0% found this document useful (0 votes)
20 views39 pages

Chương 07

The document discusses Jacobians and their use in relating the velocities of links in a robotic manipulator to joint velocities and forces. It covers linear and angular velocities of rigid bodies, velocity propagation between links, and how Jacobians can relate joint and Cartesian space velocities and forces.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
20 views39 pages

Chương 07

The document discusses Jacobians and their use in relating the velocities of links in a robotic manipulator to joint velocities and forces. It covers linear and angular velocities of rigid bodies, velocity propagation between links, and how Jacobians can relate joint and Cartesian space velocities and forces.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 39

Ho Chi Minh city University of Technology and Education

Faculty of Electrical & Electronics Engineering


Robotics and Intelligent Control Laboratory

Topic 7: Jacobians: Velocities and


Static forces

Presenter: Dr. Duc Thien, Tran

1
Contents

Topic 1: Introduction
Topic 2: Basic robotic concepts
Topic 3: Spatial Representations of Rigid Bodies
Topic 4: Forward Kinematics of Robot Manipulators
Topic 5: Inverse Kinematics of Robot Manipulators
Topic 6: Kinematic Trajectory Generation.
Topic 7: Jacobians: Velocities and Static forces

2
References

1. John, J Craig, “Introduction to Robotics: Mechanics and Control”

3
Outline:

7.1 INTRODUCTION
7.2 NOTATION FOR TIME-VARYING POSITION AND ORIENTATION
7.3 LINEAR AND ROTATIONAL VELOCITY OF RIGID BODIES
7.4 MOTION OF THE LINKS OF A ROBOT
7.5 VELOCITY "PROPAGATION" FROM LINK TO LINK
7.6 JACOBIANS
7.7 SINGULARITIES
7.8 STATIC FORCES IN MANIPULATORS
7.9 JACOBIANS IN THE FORCE DOMAIN

4
Introduction

1. In this Chapter the following concepts will be studied:


The linear and angular velocity of a rigid body to analyze the motion of the
manipulator
Forces that act on a rigid body (application on static forces of a manipulator)
➔ Jacobian matrix relations

5
Introduction
1. Velocity analysis between Cartesian Space and Joint Space
1  1 
   
v  2   2  = J −1   v 
  =  J        
   
   
 n   n 
Jacobian Matrix If J is singular
Transformation/Mapping ➔ it is not invertible
➔ Singular point/configuration
(Important issue in robot design When θ = 0/180 ➔
2
that should be avoided) singular configuration

The Jacobian is important to determine the torque in the joints needed to be


applied to give a specific contact force at the End-Effector (Static problem)

6
Notations

1. Differentiation of a position vector (Linear Velocity)


𝐵 𝐵
𝐵𝑉 =
𝑑 𝐵 𝑄 = lim
𝑄 𝑡 + ∆𝑡 − 𝑄(𝑡)
𝑄
𝑑𝑡 ∆𝑡→0 ∆𝑡
𝐵𝑉 ≡ the velocity of a position vector
𝑄
≡ Linear velocity of a point in space represented by the position vector.
Derivative is made relative to frame {B} (Frame of Differentiation)
Also the velocity vector is expressed in the same frame {B}
To express the velocity in any other frame {A}
𝐴𝑑
𝐴 𝐵𝑉 = 𝐵𝑄 = 𝐴𝑅 𝐵𝑉
𝑄 𝐵 𝑄
𝑑𝑡
𝐴 𝐵𝑉 ≡ The velocity of point (position vector Q) relative the frame {B} expressed in frame {A}.
𝑄
(draw frames for explanations)

7
Notations

• Differentiation of a position vector (Linear Velocity)


✓ Often used the velocity of the origin of a frame {C} (for example) relative to the
universe frame {U}.
✓ 𝑣Ԧ𝑐 = 𝑢 𝑉𝑂𝐶 OC is origin of frame {C}
✓ Remember that in 𝑢 𝑉𝑂𝐶 the differentiation is made relative to {U} and
expressed in {U}
✓ 𝐴 𝑣Ԧ𝐶 as v is small-letter the differentiation is made relative/ respect to {U}, but
the velocity is expressed in {A}.
• Example:

8
Notations

1. Example 5.1:

9
Notations

1. The Angular Velocity Vector


Ω is angular velocity vector of a body ( 𝐴 𝑣Ԧ𝐶 is for a point)
As a frame represents the orientation
of the body ≡ rotational velocity of the frame.
𝐴 Ω is rotational velocity of frame {B} relative
𝐵
to frame {A}:
- it’s direction represents the instantaneous
axis of rotation of {B} relative to {A}
- it’s magnitude represents the speed of rotation.
𝐶 𝐴
Ω𝐵 the angular velocity of {B} relative to {A} expressed in {C}
𝑈
In the universe frame {U} Ω𝐶 = 𝑤𝑐

10
Linear and rotational velocities of rigid
bodies

1. Linear velocity: (Translation only)


𝐴𝑄 𝐴𝑃 𝐴 𝐵
= 𝐴,𝐵 + 𝐵𝑅 𝑄
Translation only ➔ the orientation of {B} relative to {A} is not changing
𝐴𝑉 = 𝐴 𝑉 𝐴 𝐵
𝑄 𝑂𝐵 + 𝐵 𝑅 𝑉𝑄

B
Q
A
Q

A
PA,B

11
Linear and rotational velocities of rigid
bodies

2. Rotational velocity: (Rotation only)


Two frames {A} and {B} have the same origin for all the time only the relative
orientation is changing in time ➔ 𝐴 Ω𝐵
𝐵 𝑄 is fixed in {B} ➔ 𝐵 𝑉 = 0
𝑄

What is the velocity of point Q in {A} 𝐴𝑉 ?


𝑄
A
VQ = A  B  AQ
= A B  ( A
B R BQ )
( ) (V)
If 𝐵𝑄 is changing in {B} A
A
VQ =  B A A
B R Q +
B B
Q

= A B ( A
B RB Q) + R V
A
B
B
Q

12
Linear and rotational velocities of rigid
bodies

3. Linear and rotational velocity at the same time


𝐴
The same as point 2, however frame {B} is moving relative to frame {A}, i.e. 𝑉𝑂𝐵
A
VQ = AVOB + A  B  ( A
B )
R BQ + BA R BVQ

13
Motion of the links of a robot (notations)

𝑖𝑣
≡ linear velocity of the origin of frame {i} relative to the reference frame {O} (fixed),
Ԧ𝑖
expressed in {i}
𝑖𝑤 ≡ angular velocity of frame {i} relative to the reference frame {O} (fixed), expressed
𝑖
in {i}

i
vi

Oi i
i

14
Velocity propagation from link to link
Starting from the base the velocity of any link
(i+1) equal to the previous link (i) + the relative
Velocity between (i+1) and (i)
• Angular velocity propagation
𝑖𝑤 𝑖 𝑖
𝑖+1 = 𝑤𝑖 + Ω𝑖+1
𝑖Ω 𝑖 𝑖+1
𝑖+1 = 𝑖+1𝑅 𝑧Ԧ𝑖+1 𝜃ሶ𝑖+1
𝑖𝑤 𝑖
𝑖+1 = 𝑤𝑖 + 𝑖+1𝑅
𝑖 𝑖+1
𝑧Ԧ𝑖+1 𝜃ሶ𝑖+1 i
i +1
i +1
vi
𝑖+1 𝑤
𝑖+1 =
𝑖+1 𝑖
𝑖 𝑅 𝑤𝑖 +
𝑖+1 𝑧Ԧ ሶ
𝑖+1 𝜃𝑖+1
Where 𝑖+1 𝑧Ԧ𝑖+1 = 0,0,1 𝑇

i
i

15
Velocity propagation from link to link

• Linear velocity propagation


𝐴 𝐵
𝐴𝑉
𝑄 = 𝐴 𝑉OB + 𝐴 Ω𝐵 × 𝐵𝑅 𝑄 + 𝐵𝐴𝑅𝐵 𝑉𝑄
A ≡ 0; Q ≡ 𝑃𝑖+1 ; B ≡ i;
0𝑉 0 𝑖
𝑖+1 = 0 𝑉𝑖 + 0 Ω𝑖 × 𝑖 𝑅 𝑃𝑖+1
0 𝑖
𝑣Ԧ𝑖+1 = 𝑣Ԧ𝑖 + 𝑤i × 𝑖 𝑅 𝑃𝑖+1
𝑖
𝑣Ԧ𝑖+1 = 𝑖 𝑣Ԧ𝑖 + 𝑖 𝑤i × 𝑖 𝑃𝑖+1 i
i +1
i +1
vi
𝑖+1
𝑣Ԧ𝑖+1 = 𝑖+1𝑖𝑅 𝑖 𝑣Ԧ𝑖 + 𝑖 𝑤i × 𝑖 𝑃𝑖+1

i
i

16
Velocity propagation from link to link

Prismatic joint
𝑖
𝑤𝑖+1 = 𝑖 𝑤𝑖
𝑖+1 𝑖+1 𝑖
𝑤𝑖+1 = 𝑖 𝑅 𝑤𝑖
𝐴 𝐵
𝐴
𝑉𝑄 = 𝐴 𝑉OB + 𝐴 Ω𝐵 × 𝐵𝑅 𝑄 + 𝐵𝐴𝑅𝐵 𝑉𝑄
A ≡ 0; Q ≡ 𝑃𝑖+1 ; B ≡ i;
0𝑉 0 𝑖
𝑖+1 = 0 𝑉𝑖 + 0 Ω𝑖 × 𝑖 𝑅 𝑃𝑖+1
𝑣Ԧ𝑖+1 = 𝑣Ԧ𝑖 + 𝑤i × 0 𝑖 0 𝑖+1
𝑖 𝑅 𝑃𝑖+1 + 𝑖+1𝑅 𝑧Ԧ𝑖+1 𝑑ሶ 𝑖+1
𝑖𝑣Ԧ𝑖+1 = 𝑖 𝑣Ԧ𝑖 + 𝑖 𝑤i × 𝑖 𝑃𝑖+1 + 𝑖+1𝑖𝑅 𝑖+1 𝑧Ԧ𝑖+1 𝑑ሶ 𝑖+1
Ԧ𝑖+1 = 𝑖+1𝑖𝑅 𝑖 𝑣Ԧ𝑖 + 𝑖 𝑤i × 𝑖 𝑃𝑖+1 + 0 0 𝑑ሶ 𝑖+1 𝑇
𝑖+1 𝑣

17
Example

1. A two-link manipulator with rotational joints is shown in Fig. 5.8. Calculate the velocity of
the tip of the arm as a function of joint rates. Give the answer in two forms—in terms of
frame {3} and also in terms of frame {O}.

18
Example

1. The transformation matrices:


𝑐1 −𝑠1 0 0
0 𝑠1 𝑐1 0 0
1 𝑇 =
0 0 1 0
0 0 0 1
𝑐2 −𝑠2 0 𝑙1
1 𝑠2 𝑐2 0 0
2𝑇 = 0 0 1 0
0 0 0 1
1 0 0 𝑙2
2 0 1 0 0
3 𝑇 =
0 0 1 0
0 0 0 1

19
Example

1. Velocities:
0 0
1𝑤 = 0 1𝑣 =
1 1 0
𝜃ሶ1 0
0 𝑐2 𝑠2 0 0 𝑙1 𝑠2 𝜃ሶ1
2𝑤 = 2𝑣 =
2 0 2 −𝑠2 𝑐2 0 𝑙1 𝜃ሶ1 = 𝑙1 𝑐2 𝜃ሶ1
𝜃ሶ1 + 𝜃ሶ2 0 0 1 0 0
𝑙1 𝑠2 𝜃ሶ1
3
𝑤3 = 2 𝑤2 3
𝑣3 = 𝑙1 𝑐2 𝜃ሶ1 + 𝑙2 𝜃ሶ1 + 𝜃ሶ 2
0

20
Example

1. Rotation matrix
𝑐12 −𝑠12 0
0 0 1 2
3 𝑅 = 1 𝑅 2 𝑅 3 𝑅 = 𝑠12 𝑐12 0
0 0 1
The velocity:
−𝑙1 𝑠1 𝜃ሶ1 − 𝑙2 𝑠12 𝜃ሶ1 + 𝜃ሶ 2
0𝑣 = 0𝑅 3𝑣 =
3 3 3 −𝑙1 𝑐1 𝜃ሶ1 − 𝑙2 𝑐12 𝜃ሶ1 + 𝜃ሶ 2
0

21
Jacobians

If we have the function:


𝑦1 = 𝑓1 𝑥1 , 𝑥2 , … , 𝑥6
𝑦2 = 𝑓2 𝑥1 , 𝑥2 , … , 𝑥6

𝑦6 = 𝑓6 𝑥1 , 𝑥2 , … , 𝑥6
Or, in vector form 𝑌 = 𝐹(𝑋) with 𝑌 = 𝑦1 , 𝑦2 , … , 𝑦6 𝑇 , 𝐹 𝑋 = 𝑓1 𝑋 , 𝑓2 𝑋 , … , 𝑓6 𝑋 𝑇, and 𝑋 =
𝑥1 , 𝑥2 , … , 𝑥6 𝑇 .
The differentiation of yi can be considered as follows:
𝛿𝑓1 𝛿𝑓1 𝛿𝑓1
𝛿𝑦1 = 𝛿𝑥 + 𝛿𝑥 + ⋯ + 𝛿𝑥
𝛿𝑥1 1 𝛿𝑥2 2 𝛿𝑥6 6
𝛿𝑓2 𝛿𝑓2 𝛿𝑓2
𝛿𝑦2 = 𝛿𝑥 + 𝛿𝑥 + ⋯ + 𝛿𝑥
𝛿𝑥1 1 𝛿𝑥2 2 𝛿𝑥6 6

𝛿𝑓6 𝛿𝑓6 𝛿𝑓6
𝛿𝑦6 = 𝛿𝑥 + 𝛿𝑥 + ⋯ + 𝛿𝑥
𝛿𝑥1 1 𝛿𝑥2 2 𝛿𝑥6 6

22
Jacobians

𝛿𝐹
Or, simply, in matrix form 𝛿𝑌 = 𝛿𝑋
𝛿𝑋
𝛿𝐹
Remark: for nonlinear functions f1, f2, …, fn of X the partial derivative is also a function of X
𝛿𝑋
𝜕𝐹
𝐽 𝑋 =
𝜕𝑋
𝑑𝑌 = 𝐽 𝑋 𝑑𝑋
𝑑𝑌 𝑑𝑋
=𝐽 𝑋
𝑑𝑡 𝑑𝑡
𝑌ሶ = 𝐽 𝑋 𝑋ሶ
➔ Jacobian maps/transform velocities from X to Y
As X=X(t) (time dependent) ➔J=J(X) is also time dependent.

23
Jacobians

In robotics, the Jacobian relates the Cartesian velocities with joint velocities
0
𝑣 = 0 𝐽 Θ Θሶ
Θ ≡ vector of joint angles
Θሶ ≡ vector of joint velocities
General Case:
0𝜐
6×1 =
0𝐽(𝛩) 𝛩ሶ
6×6 6×1
0𝑣
Ԧ
0𝜐
6×1 = 0
𝜔
𝐽11 𝐽12 ⋯ 𝐽1𝑛
0𝐽 𝛩 = 𝐽21 𝐽22 ⋯ 𝐽2𝑛
⋮ ⋮ ⋱ ⋮
𝐽𝑚1 𝐽𝑚2 ⋯ 𝐽𝑚𝑛

24
Example: 3DOF Manipulator

25
Singularities

1. Given 0𝜐, find 𝛩ሶ


𝛩ሶ = 0𝐽 −1 𝛩 0𝜐
0𝐽 is invertible if det 0𝐽 𝛩 ≠0
For some values of Θ, 0𝐽 could be non-invertible
➔ singularities of the robot
Singularity:
Boundary singularity:
Interior singularity

26
Singularities

Boundary singularity:
All robots have singularities at the boundary or their workspace.

27
Singularities

Interior singularity
Inside the workspace, away from the WS boundary.
Generally, are caused by a lining up of two or more joint axes.

When a manipulator is in a singular


configuration, it has lost one or more
degrees of freedom!

28
Singularities

Example: Find an example about the singularities in robot

29
STATIC FORCES IN MANIPULATORS

Discuss the concept of force and moment propagation


Starting from the last link to the base Force (equilibrium)
Use the notation:
𝑖Ԧ
f𝑖 ≡ force exerted on link i by link i-1
𝑖+1 Ԧf = −𝑖 Ԧf
𝑖 𝑖+1 regardless to the frame of notation
all vectors must be expressed in the same frame.
σ 𝐹Ԧ = 0
➔ 𝑖 Ԧf𝑖 − 𝑖 Ԧf𝑖+1 = 0 ➔ 𝑖Ԧ
f𝑖 = 𝑖 Ԧf𝑖+1
Different notation
is used here

30
STATIC FORCES IN MANIPULATORS

Moment equation (equilibrium)


Use the notation:
𝑖 n ≡ Moment exerted on link i by link i-1
𝑖
𝑖+1
n𝑖 = −𝑖 n𝑖+1
σ 𝑀𝑂𝑖 = 0
➔ 𝑖 n𝑖 −𝑖 n𝑖+1 −1 P𝑖+1 ×𝑖 Ԧf𝑖+1 = 0
➔ 𝑖 n𝑖 = 𝑖 n𝑖+1 +𝑖 P𝑖+1 × 𝑖 Ԧf 𝑖+1

Different notation
is used here

31
STATIC FORCES IN MANIPULATORS

The important result for static force “propagation” from link to link
𝑖
𝑓𝑖 = 𝑖+1𝑖𝑅 𝑖+1 𝑓𝑖+1
𝑖𝑛 = 𝑖 𝑖+1
𝑖 𝑖+1𝑅 𝑛𝑖+1 +𝑖 𝑃𝑖+1 ×𝑖 𝑓𝑖
The joint 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:
𝜏𝑖 = 𝑖 𝑛𝑖𝑇 𝑖 𝑧Ƹ𝑖
In the case that joint i is prismatic, we compute the joint actuator force as:
𝜏𝑖 = 𝑖 𝑓𝑖𝑇 𝑖 𝑧Ƹ𝑖

32
EXAMPLE 5.7

Starting from the last link and going toward the base of
robot:
𝑓𝑥 𝑓𝑥 0
2
𝑓2 =3 𝐹 = 𝑓𝑦 , 2 𝑛2 = 𝑙2 𝑥ො2 × 𝑓𝑦 = 0
0 0 𝑙2 𝑓𝑦

𝑐2 −𝑠2 0 𝑓𝑥 𝑐2 𝑓𝑥 − 𝑠2 𝑓𝑦
1
𝑓1 = 12𝑅 2 𝑓2 = 𝑠2 𝑐2 0 𝑓𝑦 = 𝑠2 𝑓𝑥 + 𝑐2 𝑓𝑦
0 0 1 0 0
0
1
𝑛1 = 12𝑅 2 𝑛2 + 𝑙1 𝑥ො1 ×1 𝑓1 = 0
𝑙1 𝑠2 𝑓𝑥 + 𝑙2 + 𝑙1 𝑐2 𝑓𝑦
Therefore:
𝜏1 =1 𝑛1𝑇 1 𝑧Ƹ 1 = 𝑙1 𝑠2 𝑓𝑥 + 𝑙2 + 𝑙1 𝑐2 𝑓𝑦
𝜏2 =2 𝑛2𝑇 2 𝑧2Ƹ
33
JACOBIANS IN THE FORCE DOMAIN

We have
𝐹 𝑇 𝛿𝑋 = 𝜏 𝑇 𝛿Θ
where 𝐹 is a 6 x 1 Cartesian force-moment vector acting at the end-effector, 𝛿𝑋 is a 6 x 1
infinitesimal Cartesian displacement of the end-effector, 𝜏 is a 6 x 1 vector of torques at the joints,
and 𝛿Θ is a 6 x 1 vector of infinitesimal joint displacements.
The definition of the Jacobian is
𝛿𝑋 = 𝐽𝛿Θ
As a result,
𝐹𝑇 𝐽 = 𝜏 𝑇
Otherwise,
𝜏 = 𝐽𝑇 𝐹

34
Cartesian Transformation of Velocities and
Static Forces

The representation of general velocity of a body:


𝑣
𝜐=
𝑤
The representation of general force vectors
𝐹
ℱ=
𝑁
where F is a 3 x 1 force vector and N is a 3 x 1 moment vector.
The two frames involved here are rigidly conntected,
𝐵 𝐵 𝐵 𝐴 𝐴
𝑣𝐵 𝐴𝑅 − 𝐴𝑅 𝑃𝐵𝑂𝑅𝐺 × 𝑣𝐴
𝐵𝑤 = 𝐵 𝐴𝑤
𝐵 0 𝐴𝑅 𝐴

Where the cross product is expressed as follows:


0 −𝑝𝑧 𝑃𝑦
𝑃 ×= 𝑃𝑧 0 −𝑃𝑥
−𝑃𝑦 𝑃𝑥 0

35
Cartesian Transformation of Velocities and
Static Forces
A velocity transformation, 𝑇𝑣 :
𝐵𝑣
𝐵 = 𝐵𝐴𝑇𝑣 𝐴 𝑣𝐴
The description of velocity in terms of {A} given the quantities in {B}:
𝐴𝑣 𝐴 𝐴𝑃 𝐴 𝐵𝑣
𝐴 𝐵𝑅 𝐵𝑂𝑅𝐺 × 𝐵𝑅 𝐵
𝐴𝑤 = 𝐴 𝐵𝑤
𝐴 0 𝐵𝑅 𝐵
Or
𝐴𝑣
𝐴 =𝐴 𝐵
𝐵𝑇𝑣 𝑣𝐵

General force transformation, 𝑇𝑓 :


𝐴𝐹 𝐴 𝐵𝐹
𝐴 𝐵𝑅 0 𝐵
𝐴𝑁 = 𝐵𝑁
𝐴
𝐴𝑃
𝐵𝑂𝑅𝐺 × 𝐵𝐴𝑅 𝐴
𝐵𝑅 𝐵
Or
𝐴ℱ
𝐴 =𝐴 𝐵
𝐵𝑇𝑓 ℱ𝐵

Velocity and force transformations are similar to Jacobian


𝐴 𝑇 = 𝐴𝑇 𝑇
𝐵 𝑓 𝐵 𝑣

36
Example

37
Example

The kinematics of a 3-DOF robot is given by


𝑐1 𝑐23 −𝑐1 𝑠23 𝑠1 𝑙1 𝑐1 + 𝑙2 𝑐1 𝑐2
0 𝑠1 𝑐23 −𝑠1 𝑠23 −𝑐1 𝑙1 𝑠1 + 𝑙2 𝑠1 𝑐2
3𝑇 = 𝑠23 𝑐23 0 𝑙2 𝑠2
0 0 0 1
Find 0 𝐽 Θ .

38
Thank you for your listening

39

You might also like