rotation around z-axis,
A translation x-axis, y-axis or a combination of the
three.
Homogeneous Continued….
V N
O
V The (n,o,a) position of a point relative
V XY H A
V to the current coordinate frame you
are in.
1
n x ox ax Px V N
n O
oy ay Py V V X n x V N o x V O a x V A Px
V XY y
n z oz az Pz V A
0 0 0 1 1
N W N
Y O
I W
W A
T
P
X
K A O
Z
W
X WI W N
Y Point relative to the J Point relative to the O Point relative to the
W X-Y-Z frame W I-J-K frame W N-O-A frame
WZ W K W A
W Pi n i
WI ni oi ai Pi W N
I
oi ai W N
J
J
a j W O
W nj oj aj Pj W O
W Pj n j oj
W K Pk n k W K n k ok ak Pk W A
ok a k W A
1 0 0 0 1 1
J
W N
Y N O
I W
W A
T
P
X
K A O
Z
W X i x jx kx Tx W I
W Tx i x
X
jx k x W I
Y
Y W i y jy ky Ty W J
W Ty i y jy k y W J
W Z i z jz kz Tz W K
W Z Tz i z jz k z W k
1 0 0 0 1 1
W X i x jx kx Tx n i oi ai Pi W N
W I
Y
J
W i y jy ky Ty n j oj aj Pj W O
Substituting W
for W K W Z i z jz kz Tz n k ok ak Pk W A
1 0 0 0 1 0 0 0 1 1
W X W N i x jx kx Tx n i oi ai Pi
Y O i
W H W jy ky Ty n j oj aj Pj
W Z W A H y
i z jz kz Tz n k ok ak Pk
1 1 0 0 0 1 0 0 0 1
Product of the two
Notice that H can also be written as: matrices
1 0 0 Tx i x jx kx 0 1 0 0 Pi n i oi ai 0
0 1 0 Ty i y jy ky 0 0 1 0 Pj n j oj aj 0
H
0 0 1 Tz i z jz kz 0 0 0 1 Pk n k ok ak 0
0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1
The Quest:
What is the position of the end
of the robotic arm?
Solution:
1. Geometric Approach
This might be the easiest solution for the simple situation. However,
notice that the angles are measured relative to the direction of the
previous link. (The first link is the exception. The angle is measured
relative to it’s initial position.) For robots with more links and whose arm
extends into 3 dimensions the geometry gets much more tedious.
2. Algebraic Approach
Involves coordinate transformations.
Example Problem:
You are have a three link arm that starts out aligned in the x-axis.
Each link has lengths l1, l2, l3, respectively. You tell the first one to move by
1 , and so on as the diagram suggests. Find the Homogeneous matrix to
get the position of the yellow dot in the X0Y0 frame.
Y3
3
Y2 2 X3 3
2 X2
Y0
1 i.e. Rotating by 1 will put you in the X1Y1 frame.
Translate in the along the X1 axis by l1.
X1
Y4
3
Y2 2 X3 3
2 X2
X4
1
The position of the yellow dot relative to the X4Y4
Y1
X0
frame is (0,0).
X 0
Y 0
H Notice that multiplying by the (0,0,0,1) vector
Z 0 will equal the last column of the H matrix.
1 1
More on Forward Kinematics…
Denavit - Hartenberg
Parameters
More on Forward Kinematics: Denavit - Hartenberg Parameters
Z(i - 1)
Y(i -1) Yi Zi
Xi ai
a(i - 1 ) di
X(i -1)
i
( i - 1)
THE PARAMETERS/VARIABLES: , a , d,
The Parameters
1) a(i-1)
Technical Definition: a(i-1) is the length of the perpendicular
between the joint axes. The joint axes is the axes around which
revolution takes place which are the Z(i-1) and Z(i) axes. These two
axes can be viewed as lines in space. The common perpendicular
is the shortest line between the two axis-lines and is
a(i-1) cont...
Visual Approach - “A way to visualize the link parameter a(i-1) is to
imagine an expanding cylinder whose axis is the Z(i-1) axis - when
the cylinder just touches the joint axis i the radius of the cylinder
is equal to a(i-1).” (Manipulator Kinematics)
It’s Usually on the Diagram Approach - If the diagram already
specifies the various coordinate frames, then the common
perpendicular is usually the X(i-1) axis. So a(i-1) is just the
displacement along the X(i-1) to move from the (i-1) frame to the
i frame.
Z(i - 1)
In other words,
displacement along the
Z(i - 1)
Zi to align the X(i-1) and Xi axes.
Y(i -1) Yi Zi
4) i :- Xi ai
a(i - 1 ) di
X(i -1)
Amount of rotation around
i
the Zi axis needed to align ( i - 1)
the X(i-1) axis with the Xi axis.
The Denavit-Hartenberg Matrix
cosθ i sinθ i 0 a(i 1)
sinθ cosα cosθ i cosα (i 1) sinα (i 1) sinα (i 1) d i
i (i 1)
sinθ i sinα (i 1) cosθ i sinα (i 1) cosα (i 1) cosα (i 1) d i
0 0 0 1
Just like the Homogeneous Matrix, the Denavit-Hartenberg
Matrix is a transformation matrix from one coordinate frame to
the next.
Using a series of D-H Matrix multiplications and the D-H
Parameter table, the final result is a transformation matrix from
some frame to your initial frame.
Z(i - 1)
Y(i -1) Y Zi
Put the transformation here i X a
a(i - d
X(i -1) i i
1) i
( i
i
- 1)
3 Revolute Joints Y2 Z2
Z0 Z1
X2
d2
X0 X1
Y0 Y1
Denavit-Hartenberg Link
a0 a1 Parameter Table
i (i-1 ) a (i-1 ) di i
Notice that the table has two
uses:
0 0 0 0 0
1) To describe the robot with its
variables and parameters.
1 0 a0 0 1
2) To describe some state of the
robot by having a numerical
2 -9 0 a1 d2 2
values for the variables.
Y2 Z2
i (i-1 ) a (i-1 ) di i
Z0 Z1
0 0 0 0 0
X2
d2
X0 X1 1 0 a0 0 1
Y0 Y1
2 -90 a1 d2 2
a0 a1
V X2
Y2
X 0 Y0 Z 0 V T ( 0T)( 01T)(12T)
V T
V Z2
1 Note: T is the D-H matrix with (i-1) = 0 and i = 1.
i (i-1) a (i-1) di i cosθ0 sinθ0 0 0
sinθ cosθ0 0 0
0 0 0 0 0 0T
0
0 0 1 0
1 0 a0 0 1 0 0 0 1
This is just a rotation around the Z0 axis
2 -90 a1 d2 2
S
1 Finding S:
X
S (x 2 y 2 )
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
l1
1 “givens” two solutions are possible.
Sometimes no solution is possible.
l2 (x , y)
l2
l1 l1
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 θ 2 )
cos(180 θ 2 ) cos(θ 2 )
2 2
l1 x 2 y 2 l1 l2
cos(θ 2 )
2l1l2
1
x 2 y 2 l12 l2 2
θ 2 arccos
2l1l2
Using the Law of Cosines:
Redundant since 2 could be in the
sin B sin C
first or fourth quadrant.
b c
sin θ1 sin(180 θ 2 ) sin(θ 2 )
Redundancy caused since 2 has two
l2 x 2 y2 x 2 y2 possible values
θ1 θ1 α l sin(θ ) y
θ1 arcsin 2 2
arctan 2
y x 2 y2 x
α arctan 2
x
The Algebraic Solution
l2 (x , y)
2
c 1 cos θ 1
c 1 2 cos( θ 2 θ 1 )
l1 (1) x l1 c 1 l 2 c 1 2
(2) y l1 s 1 l 2 sin 1 2
1
(3) θ θ 1 θ 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 l12 l2 2 cos(ab) (cosa)(cosb) (sina)(sinb)
θ 2 arccos
2 l l sin(ab) (cosa)(sinb) (cosb)(sina)
1 2
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(ab) (cosa)(cosb) (sina)(sinb)
c 1 ( l1 l 2 c 2 ) s 1 ( l 2 s 2 ) sin(ab) (cosa)(sinb) (cosb)(sina)
We know what 2 is from the
y l1 s 1 l 2 sin 1 2
previous slide. We need to solve for
l1 s 1 l 2 s 1 c 2 l 2 s 2 c 1 1 .
c 1 ( l 2 s 2 ) s 1 ( l1 l 2 c 2 )
Now we have two equations
x s1 ( l 2 s 2 )
and two unknowns (sin 1 and cos
c1 1 )
( l1 l 2 c 2 )
x s1 ( l 2 s 2 ) Substituting for c1 and simplifying many
y ( l 2 s 2 ) s 1 ( l1 l 2 c 2 ) 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 θ 1 arcsin
x2 y2 x2 y2
6.13 ROBOT ARM DYNAMICS
1. http://nptel.ac.in/courses/112103174/module7/lec5