MEV442 Module 2

MEV442
Introduction to Robotics
Module 2
Dr. Santhakumar Mohan

Assistant Professor
Mechanical Engineering
National Institute of Technology Calicut
Jacobians: Velocities and static forces
{ Introduction
{ Notation for time-varying position and orientation
{ Linear and rotational velocity of rigid bodies
{ More on angular velocity
{ Motion of the links of a robot
{ Velocity “propagation” from link to link
{ Jacobians
{ Singularities
{ Static forces in manipulator
{ Jacobians in the force domain
{ Cartesian transformation of velocities and static
forces
1
Introduction
• In this chapter, we expand our consideration of
robot manipulators beyond static position problems.
• We examine the notions of linear and angular

velocity of a rigid body and use these concepts
to analyze the motion of a manipulator.
• We also will consider force acting on a rigid

body, and then use these ideas to study the
application of static force with manipulators.
• It turns out that the study of both velocity and

static forces leads to a matrix entity called the
Jacobian of the manipulator, which will be
introduced in this chapter.
2
Central Topic -
Simultaneous Linear and Rotational Velocity
• Vector Form
• Matrix Form
3
Definitions - Linear Velocity
• Linear velocity - The instantaneous rate of change
in linear position of a point relative to some frame.
4
Definitions - Linear Velocity
• The position of point Q in frame {A} is represented
by the linear position vector
• The velocity of a point Q relative to frame {A} is

represented by the linear velocity vector
5
Definitions - Angular Velocity - Vector
• Angular Velocity Vector: A vector whose direction is the
instantaneous axis of rotation of one frame relative to another
and whose magnitude is the rate of rotation about that axis.
6
Definitions - Angular Velocity
• Angular Velocity: The instantaneous rate of change
in the orientation of one frame relative to another.
7
Definitions - Angular Velocity
• Just as there are many ways to represent orientation (Euler
Angles, Roll-Pitch-Yaw Angles, Rotation Matrices, etc.) there
are also many ways to represent the rate of change in
orientation.
• The angular velocity vector is convenient to use because it

has an easy to grasp physical meaning. However, the matrix
form is useful when performing algebraic manipulations.
8
Linear & Angular Velocities - Frames
• When describing the velocity (linear or angular) of an
object, there are two important frames that are being
used:
– Represented Frame (Reference Frame) :

This is the frame used to represent (express)
the object’s velocity.
– Computed Frame
This is the frame in which the velocity is measured
(differentiate the position).
9
Frame - Velocity
• As with any vector, a velocity vector may be described in
terms of any frame, and this frame of reference is noted with
a leading superscript.
• A velocity vector computed in frame {B} and represented in

frame {A} would be written
Represented
(Reference Frame)
Computed
(Measured)
10
Linear Velocity - Rigid Body
• Given: Consider a frame {B}
attached to a rigid body whereas
frame {A} is fixed. The
orientation of frame {A}
with respect to frame {B} is not
changing as a function of time
• Problem: describe the motion of

of the vector relative to frame
{A}
• Solution: Frame {B} is located

relative to frame {A} by a position
vector and the rotation matrix
(assume that the orientation is not
changing in time ) expressing
both components of the velocity in
terms of frame {A} gives
11
Linear Velocity - Rigid Body
12
Angular Velocity - Rigid Body
• Given: Consider a frame {B}
attached to a rigid body whereas
frame {A} is fixed. The vector
is constant as view from frame
{A}
• Problem: describe the velocity

of the vector representing the
point Q relative to frame {A}
• Solution: Even though the vector

is constant as view from frame {B}
it is clear that point Q will have a
velocity as seen from frame {A}
due to the rotational velocity
13
Angular Velocity - Rigid Body -
Intuitive Approach
14
Intuitive Approach
15
Intuitive Approach
16
Intuitive Approach
17
Intuitive Approach
18
Intuitive Approach
19
Intuitive Approach
• The figure shows to instants
of time as the vector
rotates around .
This is what an observer in
frame {A} would observe.
• The Magnitude of the

differential change is
• Using a vector cross product

we get
20
Intuitive Approach
• In the general case, the vector Q may also be changing
with respect to the frame {B}. Adding this component
we get.
• Using the rotation matrix to remove the dual-superscript,

and since the description of at any instance is
we get
21
MORE ON ANGULAR VELOCITY
• A property of the derivative of an orthonormal matrix
R SR
Note: S is called the skew-symmetric matrix.
22
Velocity of a point due to rotating
reference frame
B
P cons tan t
• or
23
Skew-symmetric matrices and the
vector cross product
angular velocity matrix angular velocity vector
• Matrix Form • Vector Form
24
Gaining physical insight concerning the
angular velocity vector
Rotating about the

fixed frame: operator
ª kxkxvT cT kxkyvT kzsT kxkzvT kysT º

« » (2.80)
RK (T) «kxkyvT kzsT kykyvT cT kykzvT kxsT »
« »
k k
¬« x zvT kysT k k
y zvT kxsT k k
z zvT cT ¼»
sT sinT, cT cosT, vs 1 cosT AKˆ [kx , ky , kz ]T
25
ª 1 kz'T k y'T º
« »
RK ('T ) « kz'T 1 kx'T »
«k 'T kx'T 1 »
¬ y ¼
ª 0 k z 'T k y 'T º
« »
ª R ('T ) I3 º « k z 'T k x 'T »
R lim « k » R (t ) 0
't o 0 ¬ 't ¼ « k 'T k x 'T 0 »¼
lim ¬
y
R R(t )
't o0 't
ª 0 kzT kyT º ª 0 kzT kyT º

«

» 1 «
»
R « kzT 0 kxT » R(t) RR « kzT 0 kxT »
«k T k T » «k T k T 0 »
«¬ y x 0 »¼ «¬ y x »¼
26
ª 0 kzT kyT º ª 0 :z :y º
« »
RR1 « k T 0 k T» « : :
» angular velocity matrix
z x « z 0 x» S
«k T k T 0 » «: : 0 »¼
«¬ y x »¼ ¬ y x
ª : x º ª k xT º
« » « »
R SR : « : y » « k yT » T Kˆ angular velocity vector
« : » « k T »
¬ z¼ ¬ z ¼
• The physical meaning of the angular-velocity vector is that, at
any instant, the change in orientation of a rotating frame can be
viewed as a rotation about some axis K̂.
• This instantaneous axis of rotation, taken as a unit vector and then
scaled by the speed of rotation about the axis (T ) , yields the
angular-velocity vector.
27
Angular Velocity - Matrix & Vector Forms
S=
28
Simultaneous Linear and Rotational
Velocity
• The final results for the derivative of a vector in a moving
frame (linear and rotation velocities) as seen from a stationary
frame
• Vector Form
• Matrix Form
29
Simultaneous Linear and Rotational Velocity -
Vector Versus Matrix Representation
• Vector Form
• Matrix Form
30
Position Propagation
• The homogeneous transform matrix provides a complete
description of the linear and angular position relationship
between adjacent links.
• These descriptions may be combined together to describe

the position of a link relative to the robot base frame {0}.
• A similar description of the linear and angular velocities

between adjacent links as well as the base frame
would also be useful.
31
Motion of the Link of a Robot
• In considering the motion of a robot link we will always use link
frame {0} as the reference frame
Where: - is the linear velocity of the origin of link frame (i)
with respect to frame {0}
- is the angular velocity of the origin of link frame (i)
with respect to frame {0}
32
Frame – Velocity-review
• As with any vector, a velocity vector may be described in
terms of any frame, and this frame of reference is noted with
a leading superscript.
• A velocity vector computed in frame {B} and represented in

frame {A} would be written
Represented
(Reference Frame)
Computed
(Measured)
33
Velocities - Frame & Notation
• Expressing the velocity of a frame {i} (associated with link i )
relative to the robot base (frame {0}) using our previous
notation is defined as follows:
• The velocities differentiate (computed) relative to the base

frame {0} are often represented relative to other frames {k}.
The following notation is used for this conditions
34
Velocity Propagation
• Given: A manipulator - A
chain of rigid bodies each one
capable of moving relative to
its neighbour
• Problem: Calculate the linear

and angular velocities of the
link of a robot
• Solution (Concept): Due to

the robot structure we can
compute the velocities of
each link in order starting
from the base.
The velocity of link i+1 will be

that of link i , plus whatever
new velocity components were
35 added by joint i+1
Velocity of Adjacent Links - Angular
Velocity 1/5
• From the relationship developed previously
• we can re-assign link names to calculate the velocity of an

link i relative to the base frame {0}
• By pre-multiplying both sides of the equation by ,we can

convert the frame of reference for the base {0} to frame {i+1}
36
Velocity 2/5
• Using the recently defined notation, we have
-Angular velocity of frame {i+1} measured relative to the

robot base, and expressed in frame {i+1}
-Angular velocity of frame {i} measured relative to the robot

base, and expressed in frame {i+1}
-Angular velocity of frame {i+1} measured relative to frame

{i} and expressed in frame {i+1}
37
Velocity 3/5
• Angular velocity of frame {i} measured relative to the robot

base, expressed in frame {i+1}
In the second term
• Angular velocity of frame {i+1} measured (differentiate) in

frame {i} and represented (expressed) in frame {i+1} is shown
in the second term
38
Velocity 4/5
• Assuming that a joint has only 1 DOF. The joint configuration

can be either revolute joint (angular velocity) or prismatic joint
(Linear velocity).
• Based on the frame attachment convention in which we assign
the Z axis pointing along the i+1 joint axis such that the two are
coincide (rotations of a link is preformed only along its Z- axis) we
can rewrite this term as follows:
39
Velocity 5/5
The result is a recursive equation that shows the angular velocity
of one link in terms of the angular velocity of the previous link plus
the relative motion of the two links.
Since the term depends on all previous links through this

recursion, the angular velocity is said to propagate from the
base to subsequent links.
40
Velocity of Adjacent Links - Linear
Velocity 1/5
• Simultaneous Linear and Rotational Velocity
• The derivative of a vector in a moving frame (linear and rotation

velocities) as seen from a stationary frame
• Vector Form
• Matrix Form
41
Velocity 2/5
• From the relationship developed previously (matrix form)
• we re-assign link frames for adjacent links (i and i +1) with

the velocity computed relative to the robot base frame {0}
• By pre-multiplying both sides of the equation by , we can

convert the frame of reference for the left side to frame {i+1}
42
Velocity 3/5
-Linear velocity of frame {i+1} measured relative to

frame {i} and expressed in frame {i+1}
• Assuming that a joint has only 1 DOF. The joint configuration

can be either revolute joint (angular velocity) or prismatic joint
(Linear velocity).
• Based on the frame attachment convention in which we assign
the Z axis pointing along the i+1 joint axis such that the two are
coincide (translation of a link is preformed only along its Z- axis)
we can rewrite this term as follows:
43
Velocity 4/5
• With replacement:
Definition
44
Angular Velocity - Matrix & Vector Forms
S=
45
Velocity 5/5
• The result is a recursive equation that shows the linear velocity
of one link in terms of the previous link plus the relative motion
of the two links.
• Since the term depends on all previous links through this

recursion, the angular velocity is said to propagate from the
base to subsequent links.
46
Velocity of Adjacent Links - Summary
• Angular Velocity
0 - Prismatic Joint
• Linear Velocity
0 - Revolute Joint
47
Example-2R Robot
Find and
48
Example-2R Robot
49
Example-2R Robot
• For i=0
= =0
50
Example-2R Robot
• For
=0
51
Example-2R Robot
•For =0
• or
52
Example-2R Robot
•or
53
Angular and Linear Velocities - 3R
Robot - Example
• For the manipulator shown in the figure, compute the angular
and linear velocity of the “tool” frame relative to the base frame
expressed in the “tool” frame (that is, calculate
54
Robot - Example
• Frame attachment
55
Robot - Example
• DH Parameters
56
Robot - Example
• From the DH parameter table, we can specify the homogeneous
transform matrix for each adjacent link pair:
0
1 R
57
Robot - Example
• Compute the angular velocity of the end effector frame relative
to the base frame expressed at the end effector frame.
1 0 T
• For i=0 0 R 1 R
58
Robot - Example
59
Robot - Example
• Compute the linear velocity of the end effector frame relative
• Note that the term involving the prismatic joint has been
dropped from the equation (it is equal to zero).
60
Robot - Example
• For i=0
• For i=1
61
Robot - Example
62
Robot - Example
ª 0 L2 s3 0 º ªT1 º
« « »
0 L2c3 L3 L3 » «T2 » 4
J (T )T
« »
«¬ L1 L2c2 L3c23 0 0 »¼ «¬T3 »¼
63
Robot - Example
• Note that the linear and angular velocities ( ) of the end
effector where differentiate (measured) in frame {0} however
represented (expressed) in frame {4}
• In the car example: Observer sitting in the “Car”
Observer sitting in the “World”
64
Robot - Example
• Multiply both sides of the equation by the inverse transformation
matrix, we finally get the linear and angular velocities expressed
and measured in the stationary frame {0}
65
Kinematics Relations - Joint &
Cartesian Spaces
• A robot is often used to manipulate object attached to its tip
(end effector).
• The location of the robot tip may be specified using one of the
following descriptions:
• Joint Space
• Cartesian Space
Euler Angles
66
Kinematics Relations - Forward &
Inverse
• The robot kinematic equations relate the two description of the
robot tip location
Tip Location in Tip Location in

Joint Space Cartesian Space
67
Kinematics Relations - Forward &
Inverse
Tip Velocity in Tip velocity in

Joint Space Cartesian Space
68
Jacobian Matrix - Introduction
• The Jacobian is a multi dimensional form of the derivative.
• Suppose that for example we have 6 functions, each of which is

a function of 6 independent variables
• We may also use a vector notation to write these equations as
69
• If we wish to calculate the differential of as a function of the
differential we use the chain rule to get
• Which again might be written more simply using a vector

notation as
70
• The 6x6 matrix of partial derivative is defined as the Jacobian
matrix
• By dividing both sides by the differential time element, we

can think of the Jacobian as mapping velocities in X to those
in Y
• Note that the Jacobian is time varying linear transformation
71
• In the field of robotics the
Jacobian matrix describe the
relationship between the joint
angle rates ( ) and the
translation and rotation
velocities of the end effector
( ).
This relationship is given by:
72
• This expression can be expanded to:
• Where:
- is a 6x1 vector of the end effector linear and angular velocities
– is a 6xN Jacobian matrix
– is a Nx1 vector of the manipulator joint velocities
– N is the number of joints
73
How to obtain the Jacobian for a given
robot?
• According to definition of the Jacobian is given in the Eqs.

(5.58) through (5.62), a Jacobian can be obtained by directly
differentiating the kinematic equations of the mechanism of the
given robot. i.e.,
• The kinematic equations of a robot can be obtained via the

geometric approach or algebraic approach (chapter three).
• However, while this is straightforward for linear velocity, there

is no 3 × 1 orientation vector whose derivative is Ȧ.
74
How to obtain the Jacobian for a given
robot?
• Alternatively, we can use the velocity “propagation” method
to derive the Jacobian using successive application of the
recursive equation
• For revolute joint:
• For prismatic joint:
75
Jacobian Matrix - Calculation
Methods
76
Jacobian Matrix by Differentiation -
3R - 1/3 (Geometric approach)
• Consider the following 3 DOF Planar manipulator
x L1 cosT1 L2 cos(T1 T2 ) L3 cos(T1 T2 T3 )

y L2 sinT1 L2 sin(T1 T2 ) L3 sin(T1 T2 T3 )
77 D T1 T2 T3
3R - 2/3 (Geometric approach)
• The forward kinematics gives us relationship of the end effector
to the joint angles:
• Differentiating the three expressions gives
78
3R - 3/3
• Using a matrix form we get
• The Jacobian provides a linear transformation, giving a velocity map

and a force map for a robot manipulator.
• For the simple example above, the equations are trivial, but can
easily become more complicated with robots that have additional
degrees a freedom.
• Before tackling these problems, consider this brief review of linear
algebra.
79
Jacobian: Velocity propagation
• The recursive expressions for the adjacent joint linear and
angular velocities describe a relationship between the joint
angle rates ( T ) and the transnational and rotational
velocities of the end effector ( X ):
80
Jacobian: Velocity propagation
• Therefore the recursive expressions for the adjacent joint linear
and angular velocities can be used to determine the Jacobian in
the end effector frame
• This equation can be expanded to:
81
Robot - Example
82
Robot - Example
• Compute the angular velocity of the end effector frame relative
1 0 T
• For i=0 0 R 1 R
83
Robot - Example
ª 0 L2 s3 0 º ªT1 º
« « »
0 L2c3 L3 L3 » «T2 » 4
J (T )T
« »
«¬ L1 L2c2 L3c23 0 0 »¼ «¬T3 »¼
84
Jacobian: Velocities and singularities
• Mapping from velocities in

the joint space (joint) –to
velocities in the Cartesian
space (end effector) –
• At certain points, called

singularities, this
mapping is not invertible.
85
Singularity - The Concept
• Motivation: We would like the hand of a robot (end effecror) to

move with a certain velocity vector in Cartesian space. Using linear
transformation relating the joint velocity to the Cartesian velocity
we could calculate the necessary joint rates at each instance along
the path.
• Given: a linear transformation relating the joint velocity to the
Cartesian velocity (usually the end effector)
• Question: Is the Jacobian matrix invertible? (Or) Is it nonsingular?

Is the Jacobian invertible for all values of ?
If not, where is it not invertible?
86
How to find the singularities?
• The roots of the following equations determine the singularities:
• where Det[J (ș )] is the determinant of the Jacobian under

consideration.
• Certainly, a nonsquare Jacobian is not invertible. However, even

if a Jacobian is in a square matrix format, the determinant of the
Jacobian may still be zero.
87
What is the geometric or physical
meanings of singularities?
• Cleary, when a robot is in a singular configuration, it has
lost one or more degree of freedom as viewed from Cartesian
space.
• This means that there is some direction (or subspace) in

Cartesian space along which it is impossible to move the
hand of the robot no matter which joint rates are selected.
• It is obvious this happens at the workspace boundary of robots.
88
Where do singularities occur?
• All mechanisms have singularities at the boundary of their
workspace, and most have loci of singularities inside their
workspace. Hence singularities can be classified into two
categories:
1. Workspace boundary singularities are those which

occur when the manipulator is fully stretched out or
folded back on itself such that the end-effector is near
or at the boundary of the workspace.
2. Workspace interior singularities are those which

occur away from the workspace boundary and
generally are caused by two or more joint axes
lining up, or due to the mechanism structure so
that Det[J (ș )] = f (ș ) = 0 .
Viewed from
Cartesian
space
89
More on singularity
• Since different structures of the robot yields different link
parameters, therefore, Det[J (ș )] = f (ș ) = 0 occur in the
different places.
• By modifying the link parameters, one can eliminate the
singularities inside the workspace.
• Note: Det[J (ș )] = f (ș ) = 0 provides only one equations.
For f (ș ) = 0 ,
- if there is only one variable inside the equation, the singularity
is only at individual points;
- if there are two variables inside the equation, the singularities
occur alone a line;
- if more than three variables, the singularities form a hyper-
surface.
90
Singularity -Summary
• Losing one or more DOF means that there is a some direction

(or subspace) in Cartesian space along which it is impossible to
move the hand of the robot (end effector) no matter which joint
rate are selected
91
Singularity -Summary
• Answer (Conceptual): Most manipulator have values of where
the Jacobian becomes singular . Such locations are called
singularities of the mechanism or singularities for short
92
Properties of the Jacobian -
Velocity Mapping and Singularities
• Where are the singularities?
• What is the physical explanation
of the singularities?
• Are they workspace boundary or
interior singularities?
• From example 5.3
93
•They are workspace boundary singularities because they exist at

the edge of the manipulator’ workspace.
• At both case, the motion of the end-effector is possible only along

one Cartesian direction (the one perpendicular to the arm). Therefore,
the mechanism has lost one degree of freedom.
• At singularity configuration, the inverse Jacobian blows up! This
results in joint rates approaching infinity as the singularity is
approached. It is very dangers in a robot control system.
94
95
• Example: Planar 3R
96
• The manipulator loses 1 DEF. The end effector can only move along
the tangent direction of the arm. Motion along the radial direction is
not possible.
97
Robot - Example
Workspace Interior Singularities?

98
Robot - Example
ª 0 L2s3 0 º ªT1 º
« 0 L c L L » «T » 4
J (T)T
« 2 3 3 3 »« 2»
«¬L1 L2c2 L3c23 0 0 »¼ «¬T3 »¼
0 0 4
Det ( V4 ) Det ( R V4 )
4 Det ( R ) Det ( J (T )T)
0
4
4
0 det(4
J (T )) 0
det( 4 J (T )) ( L`1 L2 c 2 L3 c 23 ) L2 L3 s 3 0
L`1 L2c2 L3c23 0 L2L3s3 0
Draw the locus of the singular points of the robot in its workspace (or in
its base frame). (Interior singularities?)
99
Robot - Example
How to find
0
V4 or 0
J (T ) ?
0
X Tip ( L1 L2 cos T 2 L3 cos( T 2 T 3 )) cos T 1 top view
Geometric approach: 0
YTip ( L1 L2 cos T 2 L3 cos( T 2 T 3 )) sin T 1 top view
0
Z Tip L2 sin T 2 L3 sin( T 2 T 3 ) side view
0
V4 0
J (T )T dPTip / dt
100
Jacobian: Velocities and Static Forces
• Mapping from velocities in

the joint space (joint) –to
velocities in the Cartesian
space (end effector) –
• Mapping from the

joint force/torques –
Ĳ to forces/torque in
the Cartesian space
applied on the end
effector - F
101
Static Analysis Protocol - Free Body
Diagram 1/
Step 1
Lock all the joints - Converting the
manipulator (mechanism) to a
structure
Step 2
Consider each link in the structure as
a free body and write the force /
moment equilibrium equations
Step 3
Solve the equations - 6 Eq. for each link.
Apply backward solution starting from
the last link (end effector) and end up at
the first link (base)
102
Diagram 2/
• Special Symbols are defined
for the force and torque
exerted by the neighbor link
- Force exerted on link i by link i-1

- Torque exerted on link i by link i-1
Reference Force f or
coordinate torque n
system {B}
Exerted on link A by link A-1
• For easy solution superscript index

(B) should the same as the subscript (A)
103
Diagram 3/
• For serial manipulator in static equilibrium (joints locked), the

sum the forces and torques acting on link i in the link frame {i}
are equal to zero.
104
Diagram 4/
• Procedural Note: The solution starts at the end effector

and ends at the base
• Re-writing these equations in order such that the known

forces (or torques) are on the right-hand side and the
unknown forces (or torques) are on the left, we find
105
Diagram 5/
• Changing the reference frame such that each force (and torque)
is expressed upon their link’s frame, we find the static force
(and torque) propagation from link i+1 to link i
• These equations provide the static force (and torque)

propagation from link to link. They allow us to start with the
force and torque applied at the end effector, and calculate
the force and torque at each joint all the way back to the
robot base frame.
106
Diagram 6/
• Question: What torques are needed at the joints in order to
balance the reaction forces and moments acting on the link.
• Answer: All the components of the force and moment vectors

are resisted by the structure of mechanism itself, except for the
torque about the the joint axis.
• Therefore, to find the joint the torque required to maintain the

static equilibrium, the dot product of the joint axis vector with
the moment vector acting on the link is computed
Revolute Joint
Prismatic Joint
107
Example - 2R Robot - Static Analysis
Problem
Given:
- 2R Robot
-A Force vector is applied by
the end effector
Compute:
The required joint torque as a
function of the robot configuration
and the applied force
108
Solution
• Lock the revolute joints
• Apply the static equilibrium equations starting from the

end effector and going toward the base
109
• For i=2
Vector cross
product
110
• For i=1
111
112
• Re-writing the equations in a matrix form
113
Jacobians in the force domain
• We have found joint torques that will exactly balance

forces at the end-effector in the static situation.
• When force act on a mechanism, work (in the technical

sense) is done if the mechanism moves through a
displacement.
• Work is defined as a force acting through a distance and

is a scalar with units of energy.
• The principle of virtual work allows us to make certain

statement about the static case by allowing the amount of
this displacement to go to an infinitesimal.
Note : The units of energy are invariant after coordinate

changing.
114
The principle of virtual work
• In the multidimensional case, work is the dot product of
a vector force or torque and a vector displacement:
F is a 6×1 Cartesian force-torque vector acting at the end-effctor,

įX is a 6×1 infinitesimal Cartesian displacement of the end-effector,
is a 6×1 vector of torques at the joints of the robot,
įș is a 6×1 vector of infinitesimal joint displacements.
• Inner product can also be expressed as:
• where
•it must hold for all įș , so we have
115
Jacobians: Velocities and static forces
• For velocity:
• For force-torque:
• If the Jacobian is with respect to frame {0}, we have
• When the Jacobian loses full rank, there are certain

directions in which the end-effector cannot exert static
forces as desired.
116
Jacobians in the force domain
• If the Jacobian is singular, F could be increased or
decreased in certain directions (those defining the
null-space of the Jacobian) with no effect on the
value calculated for .
• This also means that near singular configurations, mechanical

advatage tends towards infinity such that with small joint
torques large forces could be generated at the end-effector.
117
Force Mapping and Singularities
• The relationship between joint torque and end effector force
and moments is given by:
• The rank of is equals the rank of .
• At a singular configuration there exists a non trivial force

such that
• In other words, a finite force can be applied to the end effector

that produces no torque at the robot’s joints. In the singular
configuration, the manipulator can “lock up.”
118
Force Mapping and Singularities
• This situation is an old and famous one in mechanical
engineering.
• For example, in the steam locomotive, “top dead center”

refers to the following condition
• The piston force, F, cannot generate any torque around the drive
wheel axis because the linkage is singular in the position shown.
119
Cartesian transformation of velocities
and static forces
120
and static forces
If the frames {A} and {B} are rigidly connected, the matrix
operator form to transform general velocity vectors in frame
{A} to their description in frame {B} is given as follows:
• It can be shown by substituting into equations (5.45)

and (5.47) and re-expressed the two equations in matrix form.
?
ª vi º ª i R i R Pi 1 uº ª vi º
i 1 i 1 i 1 i i
« i 1 » « »« i »
¬ Z i¼ ¬ 0 i 1
i R ¼ ¬ Zi ¼
121
and static forces
By defining a 6 ×6 operator: a velocity transformation,
which maps velocities in {A} into velcoties in {B}, we have
• The inverse transformation which maps velocities in {B} to

{A} is given in the similar matrix form as
• or
i 1 i 1 i
i
vi i
R i 1vi 1 i Zi ui Pi 1
i R ( vi Z i u Pi 1 )
i i i 1
vi 1 i
Zi i
R i 1Z i 1
•Note that i 1
Zi 1
i 1
R iZi
i 1
i 1
i
R i 1vi 1 iPi 1ui Zi
i
122
and static forces
• Similarly, with the static forces relations
ªi fi º ª i1iR 0 ºªi1 fi1º

«i » «i i »«i1 »
n P u i
¬ i ¼ ¬ i1 i1 i1R¼¬ ni1¼
R
• We have the
•where is used to denote a force-momonet transformation.
123
and static forces
Velocity and force transformations are similar to Jacobians
in that they related velcotities and force in different
coordinate systems. Similarly to Jacobians we have that
• as can be verified by examining by the following equations
124
Example
• Given:
-the output of the force
Sensor.
-the transformation
related the tool frame to
the sensor frame.
• Find:
• Solution:
where
125

MEV442 Module 2

Caricato da

Informazioni sul documento

Descrizione originale:

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

MEV442 Module 2

Caricato da

Copyright:

Formati disponibili

MEV442

Dr. Santhakumar Mohan

• We examine the notions of linear and angular

• We also will consider force acting on a rigid

• It turns out that the study of both velocity and

• The velocity of a point Q relative to frame {A} is

• The angular velocity vector is convenient to use because it

– Represented Frame (Reference Frame) :

• A velocity vector computed in frame {B} and represented in

• Problem: describe the motion of

• Solution: Frame {B} is located

• Problem: describe the velocity

• Solution: Even though the vector

• The Magnitude of the

• Using a vector cross product

• Using the rotation matrix to remove the dual-superscript,

angular velocity matrix angular velocity vector

• Matrix Form • Vector Form

Rotating about the

ª kxkxvT cT kxkyvT kzsT kxkzvT  kysT º

ª 0 kzT kyT º ª 0 kzT kyT º

• These descriptions may be combined together to describe

• A similar description of the linear and angular velocities

• A velocity vector computed in frame {B} and represented in

• The velocities differentiate (computed) relative to the base

• Problem: Calculate the linear

• Solution (Concept): Due to

The velocity of link i+1 will be

• we can re-assign link names to calculate the velocity of an

• By pre-multiplying both sides of the equation by ,we can

• Using the recently defined notation, we have

-Angular velocity of frame {i+1} measured relative to the

-Angular velocity of frame {i} measured relative to the robot

-Angular velocity of frame {i+1} measured relative to frame

• Angular velocity of frame {i} measured relative to the robot

In the second term

• Angular velocity of frame {i+1} measured (differentiate) in

• Assuming that a joint has only 1 DOF. The joint configuration

Since the term depends on all previous links through this

• The derivative of a vector in a moving frame (linear and rotation

• we re-assign link frames for adjacent links (i and i +1) with

• By pre-multiplying both sides of the equation by , we can

-Linear velocity of frame {i+1} measured relative to

• Assuming that a joint has only 1 DOF. The joint configuration

• Since the term depends on all previous links through this

• In the car example: Observer sitting in the “Car”

Observer sitting in the “World”

Tip Location in Tip Location in

Tip Velocity in Tip velocity in

• Suppose that for example we have 6 functions, each of which is

• We may also use a vector notation to write these equations as

• Which again might be written more simply using a vector

• By dividing both sides by the differential time element, we

• Note that the Jacobian is time varying linear transformation

This relationship is given by:

• According to definition of the Jacobian is given in the Eqs.

• The kinematic equations of a robot can be obtained via the

• However, while this is straightforward for linear velocity, there

• For revolute joint:

• For prismatic joint:

x L1 cosT1  L2 cos(T1  T2 )  L3 cos(T1  T2  T3 )

• Differentiating the three expressions gives

ª kxkxvT cT kxkyvT kzsT kxkzvT kysT º

ª 0 kzT kyT º ª 0 kzT kyT º

x L1 cosT1 L2 cos(T1 T2 ) L3 cos(T1 T2 T3 )

ªi fi º ª i1iR 0 ºªi1 fi1º