Beruflich Dokumente
Kultur Dokumente
Robot Kinematics
Renata Melamud
4
1
Prismatic Joint
1 DOF (linear) (Variables - d)
Spherical Joint
3 DOF ( Variables - 1, 2, 3)
ax
a
y
Geometric Representation:
A B A B cos
Matrix Representation:
bx
b
y
ax bx
AB
a xb x a y b y
ay by
Unit Vector
Vector in the direction of a chosen vector but whose magnitude is 1.
B
uB
B
B
uB
a b e
c d g
f ae bg
h ce dg
af bh
cf dh
Matrix Addition:
a b e
c d g
f a e
h c g
b f
d h
Basic Transformations
Moving Between Coordinate Frames
Translation Along the X-Axis
Y
O
(VN,VO)
NO
XY
P
X
Px
Px = distance between the XY and NO coordinate planes
Notation:
XY
VX
Y
NO
VN
O
Px
P
VO
Writing V XY in terms of V NO
Y
NO
VX
P
X
V XY
VN
PX V N
NO
O
V
VO
N
Y
NO
V
Y
VX
PX
X
V XY P V NO
PX V N
O
P
V
Y
XY
Px
PY
VO
n
o
V NO
NO
V NO
VO
VN
NO
V NO cos V NO n
V N V cos
O
NO
NO
NO
V V sin V cos(90 ) V o
VY
N
VX
V XY
VX
Y
V NO
VN
O
V XY
V NO
VY
VX
V X V XY cos V NO cos V NO x
(V n V o ) x
N
V X V N (x n ) V O (x o)
V N (cos) V O (cos( 90))
V N (cos) V O (sin)
Similarly.
V Y V NO sin V NO cos(90 ) 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 )) V O (cos)
V N (sin) V O (cos)
So.
V V (cos) V (sin)
V Y V N (sin) V O (cos)
X
VX
V XY
XY
V X cos sin
Y
sin
cos
V
VN
O
V
Y1
(VN,VO)
Y0
VNO
VXY
X1
X0
XY
Px
cos
Y
P
sin
V
y
VX
sin V N
cos V O
(Note : Px, Py are relative to the original coordinate frame. Translation followed by
rotation is different than rotation followed by translation.)
HOMOGENEOUS REPRESENTATION
Putting it all into a Matrix
XY
Px
cos
Y
P
sin
V
y
sin V N
cos V O
VX
VX
V
1
VX
VY
1
cos
H sin
0
sin
Px
cos
Py sin
1
0
cos
sin
0
sin
cos
0
cos
0
sin
cos
0
0 V N
0 V O
1 1
Px V N
Py V O
1 1
Px
Py
1
cos
R z sin
0
sin
cos
cos
R y
0
sin
1
R z 0
0
0
cos
sin
0
0
1
0
0
0
sin
0
cos
0
sin
cos
Homogeneous Matrices in 3D
H is a 4x4 matrix that can describe a translation, rotation, or both in one matrix
O
Y
N
P
X
1
0
H
0
0
1
0
0
0 Px
0 Py
1 Pz
0 1
Z
Y
O
N
X
Z
A
nx
n
y
nz
ox
oy
ax
ay
oz
0
az
0
0
0
Rotation part:
Could be rotation around z-axis,
x-axis, y-axis or a combination of
the three.
Homogeneous Continued.
V XY
XY
VN
O
V
H A
V
nx
n
y
nz
ox
oy
oz
0
ax
ay
az
0
Px
Py
Pz
VN
O
V
VA
V X n x V N o x V O a x V A Px
The rotation and translation part can be combined into a single homogeneous
matrix IF and ONLY IF both are relative to the same coordinate frame.
WN
O
W
WA
T
P
X
Z
WX
Y
W
WZ
WI
W
WK
Pi n i
Pj n j
Pk n k
oi
oj
ok
J
W
K
W
WN
ai W
a j W O
a k W A
N
WI
O
W
A
W
ni
n
j
J
W
W K nk
1 0
oi
ai
oj
aj
ok
ak
Pi
Pj
Pk
1
WN
WO
WA
1
WN
O
W
WA
T
X
P
A
W
X
Y
W
Z
W
Tx i x
Ty i y
Tz i z
kx W
k y W J
k z W k
jx
jy
jz
WX
Substituting for W
WK
WX
ix
i
y
Y
W
WZ iz
1 0
ix
i
y
Y
W
Z
W iz
jx
kx
jy
ky
jz
kz
Tx
Ty
Tz
1
Tx
Ty
Tz
1
jx
kx
jy
ky
jz
kz
Pi
Pj
Pk
1
ni
oi
ai
nj
oj
aj
nk
ok
ak
WI
WJ
WK
WN
WO
WA
1
WX
WN
Y
O
W
W
WZ
WA
1
1
ix
i
y
kx
jy
ky
iz
jz
kz
jx
Tx
Ty
Tz
1
Pi
ni
oi
ai
nj
oj
aj
nk
ok
ak
Pk
Pj
1 0 0 Tx i x
0 1 0 T i
y y
H
0 0 1 Tz i z
0
0
0
1
jx
kx
jy
ky
jz
kz
0 1
0 0
0 0
1 0
Pi n i
1 0 Pj n j
0 1 Pk n k
0 0 1 0
0 0
oi
ai
oj
aj
ok
ak
0
0
0
H = (Translation relative to the XYZ frame) * (Rotation relative to the XYZ frame)
* (Translation relative to the IJK frame) * (Rotation relative to the IJK frame)
WN
T
X
K
P
A
O
W
WA
This method might seem a bit confusing, but its actually an easier way to
solve our problem given the information we have. Here is an example
Forward Kinematics
The Situation:
You have a robotic arm that
starts out aligned with the xo-axis.
You tell the first link to move by 1
and the second link to move by 2.
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 its
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
X3
X2
Y1
X1
Y0
1
X0
3
Y2
2
X2
X3
3
X4
This takes you from the X0Y0 frame to the X4Y4 frame.
X1
Y0
Y1
1
X0
X
Y
0
0
H
0
1
Denavit-Hartenberg Notation
Z(i - 1)
Y(i -1)
X(i -1)
Yi
a(i - 1 )
Zi
Xi
di
ai
( i - 1)
IDEA: Each joint is assigned a coordinate frame. Using the DenavitHartenberg notation, you need 4 parameters to describe how a frame (i)
relates to a previous frame ( i -1 ).
THE PARAMETERS/VARIABLES:
, a , d,
The Parameters
Z(i - 1)
Y(i -1)
X(i -1)
( i - 1)
Yi
a(i - 1 )
Zi
Xi
di
ai
You can
align the
two axis
just using
the 4
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 perpendicular
to both axis-lines.
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)
Its 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.
If the link is prismatic, then a(i-1)
is a variable, not a parameter.
Z(i - 1)
Y(i -1)
X(i -1)
( i - 1)
Yi
a(i - 1 )
di
Zi
Xi
ai
i
2) (i-1)
Technical Definition: Amount of rotation around the common perpendicular so that
the joint axes are parallel.
i.e. How much you have to rotate around the X (i-1) axis so that the Z(i-1) is pointing in
the same direction as the Zi axis. Positive rotation follows the right hand rule.
3) d(i-1)
Technical Definition: The displacement
along the Zi axis needed to align the a(i-1)
common perpendicular to the ai common
perpendicular.
In other words, displacement along the
Zi to align the X(i-1) and Xi axes.
Z(i - 1)
Y(i -1)
X(i -1)
Yi Z
i
a(i - 1 )
di
Xi
( i - 1)
4) i
Amount of rotation around the Zi axis needed to align the X(i-1) axis with the Xi
ai
i
cos i
sin i cos (i 1)
sin i sin (i 1)
0
sin i
cos i cos (i 1)
cos i sin (i 1)
0
0
sin (i 1)
cos (i 1)
0
a(i 1)
sin (i 1) d i
cos (i 1) d i
1)
X(i -1)
( i -
Y Zi
a(i -
X a
i
1)
1)
3 Revolute Joints
Y2
Z1
Z0
X2
X0
d2
X1
Y0
Y1
a0
Denavit-Hartenberg Link
Parameter Table
a1
i
(i-1)
a(i-1)
di
a0
-90
a1
d2
Y2
Z1
Z0
X2
X0
X1
Y0
Y1
a0
a1
X 0 Y0 Z 0
V X2
Y2
V
V Z2
(i-1)
a(i-1)
di
a0
-90
a1
d2
d2
T ( 0T)( 01T)(12T)
Note: T is the D-H matrix with (i-1) = 0 and i = 1.
(i-1)
a(i-1)
di
a0
cos1
sin
1
0
1
0
-90
a1
sin1
cos1
0
0
d2
0 a0
0 0
0 0
0 1
cos 0 sin 0
sin
cos 0
0
0T
0
0
0
0
0
0
1
0
0
0
0
cos 2
0
1
2T
sin 2
sin 2
0
cos 2
0
0 a1
1 d 2
0 0
0 1
Inverse Kinematics
From Position to Angles
A Simple Example
Revolute and
Prismatic Joints
Combined
Finding :
y
arctan( )
x
More Specifically:
(x , y)
y
arctan 2( )
x
1
X
Finding S:
S (x 2 y 2 )
Find:
l2
l1
Given: l1, l2 , x , y
1, 2
Redundancy:
A unique solution to this problem
does not exist. Notice, that using the
givens two solutions are possible.
Sometimes no solution is possible.
(x , y)
l2
l2
l1
l1
l2
(x , y)
c 2 a 2 b 2 2ab cos C
2
( x 2 y 2 ) l1 l2 2l1l2 cos(180 2 )
cos(180 2 ) cos( 2 )
2
l1
sin B sin C
b
c
sin 1 sin(180 2 )
sin( 2 )
l2
x 2 y2
x 2 y2
1 1
y
arctan 2
x
x 2 y 2 l1 l2
cos( 2 )
2l1l2
x 2 y 2 l12 l2 2
2 arccos
2l1l2
1 arcsin
l sin( )
2
2
y
arctan 2
x
x 2 y 2
l2
(x , y)
c1 cos1
c1 2 cos( 2 1 )
(1) x l1 c1 l2 c1 2
l1
(2) y l1 s1 l2 sin 1 2
(3) 1 2
(1) 2 (2) 2 x 2 y 2
l1 l2 2l1l2 c 2
Only Unknown
x 2 y 2 l12 l2 2
2 arccos
2
l
l
1 2
Note :
cos(ab) (cos a )(cos b) (sin a )(sin b)
x l1 c1 l2 c1 2
l1 c1 l2 c1c 2 l2 s1s2
c1 (l1 l2 c 2 ) s1 (l2 s2 )
y l1 s1 l2 sin 1 2
l1 s1 l2 s1c 2 l2 s 2 c1
c1 (l2 s 2 ) s1 (l1 l2 c 2 )
c1
x s1 (l 2 s 2 )
(l1 l 2 c 2 )
x s1 (l 2 s 2 )
y
(l 2 s 2 ) s1 (l1 l 2 c 2 )
(l1 l 2 c 2 )
s1
Note :
cos(a b) (cos a )(cos b) (sin a )(sin b)
1
x l 2 s 2 s1 (l12 l 2 2 2l1l 2 c 2 )
(l1 l 2 c 2 )
y(l1 l 2 c 2 ) x l 2 s 2
2
x y
y(l1 l 2 c 2 ) x l 2 s 2
1 arcsin
x y