Ch. 3: Forward and Inverse Kinematics: KON 318E: Introduction To Robotics 1

İTÜ-EEF Dept.
Of Control Engineering
Ch. 3: Forward and Inverse

Kinematics
KON 318E : Introduction to Robotics 1 1

İTÜ-EEF Dept. Of Control Engineering
Recap: Rigid Motions

• Rigid motion is a combination of rotation and translation
– Defined by a rotation matrix (R) and a displacement vector (d)
– the group of all rigid motions (d,R) is known as the Special Euclidean group,
SE(3)
• We can represent rigid motions (rotations and translations) as
matrix multiplication
• The matrix multiplication H is known as a homogeneous transform
and we note that R T d 
H 
 0 1 
1 R T  RT d 
H  
• Inverse transforms: 0 1 

Recap:Homogeneous Transforms
• Basic transforms:
– Three pure translation, three pure rotation
1 0 0 a 1 0 0 0
0 0 c 0
1 0 0   s
Trans x ,a  Rot x ,  
0 0 1 0 0 s c 0
   
0 0 0 1 0 0 0 1
1 0 0 0  c  0 s 0
0  0 0
1 0 b  1 0
Trans y ,b  Rot y ,  
0 0 1 0  s  0 c  0
   
0 0 0 1  0 0 0 1
1 0 0 0 c  s 0 0
0 1 0 0 s c 0 0
Trans z,c  Rot z,   
0 0 1 c 0 0 1 0
   
0 0 0 1 0 0 0 1

Example
• Euler angles: we have only discussed ZYZ Euler angles. What is the set of
all possible Euler angles that can be used to represent any rotation matrix?
– XYZ, YZX, ZXY, XYX, YZY, ZXZ, XZY, YXZ, ZYX, XZX, YXY, ZYZ
– ZZY cannot be used to describe any arbitrary rotation matrix since two
consecutive rotations about the Z axis can be composed into one rotation

Example
• Compute the homogeneous transformation representing a translation of 3
units along the x-axis followed by a rotation of p/2 about the current z-axis
followed by a translation of 1 unit along the fixed y-axis
T  Ty ,1Tx,3Tz ,p / 2
1 0 0 0  1 0 0 3 0 1 0 0
0 1 0 1  0 1 0 0  1 0 0 0 

0 0 1 0 0 0 1 0 0 0 1 0
   
0 0 0 1  0 0 0 1  0 0 0 1
1 1 0 3
1 0 0 1 

0 0 1 0
 
0 0 0 1

Forward Kinematics Introduction

• Challenge: given all the joint parameters of a manipulator,
determine the position and orientation of the tool frame
– Tool frame: coordinate frame attached to the most distal link of the
manipulator
– Inertial frame: fixed (immobile) coordinate frame fixed to the most
proximal link of a manipulator
• Therefore, we want a mapping between the tool frame and the

inertial frame
– This will be a function of all joint parameters and the physical geometry
of the manipulator
• Purely geometric: no need to worry about joint torques or dynamics

yet !!!

Convention
• A n-DOF manipulator will have n joints (either revolute or prismatic) and n+1
links (since each joint connects two links)
joint i-1 joint i
link i-1 link i
• We assume that each joint only has one DOF.
• Although this may seem like it does not include spherical or

universal joints, we can consider multi-DOF joints as a combination
of 1DOF joints with zero length

Convention
– “Frame o0” is the inertial frame
– “Frame on” is the tool frame
– “Joint i ” connects “link i-1” and “link i”
– “Frame oi” is connected to link i
• Joint variables, qi
 if joint i is revolute
qi   i
d i if joint i is prismatic

Convention
• A homogeneous transformation allows to express the position and orientation
of oj with respect to oi
– It is desired that the position and orientation of the tool frame with respect to the
inertial frame
– An intermediate step is to determine the transformation matrix that gives position
and orientation of oi with respect to oi-1: Ai ( Homogeneous Trans.)
zi
zi-1
Ai
yi
oi
oi-1 yi-1 xi

xi-1  Ai 1 Ai  2 ... Aj i Aj if i  j
– Now we can define the transformation oj to oi as: T j   if i  j
i
I

 T j i 1 if j  i

Convention
• Finally, the position and orientation of the tool frame with respect to the
inertial frame is given by one homogeneous transformation matrix:
– For a n-DOF manipulator
Rn0 on0 
H  n  A1 q1 A2 q 2     An q n 
0 qi :Joint variables,
 T
0 1
“FORWARD KINEMATIC MODEL”
• Ai homogeneous Trans formation Matrix is not a constant matrix, it is a

function of qi joint variable Ai(qi)
• Thus, to fully define the forward kinematics for any serial manipulator, all we
need to do is create the Ai transformations and perform matrix multiplication

Convention
• Each Homogeneous transformation matrix in Tn0 :
R i 1
o 
i 1
Ai   i i

 0 1 
 i
Rj o 
i
T j  Ai 1 qi 1     A j qn   
i j

0 1
Rii1    R jj 1 o ij 1  R ij 1  o jj 1
Question : How the coordinate frames are attached onto the links??

The Denavit-Hartenberg (DH) Convention

Aim : to develop Forward Kinematic model ( Tn0 )
DEFINITION : DH convention is a method selecting frames
• Representing each individual homogeneous transformation as the

product of four basic transformations:
Ai  Transzi1 , di Rot zi1 , i Transxi , ai Rot xi , i


• Four DH parameters of linki and jointi:
– ai: link length
– i: link twist
– di: link offset
– i: joint angle
• Since each Ai is a function of only one variable, three of these will

be constant for each link
– di will be variable for prismatic joints and i will be variable for revolute
joints
• But we said any rigid body needs 6 parameters to describe its

position and orientation
– Three angles (Euler angles, for example) and a 3x1 position vector
– So how can there be just 4 DH parameters?...

Assigning Coordinate Frames

• AIM : Choosing “Frame 0” ….. “Frame n” for a given robot
manipulator
• The choice is not unique, but the result will always be the same
• The following figure, a part serial robot, are considered for

generating D-H frames.
Link_i
Link_i+1
Joint_i Joint_i+1


• Assingning coordinate frames can be presented as an algorithm
given below:
Start:
for i=0 to n-1
STEP 1 : Assign zi axes
end :
STEP 2 : Establis the base frame
for i=1 to n-1
STEP 3 : Locate origin oi
STEP 4 : Assign xi axes
STEP 5 : Assign yi axes
end
STEP 6 : Establis the “Tool Frame”
STEP 7 : Create DH Table
end.

STEP 1 : Assign zi axis
• Choose zi as the motion axis (= axis of actuation) of joint i+1
• Repeat this from z0 to zn-1
– If joint i+1 is revolute, zi is the axis of rotation of joint i+1
– If joint i+1 is prismatic, zi is the axis of translation for joint i+1
Joint_i Joint_i+1
Zi
Link i-1
Link i+1
Link i
Zi-1


STEP 2 : Base Frame
• Choice of base frame nearly arbitrary
• We may choose o0 of the base frame to be any point on z0
• Then choose x0 and y0 by keeping right hand rule
1
z0
x0 y0

(Now, an iterative process starts at base, assign frames
for each link from base to end-effector)
• Locate oi : an intersection of zi axes by the common
normal between zi and zi-1

Joint_i+1
Joint_i Zi
Link i-1
Link i+1
Link i
Zi-1
oi Common
normal


• Two special cases :
1. If zi intersect zi-1 (no comon normal) then locate oi at this
intersection
zi-1 oi
zi


• Two special cases :
2. If zi and zi-1 are parallel then locate oi any convinient position
along zi
zi-1 zi
oi


STEP 4 : Assign xi axis
• Define xi axis along the common normal between zi and
zi-1 axes
• Direction : from zi-1 to zi

Joint_i
Joint_i+1
Link i-1
Link i+1
Link i
Zi-1 Zi
oi
Common xi
normal


STEP 4 : Assign xi axis
• Specia Case : İf zi intersect zi-1 then define direction of xi
as a normal to zi and zi-1 plane
zi
zi-1
oi
xi


STEP 5 : Assign yi axis
• Define yi axis by keeping right-hand rule

Joint_i Joint_i+1
Link i-1
Link i+1
Link i
Zi-1 Zi
yi
oi
Common xi
normal


STEP 6 : Establish “Tool Frame” ( on)
• The previous assignments are valid up to frame n-1
• “TOOL FRAME” : Frame to which the gripper is attached
– The tool frame assignment is most often defined by the axes n, s, a:
• a is the approach direction
• s is the ‘sliding’ direction (direction along which the grippers open/close)
• n is the normal direction to a and s

D-H Parameters (D-H Table)

STEP 7 : Create DH Table
• DH parameters can be divided into 2 groups :

– Link Parameters : ai , i
– Joint Parameters : di , i
• Physical Basis for DH Parameters :

– ai : distance from intersection of zi-1 and xi to oi along xi
– i : angle from zi-1 to zi around xi
– di :distance from oi-1 to intersection of zi-1 and xi along zi-1
– i : angle from xi-1 to xi around zi-1

Link Parameters :
– ai : distance from intersection of zi-1 and xi to oi along xi
– i : angle from zi-1 to zi around xi i+1
zi
Z’i
Zi-1 Joint_i+1
Joint_i Zi+1
Link i Joint_i+2
zi+1
yi+1
i oi+1 xi+1
zi
ai+1
yi
ai oi xi

Joint Parameters :
– di :distance from oi-1 to intersection of zi-1 and xi along zi-1
– i : angle from xi-1 to xi around zi-1
i Joint_i+1
i+1
Link i Link i+1
Joint_i
Zi-1
xi-1
yi+1
zi+1 di+1
yi-1
di+1 i+1 xi+1
di zi
i yi
xi

DH Parameters : Positive Conventions

(-)
(+)
ai : distance from intersection of zi-1 and xi to oi along xi

di :distance from oi-1 to intersection of zi-1 and xi along zi-1
DH Parameters : Positive Conventions
i : angle from zi-1 to zi around xi

i : angle from xi-1 to xi around zi-1
DH Frames and Parameters



D-H Table :
• Described by 4 parameters (short names):
– i : twist angle
– ai : link length
– di : joint offset
– i : joint angle
• Joint variable is di or i ( takes “*” symbol)
• Build DH Table with values for each link:
Link a  d 
1 L1 90o 0 1*
2 0 0 d2* 0
    

Link Transformations
Joint_i
Joint_i+1
Link i Link i+1

Joint_i
Zi-1 yi+1
zi+1
xi-1
yi-1
xi+1
zi
yi
Ai+1
Ai xi
Ti-1i+1=Ai Ai+1
Link Transformations
Joint_i
Joint_i+1
Link i Link i+1

Joint_i
Zi-1 yi+1
zi+1
xi-1
yi-1
xi+1
zi
yi
Ai xi
Ai  Transzi1 , di Rotzi1 , i Transxi , ai Rotxi , i


Ai  Trans zi1 ,di Rot zi1 ,i Trans xi ,ai Rot xi ,i
1 0 0 0  ci  s i 0 0 1 0 0 ai  1 0 0 0
0 1 0 0   si ci 0 0 0 1 0 0  0 c i  s i 0

0 0 1 di   0 0 1 0  0 0 1 0  0 s i c i 0
    
0 0 0 1  0 0 0 1  0 0 0 1  0 0 0 1
ci  si c i si s i ai ci 
s ci c i  ci s i ai si  Ai is function of i , di , i and ai
  i
0 s i c i di 
 
 0 0 0 1 
Example 1: Two-Link Planar Manipulator

• 2DOF: need to assign three coordinate frames
1. Choose z0 axis (axis of rotation for joint 1, base frame)
2. Choose z1 axis (axis of rotation for joint 2)
3. Choose z2 axis (tool frame)
• This is arbitrary for this case since we have described no wrist/gripper
• Instead, define z2 as parallel to z1 and z0 (for consistency)
4. Choose xi axes
• All zi’s are parallel
• Therefore choose xi to intersect oi-1


y2
x2
y0
y1
x1
x0


• Now define DH parameters
– First, define the constant parameters ai, i
– Second, define the variable parameters i, di
link ai  i di i
1 a1 0 0 1
2 a2 0 0 2
– The i terms are 0 because all zi are parallel
– Therefore only i are variable
c1  s1 0 a1c1  c2  s2 0 a2c2  T10  A1

s c1 0 a1s1  s c2 0 a2s2  c12  s12 0 a1c1  a2c12 
A1   1 , A2   2 s
0 0 1 0  0 0 1 0  c12 0 a1s1  a2s12 
    T2  A1A2   12
0
0 0 1 0 
0 0 0 1  0 0 0 1 
 
0 0 0 1 

DH Example 2:
3 revolute joints
joint 1 Shown in home position
R
Link 2
Link 1 Link 3
joint 2 joint 3
L1 L2

DH Example 2:
Zi is axis of actuation for joint i+1
Z0 Z0 and Z1 are not co-planar

Z1 and Z2 are parallel
1
2 3
Z1 Z2

DH Example 2:
Z0 and Z1 are not co-planar:
Z0 x0 is the common normal
1
x1 x2 x3
x0
2 3
Z1 Z3
Z2
Z2Zand
1 and Z3Zare
2 are parallel
parallel : :
x1 is selected
x2 is selected on the common
on the common normal that
normal that
lies along lies
the alongofthe
center thecenter
link of
the link

DH Example 2:
Shown with joints in non-zero positions
Z0 x3
z3
2 3
x2
x1
1 Z2
x0
Z1
Observe that frame i moves with link i

DH Example 2:
• Link lengths given
• 1 = 90o (rotate by +90o around x0 to
align Z0 and Z1)
R
Z0
L1 L2
x1 x2 x3
x0
1
Z1 Z3
Z2

DH Example 2:
Build table
R
Z0
L1 L2
1
x1 x2 x3
1 x0
2 3
Z1 Z3
Z2
Link  d  a
1 1* 0 90o R
2 2* 0 0 L1
3 3* 0 0 L2
DH Example 2:
ci  si c i si s i ai ci 

s ci c i  ci s i 
ai si 
i 1
Ai  Ti   i
0 s i c i di 
 
0 0 0 1 

DH Example 2:
c1 0 s1 Rc1 
s 0  c1 Rs1 
A1  T10   1
0 1 0 0 
 
0 0 0 1 
c 2  s 2 0 L1c 2 
s c 2 0 L1s 2 
A2  T21    2
0 0 1 0 
 
0 0 0 1 
c3  s 3 0 L2 c3 
s c3 0 L2 s3 
A3  T32   3
0 0 1 0 
 
0 0 0 1 
DH Example 2: x3
z0
y2 z3
y1 3
x2
y0 2
x1
1 z2
x0
z1
x1 axis expressed y1 axis expressed z1 axis expressed

wrt {0} wrt {0} wrt {0}
c1 0 s1 Rc1 

s Origin of {1}
 1 0  c1 Rs1  w.r.t. {0}
A1  T1 
0
0 1 0 0 
 
0 0 0 1 
DH Example 2: x3
z0
y2 z3
y1 3
x2
y0 2
x1
1 z2
x0
z1
wrt {1} wrt {1} wrt {1}
c 2  s 2 0 L1c 2 
s Origin of {2}
 2 c 2 0 L1s 2  w.r.t. {1}
A2  T2 
1
0 0 1 0 
 
0 0 0 1 
DH Example 2: x3
z0
y2 z3
y1 3
x2
y0 2
x1
1 z2
x0
z1
wrt {2} wrt {2} wrt {2}
c3  s3 0 L2c3 

Origin of {3}
s c3 0 L2 s3  w.r.t. {2}
A3  T32   3
0 0 1 0 
 
0 0 0 1 
DH Example 2:
T30  A1 A2 A3
c1 c 23  c1 s 23 s1 
c1 L2 c 23  L1c 2  R  

s c
 1  23  s1 s 23  c1  
s1 L2 c 23  L1c 2  R  
 s 23 c 23 0 L2 s 23  L1c 2 
 
 0 0 0 1 
where
ci  cos i 
si  sin  i 
c 23  cos 2   3 
s 23  sin  2   3 
Example 3: Three-Link Cylindrical Robot


• 3DOF: need to assign four coordinate frames
2. Choose z1 axis (axis of translation for joint 2)
3. Choose z2 axis (axis of translation for joint 3)
4. Choose z3 axis (tool frame)
• This is again arbitrary for this case since we have described no wrist/gripper
• Instead, define z3 as parallel to z2


– First, define the constant parameters ai, i link ai i di i
– Second, define the variable parameters i, di 1 0 0 d1 1
2 0 -90 d2 0
c1  s1 0 0 1 0 0 0 1 0 0 0 3 0 0 d3 0
s c1 0 0  0 0 1 0  0 1 0 0 
A1   1 , A2   , A3  
0 0 1 d1  0  1 0 d2  0 0 1 d3 
     
0 0 0 1 0 0 0 1 0 0 0 1
c1 0  s1  s1d 3 
s 0 c c d 
T3  A1A2 A3  
0 1 1 1 3 
 0  1 0 d1  d 2 
 
0 0 0 1 

Example 4: Spherical Wrist

– yaw, pitch, roll (4, 5, 6) all intersecting at one point o (wrist center)

Example 4: spherical wrist

– yaw, pitch, roll (4, 5, 6) all intersecting at one point o (wrist center)
4. Choose tool frame:
• z6 (a) is collinear with z5
• y6 (s) is in the direction the gripper closes
• x6 (n) is chosen with a right-handed convention


– First, define the constant parameters ai, i
link ai i di i
– Second, define the variable parameters i, di
4 0 -90 0 4
c 4 0  s4 0 c5 0  s5 0 c6  s6 0 0 5 0 90 0 5
s 0 c4 
0 s 0 c5 
0 s c6 0 0  6 0 0 d6 6
A4   4 , A5   5 , A6   6
0 1 0 0 0 1 0 0 0 0 1 d6 
     
0 0 0 1 0 0 0 1 0 0 0 1
c 4c5c6  s4s6  c 4 c 5 s6  s 4 c 6 c 4 s5 c 4 s5 d 6 
s c c  c s  s 4 c 5 s6  c 4 c 6 s 4 s5 s4s5d 6 
T6  A4 A5 A6   4 5 6
3 4 6
  s5 c 6 s5 c 6 c5 c5d 6 
 
 0 0 0 1 


• The forward kinematics of the spherical wrist were identical to a ZYZ Euler
angle transformation:
c 4c5c6  s4s6  c 4 c 5 s6  s 4 c 6 c 4 s5 c 4 s5 d 6 
s c c  c s  s 4 c 5 s6  c 4 c 6 s 4 s5 s4s5d 6 
T63  A4 A5 A6   4 5 6 4 6
  s5 c 6 s5 c 6 c5 c5d 6 
 
 0 0 0 1 
4  
 5  
c c c  s s  c c s  s c c s  6  
 
R   s c c  c s  s c s  c c s s 
  s c s s c 


Example 5: Matlab “Symbolic Toolbox”

• Find the robot forwaed kinematic model using MATLAB and “Symbolic
Toolbox”


• D-H parameter are defined after coor. frames assigments.
Link i i ai di i Ti i-1
1 90 0 d1 1 * T1o
2 0 0 d2* 0 T21
3 0 0 d3 3 * T32


• Transformations
– The resulting transformation matrices for the manipulator :


• MATLAB Code (Sybolic Toolbox) : Transformation
%Define symbolic variables
syms theta1 theta3 d1 d2 d3
%Define transformations:
T1_0 = [cos(theta1), 0, sin(theta1), 0; sin(theta1), 0, -cos(theta1), 0;0, 1, 0, d1; 0, 0, 0, 1];
T2_1 = [1, 0, 0, 0; 0, 1, 0, 0; 0, 0, 1, d2; 0, 0, 0, 1];
T3_2 = [cos(theta3), -sin(theta3), 0, 0; sin(theta3), cos(theta3), 0, 0; 0, 0, 1, d3; 0, 0, 0, 1];


• MATLAB Code: Transformation
%Determine transformation
T2_0 = T1_0 * T2_1;
T2_0 = simple(T2_0)
T3_0 = T2_0*T3_2;
T3_0 = simple(T3_0)
After executing the program, the output should be the following:

T2_0 =
[ cos(theta1), 0, sin(theta1), sin(theta1)*d2]
[ sin(theta1), 0, -cos(theta1), -cos(theta1)*d2]
[ 0, 1, 0, d1]
[ 0, 0, 0, 1]
T3_0 =
[ cos(theta1)*cos(theta3), -cos(theta1)*sin(theta3), sin(theta1), sin(theta1)*(d3+d2)]
[ sin(theta1)*cos(theta3), -sin(theta1)*sin(theta3), -cos(theta1), -cos(theta1)*(d3+d2)]
[ sin(theta3), cos(theta3), 0, d1]
[ 0, 0, 0, 1]


• MATLAB Code: Transformation
%give numeric values for symbolic variables
T2_0Numeric = subs(T2_0,{theta1,d1,d2},{0,0,0})
T3_0Numeric = subs(T3_0,{theta1,theta3,d1,d2,d3},{0,0,0,0,0})
After executing the program, the output should be the following:
T2_0Numeric = T3_0Numeric =
1 0 0 0 1 0 0 0
0 0 -1 0 0 0 -1 0
0 1 0 0 0 1 0 0
0 0 0 1 0 0 0 1


• MATLAB Code (MuPad) : Transformation
Define transformations:
T1_0 :=matrix([ [cos(theta1), 0, sin(theta1), 0],[ sin(theta1), 0, -cos(theta1), 0],[0, 1, 0, d1],[ 0, 0, 0, 1]]):
T2_1 :=matrix([ [1, 0, 0, 0],[ 0, 1, 0, 0],[0, 0, 1, d2],[ 0, 0, 0, 1]]):
T3_2 :=matrix([ [cos(theta3), -sin(theta3), 0, 0],[sin(theta3), cos(theta3), 0, 0],[ 0, 0, 1, d3],[0, 0, 0, 1]]):
Define Symbols:
theta1:=`θ_1`;theta3:=`θ_3`;d1:=`d_1`;d2:=`d_2` ;d3:=`d_3`:
print(`T_1^0 `=T1_0):
print(`T_2^1 `=T2_1):
print(`T_3^2 `=T3_2):
Define Forwar Kinematic model: (Total Transformations):

T3_0:=Simplify(T1_0 * T2_1 * T3_2):
print(`T_3^0 `=T3_0):
Obtain numerical values:
`θ_1`:=0;`θ_3`:=0;`d_1`:=1;`d_2`:=1;`d_3`:=1:
print(`T_3^0 `=T3_0):
( ps: after formulation choose `Copy as LaTex code` to create Word formulas ! )
Example 6: cylindrical robot with spherical wrist
• 6DOF: need to assign seven coordinate frames

– But we already did this for the previous two examples, so we can fill in the table
of DH parameters:
1 0 0 d1 1*
2 0 -90 d2* 0
3 0 0 d3* 0
4 0 -90 0 4*
o3, o4, o5 are all at
the same point oc 5 0 90 0 5*
6 0 0 d6 6*

Example 6: cylindrical robot with spherical wrist
• Note that z3 (axis for joint 4) is collinear with z2 (axis for joint 3), thus we can
make the following combination:
r11  c1c 4c5c 6  c1s4s6  s1s5c 6

r11 r12 r13 dx 
r r r23 d y  r21  s1c 4c5c 6  s1s4s6  c1s5c 6
T6  T3 T6   21 22
0 0 3
r31  s4c5c 6  c 4s6
r31 r32 r33 dz 
  r12  c1c 4c5s6  c1s4c 6  s1s5c 6
0 0 0 1
r22  s1c 4c5s6  s1s4s6  c1s5c 6
r32  s4c5c 6  c 4c 6
r13  c1c 4s5  s1c5
r23  s1c 4s5  c1c5
r33  s4s5
d x  c1c 4s5d 6  s1c5d 6  s1d 3
d y  s1c 4s5d 6  c1c5d 6  c1d 3
d z  s4s5d 6  d1  d 2

Example 7: the Stanford manipulator

• 6DOF: need to assign seven coordinate frames:
2. Choose z1-z5 axes (axes of rotation/translation for joints 2-6)
3. Choose xi axes
4. Choose tool frame
5. Fill in table of DH parameters:
1 0 -90 0 1*
2 0 90 d2 2*
3 0 0 d3* 0
4 0 -90 0 4*
5 0 90 0 5*
6 0 0 d6 6*


• Now determine the individual homogeneous transformations:
c1 0  s1 0 c 2 0 s2 0 1 0 0 0
s 0 c1 0 s 0  c 2 0  0 1 0 0 
A1   1 , A2   2 , A3  
0 1 0 0 0 1 0 d2  0 0 1 d3 
     
0 0 0 1 0 0 0 1 0 0 0 1
c 4 0  s4 0 c5 0 s5 0  c 6  s6 0 0
s 0 c4 0 s 0  c5 0 s c6 0 0 
A4   4 , A5   5 , A6   6
 0 1 0 0 0  1 0 0 0 0 1 d6 
     
0 0 0 1 0 0 0 1 0 0 0 1


• Finally, combine to give the complete description of the forward kinematics:
r11  c1c 2 c 4c5c 6  s4s6   s2s5c 6   d 2 s4c5c 6  c 4s6 

r21  s1c 2 c 4c5c 6  s4s6   s2s5c 6   c1 s4c5c 6  c 4s6 
r31  s2 c 4c5c 6  s4s6   c 2s5c 6
r12  c1 c 2 c 4c5s6  s 4c 6   s2s5s6   s1  s4c5s6  c 4c 6 
r11 r12 r13 dx  r22  s1 c 2 c 4c5s6  s4c 6   s2s5s6   c1  s4c5s6  c 4s6 
r d y 
r r23 r32  s2 c 4c5s6  s4c 6   c 2s5s6
T6  A1    A6   21 22
0
r31 r32 r33 dz  r13  c1 c 2c 4s5  s2c5   s1s4s5
 
0 0 0 1 r23  s1 c 2c 4s5  s2c5   c1s4s5
r33  s2c 4s5  c 2c5
d x  c1s2d 3  s1d 2  d 6 c1c 2c 4s5  c1c5s2  s1s4s5 
d y  s1s2d 3  c1d 2  d 6 c1s4s5  c 2c 4s1s5  c5s1s2 
d z  c 2d 3  d 6 c 2c5  c 4s2s5 

Example 8: the SCARA manipulator

• 4DOF: need to assign five coordinate frames:
2. Choose z1-z3 axes (axes of rotation/translation for joints 2-4)
3. Choose xi axes
4. Choose tool frame
5. Fill in table of DH parameters:
1 a1 0 0? 1*
2 a2 180 0 2*
3 0 0 d3* 0
4 0 0 d4 4*

Example 8: the SCARA manipulator

• Now determine the individual homogeneous transformations:
c1  s1 0 a1c1  c2 s2 0 a2c2  1 0 0 0 c 4  s4 0 0

s c1 0 a1s1  s  c2 0 a2s2  0 1 0 0  s c4 0 0 
A1   1 , A2   2 , A3   , A4   4
0 0 1 0  0 0 1 0  0 0 1 d3  0 0 1 d4 
       
0 0 0 1  0 0 0 1  0 0 0 1 0 0 0 1
c12c 4  s12s4  c12s4  s12c 4 0 a1c1  a2c12 

s c  c s  s12s4  c12c 4 0 a1s1  a2s12 
T4  A1    A4   12 4
0 12 4
 0 0  1  d3  d 4 
 
 0 0 0 1 

Example 9: Human Arm

• As shown, the human arm consists of a series of rigid links connected by three
anatomical joints (shoulder joint, elbow joint, and wrist joint) :
Kinematic models of the Human and Robot arms.

Example 9: Human Arm

• Human arm D-H parameters :
• KUKA-Arm model D-H parameters

Dual Arm Examples
DLR Humanoid: Justin


Ch. 3: Forward and Inverse Kinematics: KON 318E: Introduction To Robotics 1

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ch. 3: Forward and Inverse Kinematics: KON 318E: Introduction To Robotics 1

Uploaded by

Copyright:

Available Formats

İTÜ-EEF Dept.

Ch. 3: Forward and Inverse

KON 318E : Introduction to Robotics 1 1

Recap: Rigid Motions

KON 318E : Introduction to Robotics 2 2

KON 318E : Introduction to Robotics 3 3

KON 318E : Introduction to Robotics 4 4

KON 318E : Introduction to Robotics 5 5

Forward Kinematics Introduction

• Therefore, we want a mapping between the tool frame and the

• Purely geometric: no need to worry about joint torques or dynamics

KON 318E : Introduction to Robotics 6 6

link i-1 link i

• We assume that each joint only has one DOF.

• Although this may seem like it does not include spherical or

KON 318E : Introduction to Robotics 7 7

KON 318E : Introduction to Robotics 8 8

KON 318E : Introduction to Robotics 9 9

“FORWARD KINEMATIC MODEL”

• Ai homogeneous Trans formation Matrix is not a constant matrix, it is a

KON 318E : Introduction to Robotics 10 10

KON 318E : Introduction to Robotics 11 11

The Denavit-Hartenberg (DH) Convention

DEFINITION : DH convention is a method selecting frames

• Representing each individual homogeneous transformation as the

Ai  Transzi1 , di Rot zi1 , i Transxi , ai Rot xi , i

KON 318E : Introduction to Robotics 12 12

The Denavit-Hartenberg (DH) Convention

• Since each Ai is a function of only one variable, three of these will

• But we said any rigid body needs 6 parameters to describe its

KON 318E : Introduction to Robotics 13 13

Assigning Coordinate Frames

• The following figure, a part serial robot, are considered for

KON 318E : Introduction to Robotics 14 14

Assigning Coordinate Frames

Assigning Coordinate Frames

KON 318E : Introduction to Robotics 16 16

Assigning Coordinate Frames

Assigning Coordinate Frames

KON 318E : Introduction to Robotics 18 18

KON 318E : Introduction to Robotics 19 19

Assigning Coordinate Frames

KON 318E : Introduction to Robotics 20 20

Assigning Coordinate Frames

KON 318E : Introduction to Robotics 21 21

Assigning Coordinate Frames

KON 318E : Introduction to Robotics 22 22

KON 318E : Introduction to Robotics 23 23

Assigning Coordinate Frames

KON 318E : Introduction to Robotics 24 24

Assigning Coordinate Frames

KON 318E : Introduction to Robotics 25 25

KON 318E : Introduction to Robotics 26 26

Assigning Coordinate Frames

KON 318E : Introduction to Robotics 27 27

D-H Parameters (D-H Table)

• DH parameters can be divided into 2 groups :

• Physical Basis for DH Parameters :

KON 318E : Introduction to Robotics 28 28

KON 318E : Introduction to Robotics 29 29

KON 318E : Introduction to Robotics 30 30

DH Parameters : Positive Conventions