0% found this document useful (0 votes)
82 views50 pages

Week 6 Lecture Material - Watermark PDF

The document discusses control schemes and sensors for robotics. It describes how DC motors connected at each robot joint can control torque proportional to armature current. It also discusses partitioned control schemes using proportional-derivative and proportional-integral-derivative control laws to control joint torque based on error between desired and actual positions. Finally, it categorizes sensors as internal or external and describes various sensor types used for position, force, and tactile feedback including potentiometers, encoders, and force/moment sensors.

Uploaded by

ds vidhya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
82 views50 pages

Week 6 Lecture Material - Watermark PDF

The document discusses control schemes and sensors for robotics. It describes how DC motors connected at each robot joint can control torque proportional to armature current. It also discusses partitioned control schemes using proportional-derivative and proportional-integral-derivative control laws to control joint torque based on error between desired and actual positions. Finally, it categorizes sensors as internal or external and describes various sensor types used for position, force, and tactile feedback including potentiometers, encoders, and force/moment sensors.

Uploaded by

ds vidhya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 50

WEEK 6: ROBOTICS

PROF. (DR.) DILIP KUMAR PRATIHAR


MECHANICAL ENGINEERING DEPARTMENT, IIT KHARAGPUR

1
Topic 5: Control Scheme

PROF. (DR.) DILIP KUMAR PRATIHAR


MECHANICAL ENGINEERING DEPARTMENT, IIT KHARAGPUR

2
Control of Motor

• A DC motor is connected at each joint of a robot, where


torque is proportional of the armature current.

𝝉𝝉𝒎𝒎 ∝ 𝑰𝑰𝑨𝑨
𝝉𝝉 = 𝑲𝑲𝒎𝒎 ⋅ 𝑰𝑰𝑨𝑨

3
• Joint torque 𝝉𝝉 can be represented as follows:

𝝉𝝉 = 𝑫𝑫(𝜽𝜽)𝜽𝜽 𝜽𝜽
+ 𝒉𝒉(𝜽𝜽, 𝜽̇
) + 𝑪𝑪

where
𝑫𝑫(𝜽𝜽) : inertia terms
𝒉𝒉(𝜽𝜽, 𝜽̇
): Coriolis and centrifugal terms
𝑪𝑪𝜽𝜽: gravity terms

4
If we consider friction, then

𝝉𝝉 = 𝑫𝑫(𝜽𝜽)𝜽𝜽 𝜽𝜽, 𝜽
𝜽̇
+ 𝒉𝒉(𝜽𝜽, 𝜽̇
) + 𝑪𝑪 𝜽𝜽 + 𝑭𝑭

where
𝑭𝑭𝜽𝜽, 𝜽
𝜽:̇
friction terms
5
Let us consider Partitioned Control Scheme

𝝉𝝉 = 𝜶𝜶𝝉𝝉′ + 𝜷𝜷

where 𝜶𝜶 = 𝑫𝑫 𝜽𝜽
𝜷𝜷 = 𝒉𝒉(𝜽𝜽, 𝜽̇
) + 𝑪𝑪 𝜽𝜽 + 𝑭𝑭𝜽𝜽, 𝜽
𝜽̇

6
Now, 𝝉𝝉′ can be written as follows:

𝝉𝝉′ = 𝜽̈+ 𝑲𝑲 𝑬𝑬 + 𝑲𝑲 𝑬𝑬 (for PD control law)


𝒅𝒅 𝑷𝑷 𝑫𝑫

𝝉𝝉′ =
𝜽̈
𝒅𝒅 + 𝑲𝑲𝑷𝑷𝑬𝑬 + 𝑲𝑲𝑰𝑰 ∫ 𝑬𝑬𝒅𝒅𝑬𝑬 + 𝑲𝑲𝑫𝑫𝑬𝑬 (for PID control law)

where E = error = 𝜽𝜽d − 𝜽𝜽


where 𝜽𝜽d : Desired value of 𝜽𝜽
𝜽𝜽 : Actually obtained value of 𝜽𝜽
7
Control Architecture

D8) Load

h(8,8)+ C (8)+ F (8,8 )

iiT r:n Ax ñ(:.r'cr‹


8
PUMA

Current Amplifier Motor


6503p D/A

Encoder

Ma ster
Control I I terfa ce
computer

iiT r:n Ax ñ(:.r'cr‹


9
Topic 6: Sensors

PROF. (DR.) DILIP KUMAR PRATIHAR


MECHANICAL ENGINEERING DEPARTMENT, IIT KHARAGPUR

10
SENSORS

 Human-beings collect information of the surroundings


using their sensors, namely eyes, ears, nose, skin etc.,
in order to perform various tasks.
 A sensor is used to take measurement of
physical variable.
 A sensor requires calibration
 Sensors are used to build intelligent robots

11
Classification of Sensors
1. SENSORS

Internal Sensors External Sensors


(used to operate the drive (used to collect information
units) of the environment)
Ex. Position sensors,
velocity sensors, Ex. Temperature sensors,
Acceleration sensors, Visual sensor, Proximity
Force/Moment sensor sensor, Acoustic sensor

12
Classification of Sensors (Cont.)
SENSORS
2.
Non-Contact Sensors
Contact Sensors
(No physical contact)
(Physical contact between sensor Proximity sensor
mounted on robot and object)
Visual sensor
Touch sensor/ Tactile Force sensor/ Analog Acoustic sensor
sensor/ Binary sensor sensor (not only the Range sensor
(indicates presence or contact is made but also
absence of an object) the force is measured)
Ex. Micro-switch, Limit Ex. Sensors using strain
switch gauges

13
Characteristics of Sensors
 Range : Difference between the maximum and minimum values
of the input that can be measured.
 Response : should be capable of responding to the changes in
minimum time.
 Accuracy : deviation from exact quantity
 Sensitivity = change in output/ change in input
 Linearity : constant sensitivity
 Repeatability : Deviation from reading to reading, when these
are taken for a number of times under identical conditions.
 Resolution

14
Touch Sensor

 Used to indicate whether contact has been made between


two objects

 Does not determine the magnitude of contact force

 Ex. : Micro-switch, Limit switch

15
Connected to robot’s wrist

Finger

Micro – switches

Figure: Micro – switches placed on


two fingers of a robotic hand

16
Position sensor
Linear Potentiometer
1. Potentiometer
Angular Potentiometer
Angular
Potentiometer

Θ: Angular displacement of the wiper


with respect to the reference
𝑹𝑹: Total resistance
𝒓𝒓 ∶ Resistance of the coil between
the wiper and the reference
𝑽𝑽𝒊𝒊𝒊𝒊 : Input voltage
𝑽𝑽𝒐𝒐𝒐𝒐𝑬𝑬 : Output voltage
17
Angular Potentiometer (contd.)
𝑽𝑽𝒐𝒐𝒐𝒐𝑬𝑬 𝑽𝑽
𝒊𝒊𝒊𝒊
𝒓𝒓 = 𝑹𝑹
⇒ 𝑽𝑽 𝒓𝒓
𝒐𝒐𝒐𝒐𝑬𝑬 = 𝑽𝑽𝒊𝒊𝒊𝒊
𝑹𝑹
For the known values of 𝑽𝑽𝒊𝒊𝒊𝒊 ,𝑹𝑹; 𝑽𝑽𝒐𝒐𝒐𝒐𝑬𝑬 = 𝒇𝒇(𝒓𝒓)
By measuring 𝑽𝑽𝒐𝒐𝒐𝒐𝑬𝑬 , 𝒓𝒓 can be determined and
hence, angular displacement 𝜽𝜽.
Demerit
• Resistance of the wire is temperature
dependent. Potentiometer is temperature
sensitive.
18
2. Absolute Optical Encoder
Optical Encoder
Incremental Optical Encoder
Absolute Optical
Encoder
 It is mounted on the shaft
a rotary device
 To generate digital
word identifying actual
position of the shaft
measured from zero
position

19
Absolute Optical Encoder (contd.)
Resolution : 𝟏𝟏 part in 𝟐𝟐𝒊𝒊 , where 𝒊𝒊 : number of concentric rings (tracks)
𝟐𝟐𝟑𝟑 𝟐𝟐𝟐𝟐 𝟐𝟐𝟏𝟏 𝟐𝟐𝟎𝟎 𝑫𝑫𝑽𝑽
𝟎𝟎 𝟎𝟎 𝟎𝟎 𝟎𝟎 𝟎𝟎
𝟎𝟎 𝟎𝟎 𝟎𝟎 𝟏𝟏 𝟏𝟏
𝟎𝟎 𝟎𝟎 𝟏𝟏 𝟎𝟎 𝟐𝟐
∶ ∶
∶ ∶
𝟏𝟏 𝟏𝟏 𝟏𝟏 𝟎𝟎 𝟏𝟏𝟏𝟏
𝟏𝟏 𝟏𝟏 𝟏𝟏 𝟏𝟏 𝟏𝟏𝟏𝟏

20
Incremental optical encoder
 Consists of one coded disc
and two photo-detectors
 By counting the number of light
and dark zones, angular
displacement can be measured
with respect to known starting
position.
 It can determine the direction of
rotation also
 It is construction-wise
simpler, less accurate and
less expensive.

21
Linear Variable Differential
Transformer (LVDT)
 It consists of two parts: fixed
casing and moving magnetic
core
 In-between the fixed casing
and magnetic core, there are
one primary(𝑳𝑳𝑷𝑷) and two
secondary (𝑳𝑳𝒔𝒔𝟏𝟏, 𝑳𝑳𝒔𝒔𝟐𝟐) coils
 Produced voltage output is
proportional to the
displacement of moving part
relative to the fixed one

22
LVDT (contd.)
 AC voltage is applied to 𝑳𝑳𝒑𝒑
 𝑳𝑳𝒔𝒔𝟏𝟏 and 𝑳𝑳𝒔𝒔𝟐𝟐 are connected in
series. V
ou  VLs VLs1
t
2

Calibration curve

LVDT: equivalent electrical circuit


23
Force/Moment sensor
Fx
Mg
W7

RIM
W,
W1
Connected to wrist end)

W3

My My
HUB

Wk
Stra
ca•es
W^

iiT r:n Ax ñ(:.r'cr‹


24
Force/Moment sensor (contd.)
 It is placed between the wrist and end-effector end
 It consists of 4 deflection bars. Two pairs of strain gauges are
mounted on each deflection bar. One end of each deflection
bar is rigidly supported by a hub, which is connected to the end-
effector end. The other ends of the deflection bars are supported
by a common rim, which is connected to the wrist end.
 External forces cause deflection of the mechanical structure,
which are measured using strain gauges.

𝑷𝑷𝑳𝑳𝟑𝟑
𝜹𝜹 =
𝟑𝟑𝑬𝑬𝑰𝑰

25
Force/Moment sensor (contd.)
 Strain gauge is connected to potentiometer circuit, whose
output voltage is proportional to the deflection and hence,
force.
 Three components of force (F) and moment (M) each
are determined by adding and subtracting the
respective components of force. F = CM W
Calibration matrix
Forces/
Readings of the strain
moments
gauges
26

FX   W7C17
W3C13
FY  W1C21  W5C25
FZ  W2C32    W8C38
W4C34 W6C36
MX   W8C48
W4C44 M Y   W C
6 56
W2C52
MZ  W1C61  W3C63   W7C67
W5C65
Precautions
 Strain gauges are to be properly mounted on the deflection bars
 Sensor should be operated within the elastic limit of its
material (deflection bars).

27
Range Sensor
 It measures the distance between the sensor (detector)
mounted on the robot’s body and the object
d
 tan
a
d  a tan
Knowing the values of
Detector Receiver 𝒂𝒂 and 𝜽𝜽, 𝒅𝒅 can be
calculated.
Triangulation method

28
Proximity Sensors

Inductive Sensor Hall-effect Sensor Capacitive Sensor

Inductive Sensor
 It consists of a permanent magnet and a wound coil placed
next to it.

29
 The nature of flux lines changes, as the sensor comes closer
to a ferromagnetic object.
 The flux lines change, as the ferromagnetic object
either enters or leaves the field of the magnet.
 Rate of change of the magnetic flux is proportional to
induced current (voltage)

30
Inductive sensor:
induced voltage vs. time

 Voltage induced across the coil depends on the speed at


which the object either enters or leaves the magnetic
field
 Polarity of the voltage depends on whether the object enters
or leaves the field.
31
Calibration curve

 The figure shows the calibration curve corresponding to the


measured amplitude of voltage signal, the distance between
the sensor and object can be determined.

32
Hall-Effect Sensors (for Ferro-magnetic object)
v
 It works based on the principle of Lorentz force
 If a charge of am ount 𝒒𝒒 is moving with velocity in a magnetic
field of strength B , then the Lo rentz force
  
F  q(v B)

33
Hall-Effect Sensors (cont.)

 Voltage across the


Calibration curve for Hall-effect sensor
semiconductor will
be reduced

34
Capacitive Sensor (suitable for
any material)
 When an object is brought near
to the sensitive electrode, there
will be accumulation of charge
and consequently, its
capacitance changes.
 When the capacitance of the
sensor exceeds a predefined
threshold value, oscillation
starts.
 Oscillations are converted
into output voltage through
PCB.
35

Capacitive sensor(contd.)

Calibration curve
36
Topic 7: Robot Vision

PROF. (DR.) DILIP KUMAR PRATIHAR


MECHANICAL ENGINEERING DEPARTMENT, IIT KHARAGPUR

37
Robot Vision

• Txotraect, characterize and interpret objects present in an


image/photograph captured using a camera.

38
Steps to be Followed

• Step 1: Capturing image of the environment using CCD camera.

• Step 2: Light intensity is measured along a particular direction


say Y using Electron Beam Scanner (in which the charge
accumulated in photo-sites is proportional to light intensity).
Analog plot of light intensity is digitized and it is known as A/D
conversion or digitizing.

39
Robot Vision

(0,0) analog

Light intensity

image elemenl/ picture elemenl/ pixel

N = M = 512,256,128,64,32
IMAGE CAPTURING
SAMPLING (A/D CONVERSION)

iiT r:n Ax ñ(:.r'cr‹


40
Steps to be Followed (contd.)

• Step 3: Image is stored as an array of pixels (each pixel may


have different light intensity values). It is known as frame
grabbing.

• Step 4: Preprocessing of the data collected in Step 3 is done for


noise reduction, restoration of lost information etc.

41
Frame Grabbing

f(0,0) f(0,1)--------------------f(0,M-1)
f(1,0) f(1,1)--------------------f(1,M-1)
. . .
. . .
. . .
. . .
f(N-1,0) f(N-1,1)----------------f(N-1,M-1)

f(x,y) : Light intensity of image at the point (x,y)

42

You might also like