100% found this document useful (1 vote)
996 views23 pages

Tool Configuration Jacobian

The document discusses tool configuration vectors and Jacobians. It defines: - Tool configuration as a pair of position and orientation vectors relative to the base. - The tool configuration Jacobian which maps joint velocities to tool velocity. - The manipulator Jacobian which relates infinitesimal joint displacements to linear and angular tool displacements. It is used to relate external forces/moments at the tool to joint torques/forces.

Uploaded by

Jisha V R
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
100% found this document useful (1 vote)
996 views23 pages

Tool Configuration Jacobian

The document discusses tool configuration vectors and Jacobians. It defines: - Tool configuration as a pair of position and orientation vectors relative to the base. - The tool configuration Jacobian which maps joint velocities to tool velocity. - The manipulator Jacobian which relates infinitesimal joint displacements to linear and angular tool displacements. It is used to relate external forces/moments at the tool to joint torques/forces.

Uploaded by

Jisha V R
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 23

Tool Configuration vector

• Tool configuration is represented by the pair {p,R},


where p represents the tool position relative to the
base and R represents the orientation relative to the
base
• Specifying tool configuration with a rotation matrix
 2/3 information redundant
• Alternate compact representation

1
Tool Configuration vector

2
Tool Configuration vector
• Last column of R, Approach vector direction n
which tool pointed
• Rotation abt approach vector-roll
• It contains pitch and yaw information
• Need to augment roll information with this approach
vector
• If simply added , 4 components, not a minimal
representation X

3
Tool Configuration vector
• The size or length of approach vector can be scaled
by a positive quantity without changing direction
• We use an invertible function of tool roll angle, qn

• f-1 exists

4
Tool Configuration vector

5
Tool Configuration vector
• Three axis planar articulated robot

C123 S123 0 a1C1  a2C12 


S C123 0 a1 S1  a2 S12 
Tool
Tbase   123

 0 0 1 d3 
 
 0 0 0 1 
a1C1  a2C12 
a S  a S 
 1 1 2 12 
w (q )   d3 
 0 
 0 
 q3
( ) 
 e 
 6
Tool Configuration vector
• Five axis robot

7
Tool configuration Jacobian
• Once a trajectory is planned in the tool configuration
space one must then address the problem of
commanding the robot to follow that trajectory

• One approach is to differentiate the direct kinematic


equations to produce an expression for tool
configuration velocity in terms of joint velocity

• The transformation from tool configuration velocity to


joint velocity is called Tool Configuration Jacobian
Matrix (not a square matrix)
8
Tool configuration Jacobian(contd…)
• At certain locations in joint space tool
configuration matrix loses rank and these points
are referred to as singular configurations of the
arm

• If x  R 6 denote the tool configuration vector,


then the direct kinematic equations can be
formulated as x=w(q)

• Given x(t) find q(t) solving inverse kinematic


problem
9
Tool configuration Jacobian (contd…)
• But for driving joints we need velocity information as
well.
• Differentiating with respect to time x  V (q )q

• V(q) is a 6×n matrix of partial derivatives of w with


respect to q referred to as the tool configuration
Jacobian matrix
Operating point(q)

V(q)
Joint velocity Tool velocity

10
Tool configuration Jacobian (contd…)
• Linear transformation which maps instantaneous
joint space velocity into instantaneous tool velocity
• For a 3 axis planar robot C123 S123 0 a1C1  a2C12 
S C 0 a S  a S 
Tool
Tbase   123 123 1 1 2 12 

 0 0 1 d3 
 
 0 0 0 1 
a1C1  a2C12   a1 S1  a 2 S12  a 2 S12 0 
a S  a S   a C a C a 2C12 0 
 1 1 2 12   1 1 2 12
 0 
w (q )   
d3 0 0
  V (q )   0 0 0 

0 
 0   0 0 0 
 q3
( )   q3 
 e 
  0 0
e 
  
11
Manipulator Jacobian

• spray painting robot- force & moment generated at the


tool tip because of load carried by the robot
12
Manipulator Jacobian
• To develop a relationship between the external forces
and moments applied at the tool tip and the internal joint
torques and forces they induce

• Let dp  R3 represent an infinitesimal translation of tool


tip expressed in base co-ordinates

• d  R 3 represent infinitesimal rotation of the tool also


expressed as infinitesimal rotation of the tool

• Combining both into a six dimensional vector du called


the infinitesimal displacement vector du   
dp
d 

13
Manipulator Jacobian (contd…)
• Thus du represent both linear and angular
displacement of the tool

• The infinitesimal tool displacement is directly related


to the infinitesimal joint displacement
du  J (q)dq J(q) is the manipulator Jacobian
q  R 6 & J (q)  6  nMatrix

Operating point(q)

J(q)
Joint Displacement Tool Displacement
14
Manipulator Jacobian (contd…)
 A(q)
• Let J (q)    where, dp  A(q)dq & d  B(q)dq
 B(q)

• A(q) is the part of the manipulator Jacobian


associated with linear tool displacement and B(q) is
the part of the manipulator Jacobian associated with
angular tool displacement

dp dq
•  A(q ) The 3×n Jacobian submatrix A(q)can be
dt dt
interpreted as a linear transformation which maps

 to tool tip velocity p
instantaneous joint velocity q

15
Manipulator Jacobian (contd…)
• Differentiating tool tip position function p(q) with
respect to time p
 dp ( q ) 
 (q)   q
 dq 

A( q )  dp ( q ) 

 dq 

pk ( q )
Akj ( q )  ;1  k  3 & 1  j  n
q j

• If bk(q) denotes the kth column of B(q)


n

d  B(q)dq d   dqk b k (q)


k 1

16
Manipulator Jacobian (contd…)
• bk(q) represents a set of gains or constants of
proportionality which relate the displacement of
joint k to the angular tool displacement dφ when all
other joints are motionless
• If joint k is prismatic no angular tool displacement
induced by the linear motion along zk-1 axis

17
Manipulator Jacobian (contd…)
• Hence bk(q) can be written as
b k (q)   k z k 1 (q);1  k  n
• k  0 Prismatic Jt
•  k 1 Revolute Jt
• Let i3 be the co-ordinates of zk-1 with respect to
frame Lk-1.
• Then transforming to base co-ordinates
k 1 k 1
z (q )  R
0 (q ) i 3

18
Manipulator Jacobian (contd…)
• Let F tool  R6 denote the vector of external forces
and moments applied at the tool tip due to either a
load or contact with the environment
 f tool 
F tool   tool 
 
• ftool 3×1 force vector acting at the tool tip
• ηtool 3×1 moment vector acting at the tool tip
• Let   R 6 denote the vector of joint torques and
forces
• Let J(q) be the manipulator jacobian matrix of the
robot arm with frictionless joints
19
Manipulator Jacobian (contd…)
• If joint k is revolute τk represents the torque about k
axis
• If joint k is prismatic τk represents the force along k
axis

20
Manipulator Jacobian (contd…)
• The work performed which has units of energy must
be same regardless of the coordinate system within
which it is measured or expressed.
• The work done by an infinitesimal displacement du
of the tool is

• Alternately work done by the corresponding


infinitesimal displacement dq of the joints is

21
Manipulator Jacobian (contd…)

• But

22
Manipulator Jacobian (contd…)
• Then   J T (q)F tool


Operating point(q)
F tool JT(q)
Tool Forces & Moments Joint torques & forces

• If only one among these can be measured


using sensors, other can be calculated.

23

You might also like