ROBOTICS

AND
AUTOMATION
 Sensors- Need

Sensory–motor
and
process control
systems

Department of Electronics and Instrumentation Engineering,

3
B.M.S.C.E., B'luru
Sensors- Need

Simple instrument model.

Department of Electronics and Instrumentation Engineering,

4
B.M.S.C.E., B'luru
Sensors- Need

Construction structure of a typical instrument

Digital instruments
Department of Electronics and Instrumentation Engineering,
5
B.M.S.C.E., B'luru
Sensors- Need

Department of Electronics and Instrumentation Engineering,

6
B.M.S.C.E., B'luru
 Sensors -Need

Sensors in a data acquisition system.

Sensor 1 - noncontact
Sensors 2,3- passive
Sensor 4 – active
Sensor 5 is internal

Department of Electronics and Instrumentation Engineering,

7
B.M.S.C.E., B'luru
 Sensors - Classification
Definition - Sensor is a device that responds to a changing phenomenon
• Can be categorized into the following areas

• Active and Passive sensors

1.

• Contact and Noncontact sensors

2.

• Absolute or Relative
3.

• Application, Physical phenomena used, Detection method, Sensor specifications

4.

• Internal and External – Robots

5

Department of Electronics and Instrumentation Engineering, 8

B.M.S.C.E., B'luru
Classification of sensors

1.
• Active and Passive sensors

• Passive sensor does not need energy source and directly generates an electric signal in
response to an external stimulus; that is, the input stimulus energy is converted by the
sensor into the output signal.
• Also called self-generating sensors
• Ex: thermocouple, photodiode, piezoelectric sensor.
• Active sensors require external power for their operation, which is called an excitation
signal. That signal is modified by the sensor to produce the output signal.
• Active sensors sometimes are called parametric because their own properties change in
response to an external effect and these properties can be subsequently converted into
electric signals.
Ex: Strain gauges (resistance changes as a function of strain)
Thermistors (resistance changes as a function of temperature)
Capacitive or inductive proximity sensors (capacitance or inductance is a function of
position) Department of Electronics and Instrumentation Engineering,
9
B.M.S.C.E., B'luru
Classification of sensors

Property Sensor Active/Passive Output

Temperature Thermocouple Passive Voltage
Silicon Active Voltage/Current
RTD Active Resistance
Thermistor Active Resistance
Force/Pressure Strain Gage Active Resistance
Piezoelectric Passive Voltage
Acceleration Accelerometer Active Capacitance
Position LVDT Active AC Voltage
Light Intensity Photodiode Passive Current
Department of Electronics and Instrumentation Engineering,
10
B.M.S.C.E., B'luru
Classification of sensors

2.
• Contact and Noncontact sensors

Sensors have been used in the Various applications.

• Strain gauges are contact sensors, but a proximity sensor is not

Department of Electronics and Instrumentation Engineering,

11
B.M.S.C.E., B'luru
Classification of sensors

3. • Absolute and Relative

• Absolute sensor reacts to a stimulus in reference to an absolute scale.

• Ex: Thermistor. Its output is absolute. Its resistance relates to the absolute
temperature.
Capacitance proximity sensor is an absolute sensor
• Relative sensor’s output depends on a relative scale.
• Ex: Output of a thermocouple depends on the temperature difference
between two junctions. The sensed (measured) quantity is the temperature
difference rather than absolute temperature.
• Pressure sensors are relative sensors.
Department of Electronics and Instrumentation Engineering,
12
B.M.S.C.E., B'luru
 Classification of sensors
• Application, Physical phenomena used,
4. Detection method, Sensor specifications
By area of By measured By physical By By area of
Others
detection output effects specifications application
• Resistive
Electrostrictive
Capacitive
Electroresistive
Electric Inductive
Electrochemical Accuracy
Magnetic Current Consumer products
Electro-optic Sensitivity
Electromagnetic Voltage Military applications
Magnetoelectric Stability Power
Acoustic Resonant Infrastructure
Magnetocaloric Response time Interfaces
Optical
Chemical Magnetostrictive Energy
Mechanical Hysteresis Structure
Optical Magnetoresistive Manufacturing
Photoelectric Frequency Heat/thermal
Thermal Linearity Automotive
Photoelastic response Transportation
Temperature Hardness Avionics
Photomagnetic Input (stimulus) Marine
Mechanical Cost Space
Photoconductive
Size range
Radiation Thermomagnetic Scientific
Weight Resolution
Biological Thermoelastic
Construction materials
Thermo-optic Department of Electronics and Instrumentation Engineering,
13
Operating temp. B.M.S.C.E., B'luru
Thermoelectric
Classification of sensors

5. • Internal and External

• Internal sensors -Measure internal state of a robot (position, velocity, acceleration,
etc.)
• Provide input to the controller for decision-making
• Types: Categorized based on the quantity measured
• Position Sensors: Measure the robot's position
• Velocity Sensors: Measure the robot's velocity
• Acceleration Sensors: Measure the robot's acceleration
• Force Sensors: Measure the forces exerted on the robot

• External sensors - are primarily used to learn more about a robot’s environment,
especially the objects being manipulated.
• Types: Contact type and Noncontact type.

Department of Electronics and Instrumentation Engineering,

14
B.M.S.C.E., B'luru
Sensor Technology- MEMS

• MEMS (Micro-ElectroMechanical Systems)

- Micro → Small in size 10-6 m
- Electro → Controllable by electrical signals such as computers
- Mechanical → Moveable for sensing and actuation
- System → Integration of many devices

Department of Electronics and Instrumentation Engineering,

15
B.M.S.C.E., B'luru
Sensor Technology -Components of MEMS

➢Microelectronics:
• “brain” that receives, processes, and makes decisions
• Data comes from microsensors
➢Microsensors:
• Constantly gather data from environment
• Pass data to microelectronics for processing
• Can monitor mechanical, thermal, biological, chemical optical,
and magnetic readings
➢Microactuator:
• Acts as trigger to activate external device
• Microelectronics will tell microactuator to activate device
➢Microstructures:
• Extremely small structures built onto surface of chip
Department of Electronics and Instrumentation Engineering,
16
B.M.S.C.E., B'luru
Sensor Technology- MEMS

• Classically micromachined from silicon.

• Various types of silicon wafers exist and silicon can be doped to
varying levels of conductivity.
• Additional functional materials can be added to provide various
capabilities, such as electrode layers or piezoelectric layers.
• MEMS design and fabrication involves a series of steps and cycles

Department of Electronics and Instrumentation Engineering,

17
B.M.S.C.E., B'luru
Sensors in Robotics - Classification of sensors

Department of Electronics and Instrumentation Engineering,

18
B.M.S.C.E., B'luru
Sensors in Robotics : Major Capabilities - Robot
Simple Touch • The presence or absence of an object.
Taction or
• The presence of an object plus some
Complex
information on its size and shape.
Touch
Simple Force • Measured force along a single axis.
Complex
• Measured force along two or more axes.
Force
Proximity • Noncontact detection of an object.
• Detection of edges, holes, corners, and so
Simple Vision
on.
Complex
• Recognition of shapes.
Vision Department of Electronics and Instrumentation Engineering,
B.M.S.C.E., B'luru
19
Sensors in Robotics : Internal Sensors
Position Sensors

1. Encoder
• Incremental Linear Encoder
• Absolute Linear Encoder
• Incremental Rotary Encoder
• Absolute Rotary Encoder
2. Potentiometer
3. LVDT
4. Synchros and Resolver

Velocity Sensors

1. All Position Sensors

2. Tachometer
3. Hall-effect Sensor

Acceleration Sensors

Force Sensors

1. Strain-gauge Based
Department of Electronics and Instrumentation Engineering,
2. Piezoelectric Based 20
B.M.S.C.E., B'luru
Control Scheme in Robotics : Internal Sensors

Department of Electronics and Instrumentation Engineering,

21
B.M.S.C.E., B'luru
 Control Scheme in Robotics : Internal Sensors

Cincinnati Milacron robot arm

PUMA 560 series robot ann

Department of Electronics and Instrumentation Engineering,
22
B.M.S.C.E., B'luru
 Sensors in Robotics : Internal Sensors
Position Sensors
• Position sensors are crucial for determining the precise position of a robot's joints.
• Information is essential for calculating the end-effector’s
position and orientation through forward kinematics.
• 1. Encoder
• An encoder is a sensor that detects rotation angle or linear
displacement.
• Encoders are used in devices that need to operate
in high speed and with high accuracy.
• Incremental Linear Encoder
• Measure relative displacement.
• Linear and rotary types.
• Work by counting pulses generated by
light beams passing through a grating.

Department of Electronics and Instrumentation Engineering,

23
B.M.S.C.E., B'luru
Sensors in Robotics : Internal Sensors

4-slit incremental 8-slit incremental

type rotary encoder type rotary encoder

Department of Electronics and Instrumentation Engineering,

Courtesy : https://www.akm.com/global/en/products/rotation-angle-sensor/tutorial/role-encoder/
B.M.S.C.E., B'luru
24
Sensors in Robotics : Internal Sensors

Position Sensors
1. Encoder
• Absolute Linear Encoder
•Directly measure absolute position.
•Less prone to error at high speeds.
•Linear and rotary types.
•Use a coded pattern of opaque and
transparent segments to provide a digital output.

Department of Electronics and Instrumentation Engineering,

25
B.M.S.C.E., B'luru
Sensors in Robotics : Internal Sensors
Position Sensors

1. Encoder
• Incremental Rotary Encoder:
• Measures angular displacement.
• Emits light pulses as it rotates.
• Counts these pulses to determine the angle of rotation.
• Can detect the direction of rotation.

Department of Electronics and Instrumentation Engineering,

26
B.M.S.C.E., B'luru
 Sensors in Robotics : Internal Sensors
Position Sensors
1. Encoder
• Absolute Rotary Encoder:
• Directly outputs the absolute angular position.
• Uses a coded disk to determine position.
• More precise than incremental rotary encoders.
• Often employs Gray code to minimize errors during transitions.

Department of Electronics and Instrumentation Engineering,

27
B.M.S.C.E., B'luru
 Sensors in Robotics : Internal Sensors
Position Sensors
2. Potentiometer
• A potentiometer, or simply a "pot," is a common type of position sensor that converts mechanical
displacement into an electrical signal.
• Wiper moves, the resistance between the wiper and the ends of the resistive element changes,
resulting in a corresponding change in voltage.

•Types of Potentiometers:
1.Linear Potentiometer

2.Rotary Potentiometer

Department of Electronics and Instrumentation Engineering,

28
B.M.S.C.E., B'luru
 Sensors in Robotics : Internal Sensors
Position Sensors
2. Potentiometer
• Types :
1.Linear Potentiometer:
1.Measures linear displacement (x).
2.The wiper moves along a straight
resistive track.
2.Rotary Potentiometer:
1.Measures angular displacement (θ).
2.The wiper rotates around a circular resistive track.

Department of Electronics and Instrumentation Engineering,

29
B.M.S.C.E., B'luru
 Sensors in Robotics : Internal Sensors
Position Sensors
3. Linear Variable Differential Transformer (LVDT)
• Reliable and accurate sensors that convert linear or angular displacement into a
measurable electrical signal
• High Accuracy Displacement Transducer: Ideal for applications requiring precise
measurement.
• AC Signal Generation: Produces an AC signal whose magnitude varies with core
displacement.
• Magnetic Field Principle: Operates based on the principle of a moving core within a
magnetic field.
• Core and Coil Configuration: Consists of a central core surrounded by two secondary coils
and a primary coil.
• Linear Relationship: Core displacement is linearly related to the voltage amplitude in the
secondary coil.
• Rotary Variable Differential Department
Transformer
of Electronics(RVDT): Similar
and Instrumentation to LVDT but measures angular
Engineering,
30
displacement. B.M.S.C.E., B'luru
 Sensors in Robotics : Internal Sensors
Position Sensors
3. Linear Variable Differential Transformer (LVDT)
• Reliable and accurate sensors that convert linear or angular displacement into a
measurable electrical signal

Department of Electronics and Instrumentation Engineering,

31
B.M.S.C.E., B'luru
 Sensors in Robotics : Internal Sensors
Position Sensors
4. Synchros and Resolver
Synchros and resolvers are electromechanical devices used to measure angular position.
Consist of a rotating rotor and a stationary stator.
• Working:
1. Rotor Rotation: The rotor rotates, inducing a voltage in the stator windings.
2. Voltage Output: The voltage output is proportional to the angular position of the rotor.
3. Digital Conversion: The analog output is converted to a digital signal for computer processing.
• Key Differences:
• Synchros:
• Three stator windings, 120 degrees apart.
• More complex and expensive.
• Resolvers:
• Two stator windings, 90 degrees apart.
• Simpler and more cost-effective.
• Brushless design for increased reliability.
Department of Electronics and Instrumentation Engineering,
• Applications: Robotics | Aerospace | Industrial B.M.S.C.E., automation B'luru
32
 Sensors in Robotics : Internal Sensors
Position Sensors one degree = 60 arc-minutes and
4. Synchros and Resolver one arc-minute = 60 arc-seconds.
• Most resolvers are specified to work over 2 V - 40 V rms (RMS) frequencies 400 Hz - 10 kHz.
• Angular accuracies range from 5 - 0.5 arc-minutes.
• Basic Operation:
1.AC Excitation: AC voltage is applied to the
rotor winding.
2.Induced Voltage in stator windings is proportional
to the sine of the angle between rotor and stator axes.
3.Output voltages from the stator windings
Represent the sine and cosine components of
the angular position.

Department of Electronics and Instrumentation Engineering,

33
B.M.S.C.E., B'luru
 Sensors in Robotics : Internal Sensors
Velocity Sensors
▪ Measure the rate of change of position.
▪ Offer more direct and efficient measurement.
• Common Velocity Sensors:
1.Derived from Position Sensors:
2.Tachometers:
3.Hall-Effect Sensors:
•These sensors provide real-time velocity information, crucial for precise control and
feedback in robotic systems.

Department of Electronics and Instrumentation Engineering,

34
B.M.S.C.E., B'luru
 Sensors in Robotics : Internal Sensors
Velocity Sensors
▪ Measure the rate of change of position.
▪ Offer more direct and efficient measurement.
• Common Velocity Sensors:
1.Derived from Position Sensors:
Incremental Encoders: By counting pulses over a specific time interval, we can
calculate the average velocity. However, this method can be computationally intensive.

Department of Electronics and Instrumentation Engineering,

35
B.M.S.C.E., B'luru
 Sensors in Robotics : Internal Sensors
Velocity Sensors
▪ Measure the rate of change of position.
▪ Offer more direct and efficient measurement.
• Common Velocity Sensors:
2.Tachometers:
Electromagnetic Tachometers: Coil rotating in a magnetic field generates a voltage proportional to
rotational speed.
Permanent Magnet Tachometers: Rotating magnet induces a voltage in a stationary coil, proportional
to rotational speed.

Department of Electronics and Instrumentation Engineering,

36
B.M.S.C.E., B'luru
 Sensors in Robotics : Internal Sensors
Velocity Sensors
▪ Measure the rate of change of position.
▪ Offer more direct and efficient measurement.
• Common Velocity Sensors:
3.Hall-Effect Sensors:
•A semiconductor device that generates a
voltage when exposed to a magnetic field.
•By using a rotating magnet, the induced
Voltage is proportional to rotational speed.

Department of Electronics and Instrumentation Engineering,

37
B.M.S.C.E., B'luru
 Sensors in Robotics : Internal Sensors
Acceleration Sensors
▪ Two methods to measure acceleration:
1.Indirect Measurement:
Velocity Measurement: Derive acceleration by differentiating the velocity obtained from
velocity sensors or calculated from position sensor data.
Drawback: This method is computationally intensive and susceptible to noise amplification
during differentiation.

Department of Electronics and Instrumentation Engineering,

38
B.M.S.C.E., B'luru
 Sensors in Robotics : Internal Sensors
Acceleration Sensors
▪ Two methods to measure acceleration:
2. Direct Measurement:
Force Measurement: Use strain gauges to measure force, then calculate acceleration using
Newton's second law (F = ma).
•The equation for force, as measured by a strain gauge, is typically expressed as:
F = (ΔR/R) * A * E
where: ΔR: Change in resistance of the strain gauge R: Initial resistance of the strain gauge
A: Cross-sectional area of the strain gauge E: Elastic modulus of the strain gaugeOr

Acceleration a = [(ΔR/R) * A * E] / m

Advantage: More direct and less prone to noise amplification.

•Recommends using integration over differentiation to obtain velocity and position from
acceleration data, as integration tends toof Electronics
Department suppress noise
and Instrumentation
B.M.S.C.E., B'luru
Engineering,
39
 Sensors in Robotics : Internal Sensors
Force Sensors

▪Force sensors are devices used to measure the magnitude of a force applied to them.
▪Here are three common types:
1.Strain Gauge-Based Sensors

2.Piezoelectric Sensors

3.Current-Based Sensors

Department of Electronics and Instrumentation Engineering,

40
B.M.S.C.E., B'luru
 Sensors in Robotics : Internal Sensors
Force Sensors
1. Strain Gauge-Based Sensors:
Principle: When a force is applied to a strain gauge, its resistance changes due to changes in its
length and cross-sectional area.

Measurement: Strain gauges are often arranged in a Wheatstone bridge configuration to measure
these resistance changes accurately.

Advantages: Cost-effective, accurate, and widely used.

Department of Electronics and Instrumentation Engineering,

41
B.M.S.C.E., B'luru
 Sensors in Robotics : Internal Sensors
Force Sensors
2. Piezoelectric Sensors:
Principle: Certain materials (like quartz, tourmaline) generate an electrical charge when subjected
to mechanical stress.
Measurement: This electrical charge is proportional to the applied force and can be measured to
determine the force.
Advantages: High sensitivity,
Fast response time,
Suitable for measuring dynamic forces.

Department of Electronics and Instrumentation Engineering,

42
B.M.S.C.E., B'luru
 Sensors in Robotics : Internal Sensors
Force Sensors
3.Current-Based Sensors:
Principle: The current drawn by an electric motor is directly related to the torque it
produces.

Measurement: By measuring the motor's current and knowing its characteristics, the
torque can be calculated.

Advantages: Simple and direct method for torque measurement.

Department of Electronics and Instrumentation Engineering,

43
B.M.S.C.E., B'luru
Sensors in Robotics : External Sensors

Department of Electronics and Instrumentation Engineering,

44
B.M.S.C.E., B'luru
 Sensors in Robotics : External Sensors
• External sensors are essential for robots to perceive and interact with
their environment. They can be categorized into two main types:
• Contact Sensors:
• Directly interact with objects.
• Examples: Limit Switch
• Non-Contact Sensors:
• Sense information without physical contact.
• Examples: Proximity Sensor (Inductive, Capacitive), Semiconductor Displacement
Sensor, vision sensors (cameras), range sensors (LIDAR, sonar), and infrared
sensors.

Department of Electronics and Instrumentation Engineering,

45
B.M.S.C.E., B'luru
 Sensors in Robotics : External Sensors
ContactSensors:
Contact Sensors:

1. Limit Switch
• Limit switches are simple, mechanical devices that detect the presence or absence of an
object.
• Work similarly to household light switches, using a pressure-sensitive arm or magnetic
activation.
• Types of Limit Switches:
• Normally Open (NO): The circuit is open until the switch is activated.
• Normally Closed (NC): The circuit is closed until the switch is activated.
• Applications in Robotics:
• End-Stop Detection: Prevent robotic arms from exceeding their physical limits.
• Safety Interlocks: Ensure safe operation by halting movement in hazardous conditions.
• Limitations:
• Mechanical Wear: Prone to physical damage and failure.
• Slower Response Time: Compared to non-contact sensors like photoelectric sensors.
Department of Electronics and Instrumentation Engineering,
46
B.M.S.C.E., B'luru
 Sensors in Robotics : External Sensors
Non- Contact
Non- Contact Sensors:
Sensors:
1. Proximity Sensors
Proximity sensors are non-contact electronic devices used to detect the presence or absence
of an object. They are widely used in automation and control systems.
• Main types of proximity sensors:
Inductive Proximity Sensors:
Detect metallic objects.
Work by generating a magnetic field that is disrupted by the presence of a metal object.
The change in the magnetic field is detected, triggering an output signal.
Capacitive Proximity Sensors:
Detect both metallic and non-metallic objects.
Work by measuring changes in capacitance between the sensor and the target object.
They are more sensitive to environmental factors like humidity and temperature.
• Key factors influencing proximity sensor performance:
• Sensing range: The distance at which the sensor can detect an object.
• Sensitivity: The ability to detect small or distant objects.
• Response time: The speed at which the sensor can detect and respond to changes.
• Environmental factors: Factors like temperature, humidity, and electromagnetic
Department of Electronics and Instrumentation Engineering,
interference can affect sensor performance. B.M.S.C.E., B'luru
47
 Sensors in Robotics : External Sensors
Non- Contact
Non- Contact Sensors:
Sensors:
1. Proximity Sensors
Proximity sensors are non-contact electronic devices used to detect the presence or absence
of an object. They are widely used in automation and control systems.
• Main types of proximity sensors:
a. Inductive Proximity Sensors:
Detect metallic objects.
Work by generating a magnetic field that is disrupted by the presence of a metal object.
The change in the magnetic field is detected, triggering an output signal.
b. Capacitive Proximity Sensors:
Detect both metallic and non-metallic objects.
Work by measuring changes in capacitance between the sensor and the target object.
They are more sensitive to environmental factors like humidity and temperature.
• Key factors influencing proximity sensor performance:
• Sensing range: The distance at which the sensor can detect an object.
• Sensitivity: The ability to detect small or distant objects.
• Response time: The speed at which the sensor can detect and respond to changes.
• Environmental factors: Factors like temperature, humidity, and electromagnetic
Department of Electronics and Instrumentation Engineering,
interference can affect sensor performance. B.M.S.C.E., B'luru
48
 Sensors in Robotics : External Sensors
Non- Contact Sensors:
1. Proximity Sensors
a. Inductive Proximity Sensors:
• An inductive proximity sensor is a non-contact device used to detect the presence of metallic objects. It
comprises four key components:
• Sensor Coil and Ferrite Core: Generates a high-frequency electromagnetic field.
• Oscillator Circuit: Produces the high-frequency signal.
• Detector Circuit: Monitors the oscillator's signal amplitude.
• Solid-State Output Circuit: Provides the output signal.
• How it works:
• Field Generation: The oscillator circuit generates a magnetic field, focused by the ferrite core.
• Target Detection: When a metal object enters the field, eddy currents are induced in the object, causing the
oscillator's signal amplitude to decrease.
• Signal Processing: The detector circuit senses this change and triggers the output circuit.
• Output Signal: The output circuit switches on, indicating the presence of the metal object.
• Key Factors Affecting Performance:
• Sensing Range: The maximum distance at which the sensor can detect a target.
• Environmental Factors: Factors like temperature, humidity, and electromagnetic interference can affect
sensor performance.
• Inductive proximity sensors are widely usedDepartment
in automation and and
of Electronics control systemsEngineering,
Instrumentation for applications such as object
detection, position sensing, and limit switching. 49
B.M.S.C.E., B'luru
 Sensors in Robotics : External Sensors
Non- Contact Sensors:
1. Proximity Sensors
b. Capacitive proximity sensors
• Non-contact devices that detect the presence of an object, both metallic and non-metallic. They work by measuring
changes in capacitance between the sensor and the target object.
• How they work:
1.Capacitance: The sensor and the target object form the two plates of a capacitor.
2.Oscillator Circuit: An oscillator circuit generates an alternating current.
3.Capacitance Change: When an object approaches the sensor, the capacitance changes.
4.Signal Detection: The change in capacitance is detected by the sensor's circuitry.
5.Output Signal: The sensor generates an output signal indicating the presence of the object.
• Key Advantages:
• Non-metallic Detection: Can detect non-metallic objects.
• High Sensitivity: Can detect small or lightweight objects.
• Fast Response Time: Suitable for high-speed applications.
• Long Lifespan: Durable and reliable.
• Noise Immunity: Less susceptible to electrical noise.
• Limitations:
• Environmental Sensitivity: Affected by moisture and humidity.
• Limited Sensing Range: Requires a closerDepartment
proximity
of to the target.
Electronics and Instrumentation Engineering,
50
• Capacitive proximity sensors are widely used in various industries,
B.M.S.C.E.,including
B'luru automation, robotics, and manufacturing.
 Sensors in Robotics : External Sensors
Non- Contact Sensors:
2. Semiconductor Displacement Sensor
• A semiconductor displacement sensor utilizes a light source (LED or laser) and a
Position-Sensitive Detector (PSD) to measure displacement.
• How it works:
1. Light Emission: The light source emits a beam of light.
2. Target Reflection: The beam is focused onto a target, which reflects it back.
3. PSD Detection: The reflected beam is focused onto the PSD.
4. Displacement Measurement: The movement of the beam spot on the PSD is
directly proportional to the displacement of the target.
• By analyzing the position of the beam spot on the PSD, the sensor can accurately
determine the displacement of the target.

Department of Electronics and Instrumentation Engineering,

51
B.M.S.C.E., B'luru
Sensors in Robotics : External Sensors
Non- Contact Sensors:
3. Vision
• Vision, a crucial field in computer science, involves extracting information from
images captured by cameras or eyes. It's widely studied and applied in various fields,
including robotics.
• Key applications of vision systems in robotics:
• Part localization: Identifying and locating objects in a scene for tasks like picking and
placing.
• Object recognition: Recognizing and classifying objects based on their visual
features.
• Inspection: Detecting defects or anomalies in products or manufacturing processes.
• Navigation: Guiding robots through environments by analyzing visual information.
• Motion tracking: Tracking the movement of objects or people for various purposes.
• Vision systems offer a cost-effective alternative to traditional Hardware components
fixtures, of a vision
enabling moresystem
Department of Electronics and Instrumentation Engineering,
flexible and adaptable robotic systems. B.M.S.C.E., B'luru
52
Sensors in Robotics : External Sensors

Non- Contact Sensors:

3. Vision
Definition: The process of extracting information about the external world from images
captured by cameras or eyes.
Key Terms: Computer Vision, Machine Vision, Robot Vision
Importance:Enables machines to perceive and interact with their environment.
Applications of Vision Systems in Robotics
• Inspection: Detecting defects and flaws
• Verifying component presence and assembly correctness
• Measuring dimensional accuracy
• Identification: Recognizing and classifying objects
• Visual Servoing and Navigation: Controlling robot movements based on visual input
• Guiding robot end-effectors towards objects
• Navigating through environments
Department of Electronics and Instrumentation Engineering,
53
B.M.S.C.E., B'luru
 Sensors in Robotics : External Sensors
Non- Contact Sensors:
3. Vision
How Vision Systems Work
➢Image Acquisition: Capturing images using cameras

➢Image Processing: Preprocessing images to enhance features

➢Feature Extraction: Identifying relevant features like edges, corners & texture

➢Object Recognition: Matching extracted features with known object models

➢Decision Making: Using the recognized objects to make decisions and control

robot actions
Department of Electronics and Instrumentation Engineering,
54
B.M.S.C.E., B'luru
 Sensors in Robotics : External Sensors

Non- Contact Sensors:

3. Vision
Challenges:
• Lighting conditions
• Occlusions
• Real-time processing
• Future Trends:
• Deep learning and AI
• 3D vision
• Autonomous systems
• Integration with other sensors
Department of Electronics and Instrumentation Engineering,
55
B.M.S.C.E., B'luru
Sensors in Robotics : External Sensors

Non- Contact Sensors:

3. Vision
Elements in a Vision Sensor
• Primary Component: The camera, including the sensor, lens, electronics, and signal
output.
• Sensor Function: Measures the intensity of light reflected from objects using
photosensitive elements (pixels).
• Sensor Types: CCD, CMOS, and others, each with unique energy transformation principles.
• Camera Types:
• RS-170/CCIR: Older analog standards (monocolor)
• NTSC/PAL: Color television standards
• Progressive Scan: Captures all lines of an image simultaneously
• Variable Scan: Adjusts scanning speed
• Line Scan: Captures a singleDepartment
line ofofthe image
Electronics at a time
and Instrumentation Engineering,
56
B.M.S.C.E., B'luru
Sensors in Robotics : External Sensors

Non- Contact Sensors:

3. Vision
Elements in a Vision Sensor
• Key Camera Parameters:
• Field of View: Area captured by the camera
• Resolution: Number of pixels in the image
• Working Distance: Distance between camera and object
• Depth of Field: Range of distances where objects are in focus
• Image Data Acquisition Rate: Speed at which the camera captures images
• Resolution Rule of Thumb: For size measurements, the sensor should have
at least twice the number of pixels as the ratio of the largest to smallest
object sizes. Department of Electronics and Instrumentation Engineering,
57
B.M.S.C.E., B'luru
Sensors in Robotics : External Sensors
Non- Contact Sensors:
3. Vision
Elements in a Vision Sensor
1. Camera Systems:
• Photosensitive sensor
• Lens (focuses light onto the image plane)
• Shutter
• Analog preprocessing electronics
• Coordinate Transformations: Necessary to compute object position and orientation using camera data.
• Camera Types:
• Analog: Requires a frame grabber (A/D converter) to interface with the computer.
• Digital: Directly outputs digital data, eliminating the need for a frame grabber.
• Vision Software:
• Processes the captured image data.
• Communicates results to control robot actions or update databases.
• System Considerations:
• Complexity: Robot vision systems are less versatile than the human visual system.
Department of Electronics and Instrumentation Engineering,
• Simplicity: System design should focus on clear and achievable goals. 58
B.M.S.C.E., B'luru
Sensors in Robotics : External Sensors

Non- Contact Sensors:

3. Vision
Elements in a Vision Sensor
• Vidicon Cameras: An Overview
• Legacy Technology: Vidicon cameras were early, bulky vacuum tube-based cameras.
They are now obsolete.
• Characteristics:
• Advantages: High resolution and light sensitivity.
• Disadvantages: Susceptible to electromagnetic noise, high power consumption.
• Operation:
• Light-Sensitive Mosaic: Converts light intensity into varying resistance.
• Electron Beam Scanning: An electron beam scans the mosaic, generating an output
voltage proportional to light intensity.
• Analog Signal: The outputDepartment
is an analog signal representing the image.
of Electronics and Instrumentation Engineering,
B.M.S.C.E., B'luru
59
Sensors in Robotics : External Sensors

Non- Contact Sensors:

3. Vision
Elements in a Vision Sensor
• Vidicon Cameras: An Overview
• Analog-to-Digital Conversion (ADC) Process:
1.Sampling:
1.Converts the continuous analog signal into discrete-time samples.
2.Sampling rate must be at least twice the highest frequency in the signal to avoid aliasing (loss
of information).
2.Quantization:
1.Assigns a discrete amplitude level to each sampled value.
2.Number of quantization levels determines the precision (e.g., 8-bit ADC allows 256 levels).
3.Encoding:
1.Converts the quantized amplitude levels into digital codes (0s and 1s).
2.The accuracy of encoding depends
Departmenton the spacing
of Electronics of quantization
and Instrumentation Engineering, levels. 60
B.M.S.C.E., B'luru
Sensors in Robotics : External Sensors
Non- Contact Sensors:
3. Vision
Elements in a Vision Sensor
• Digital Cameras: A Revolution in Imaging
• Core Technology: Solid-state silicon wafer with numerous photosensitive areas (pixels).
• Image Capture: Light falling on each pixel generates a charge proportional to its intensity.
• Common Types:
• CCD (Charge-Coupled Device): Captures light like RAM, reading charges sequentially.
• Interline Transfer: Transfers charge between lines.
• Frame Transfer: More flexible integration time.
• CID (Charge-Injection Device): Uses MOS technology and carrier recombination to generate the video signal.
Offers better image quality but is more expensive than CCD.
• Advantages over Vidicon: Smaller, more rugged, longer lifespan, less image distortion.
• Key Points:
• Digital cameras utilize solid-state technology for image capture.
• CCD and CID are the primary sensor types, each with distinct charge handling and readout mechanisms.
• Digital cameras offer significant improvements over older vidicon technology in terms of size, durability, and
Department of Electronics and Instrumentation Engineering,
image quality. B.M.S.C.E., B'luru
61
Sensors in Robotics : External Sensors

Non- Contact Sensors:

3. Vision
Elements in a Vision Sensor
2. Lighting Techniques
• Key Factors in Object Illumination for Robot Vision
• Illumination's Impact on Image Brightness:
• Incident radiance, object reflectivity, camera response, and lens
characteristics all influence image brightness.
• Importance of Proper Illumination:
• Creates contrast between object features, simplifying image processing.
• Avoids unwanted reflections and shadows unless they are necessary for
image analysis. Department of Electronics and Instrumentation Engineering,
62
B.M.S.C.E., B'luru
Sensors in Robotics : External Sensors
Non- Contact Sensors:
3. Vision
Elements in a Vision Sensor
2. Lighting Techniques
• Common Lighting Techniques:
1.Direct Incident Lighting:
• Suitable for non-reflective surfaces.
• Uses direct light, often from a ring light or halogen lamps.
• Minimizes shadows.
2.Diffuse Incident Lighting
• Necessary for reflective, polished, or glossy objects.
• Achieved through diffusers, coaxial illumination, or dome-shaped illumination.
3. Lateral Lighting:
• Illuminates objects from the side.
• Can significantly affect image appearance, even inverting information in extreme cases.
4. Dark Field Lighting:
• Illuminates at a shallow angle, making the field of view dark.
• Highlights surface defects (edges, scratches) by reflecting light towards the camera.
5. Backlighting:
• Illuminates objects from behind.
• Ideal for accurate measurements, as the objectDepartment
itself of
is Electronics
silhouetted andagainst
Instrumentation Engineering,
the light. 63
B.M.S.C.E., B'luru
 Sensors in Robotics : External Sensors

Non- Contact Sensors:

3. Vision
Steps in a Vision System
1. Image Acquisition
• Process: Capturing an image using a camera
(vidicon, CCD, or CID) and
storing it in computer memory (frame buffer).
• Key Elements:
• Light source (controlled or ambient)
• Lens (focuses light onto the image sensor)
Digitized picture
• Image sensor (converts light into electrical signals)
• Electronics (reads and transmits image
Department of Electronics anddata to
Instrumentation
B.M.S.C.E., B'luru
the computer)
Engineering,
64
Sensors in Robotics : External Sensors

Non- Contact Sensors:

3. Vision
Steps in a Vision System
2. Image Processing
• Purpose: Enhance, improve, or alter the image to prepare it for analysis.
• Key Techniques:
• Image Data Reduction
• Histogram Analysis
• Thresholding
• Masking
• Edge Detection
• Segmentation
• Morphological Operations
Department of Electronics and Instrumentation Engineering,
65
B.M.S.C.E., B'luru
 Sensors in Robotics : External Sensors

Non- Contact Sensors:

3. Vision
Steps in a Vision System
2. Image Processing
• Purpose: Enhance, improve, or alter the image to prepare it for analysis.
• Key Techniques:
• Image Data Reduction:
• Digital conversion (reducing gray levels)
• Windowing (analyzing only a portion of the image)
• Histogram Analysis:
• Analyzing the distribution of pixel intensities.
• Useful for thresholding.
• Thresholding:
Department of Electronics and Instrumentation Engineering,
• Dividing the image into different regions based
B.M.S.C.E., B'luru on pixel intensity values.
66
Sensors in Robotics : External Sensors
Non- Contact Sensors:
3. Vision
Steps in a Vision System
2. Image Processing
• Key Techniques: Masking of an image
• Masking:
• Purpose: Masks are versatile tools in image processing, serving various functions like
filtering (e.g., low-pass filtering to reduce noise), and more.
• Low-pass Filtering: A low-pass filter attenuates high-frequency components (noise) while
preserving low-frequency details in an image.
• Noise Reduction: By attenuating high-frequency noise, low-pass filtering effectively
reduces noise in images.
• Masking Process:
• Involves applying a mask (a small matrix of weights) to a localized region of an image.
• Example: A 3x3 mask is applied to a portion of an image with a single pixel significantly
Department of Electronics and Instrumentation Engineering,
different from its neighbors (considered noise)
B.M.S.C.E., using
B'luru neighborhood averaging 67
Sensors in Robotics : External Sensors

Non- Contact Sensors:

3. Vision
Steps in a Vision System
2. Image Processing
• Key Techniques:
Hough transform
• Edge Detection:
• This process identifies significant changes in image intensity, such as edges of objects,
using various techniques.
• Limitations : Often, these techniques produce discontinuous edges.
• Hough Transform: This technique addresses the discontinuity issue by:
• Representing lines in the image as points in a "Hough space" (m-c plane).
• Identifying collinear points in the image by finding points in Hough space that
intersect.
• Determining the orientation of lines and, consequently, the orientation of objects in
the image. Department of Electronics and Instrumentation Engineering,
B.M.S.C.E., B'luru
68
Sensors in Robotics : External Sensors
Non- Contact Sensors:
3. Vision
Steps in a Vision System
2. Image Processing
• Key Techniques:
• Segmentation: Skeletonization
• Dividing the image into meaningful regions.
• Methods include region growing, region splitting.
• Morphological Operations:
• Purpose: These operations are used to manipulate the shapes of objects within an image.
• Applications:
• Image Analysis: They help in understanding and extracting meaningful features from images.
• Noise Reduction: They can remove "extra" information or noise that might interfere with analysis.
• Types of Operations:
• Binary: Applicable to images with only two color values (black and white).
• Gray-scale: Applicable to images with varying shades of gray.
• Examples of Operations:
• Dilation: Expands the boundaries of objects.
• Erosion: Shrinks the boundaries of objects.
• Skeletonization: Reduces objects to their skeletal representation
Department of Electronics and Instrumentation Engineering,
• Opening/Closing: Combinations of erosion and dilation used to refine object shapes.
B.M.S.C.E., B'luru
69
Sensors in Robotics : External Sensors

Non- Contact Sensors:

3. Vision
Steps in a Vision System
3. Image Analysis
• Purpose: Extracting information from the processed image.
• Techniques:
• Feature Extraction:
• Identifying and extracting characteristic features of objects (e.g., gray-level histograms,
perimeter, area, eccentricity).
• Object Recognition:
• Identifying and classifying objects within the image.
• Position and Orientation Analysis:
• Determining the location and orientation of objects in the image.
• Depth Information Extraction:
• Determining the distance of objects from the camera.
Department of Electronics and Instrumentation Engineering,
70
B.M.S.C.E., B'luru
 Sensors in Robotics : External Sensors
Non- Contact Sensors:
3. Vision
Hierarchy of a Vision System
Visual Perception Hierarchy
• Conceptual Framework: Visual perception is often described
as a hierarchical process, moving from
"low-level" to "intermediate" and finally "high-level" stages.
• Defining Levels: While commonly used, the precise definitions of
"low," "intermediate," and "high-level" vision are not universally agreed upon.
• Classification Criteria: Figure illustrates how specific image-processing
activities are categorized within these levels.

Department of Electronics and Instrumentation Engineering,

71
B.M.S.C.E., B'luru
 Sensors in Robotics : External Sensors
Non- Contact Sensors:
3. Vision
Hierarchy of a Vision System
1. Low-Level Vision
• Focus: Deals with the initial stages of visual processing, including:
• Image acquisition and formation.
• Extraction of basic physical properties like depth, shape, object
boundaries, and surface properties.
• Key Activities:
• Feature Extraction: Identifying edges, corners, and optical flow
within the image.
• Image Processing: Analyzing pixel intensities and depths to extract
information about regions, lines, and surface characteristics (peaks, pits, ridges).
• Characteristics:
• Spatially Uniform and Parallel: Similar processing occurs across most of the visual field.
• Bottom-Up Processing: Driven primarily by the image data itself, with minimal influence from higher-
Department of Electronics and Instrumentation Engineering,
level cognitive factors or specific task demands. B.M.S.C.E., B'luru 72
 Sensors in Robotics : External Sensors
Non- Contact Sensors:
3. Vision
Hierarchy of a Vision System
2. Intermediate-Level Vision
• Focus:
• Recognizing objects and interpreting 3D scenes.
• Building upon the features extracted during low-level vision.
• Key Activities:
• Grouping: Combining individual elements (pixels, lines, etc.) into
meaningful groups.
• Object Recognition: Identifying and categorizing objects based on
their shape, size, and other properties.
• Scene Interpretation: Understanding the spatial relationships between objects in a 3D environment.
• Key Differences from Higher-Level Vision:
• Data-Driven: Primarily relies on the information extracted from the image itself.
• Limited Object-Specific Knowledge: Does not heavily depend on prior knowledge about specific objects
Department of Electronics and Instrumentation Engineering,
or their typical appearances. B.M.S.C.E., B'luru
73
 Sensors in Robotics : External Sensors
Non- Contact Sensors:
3. Vision
Hierarchy of a Vision System
3.High-Level Vision
• Focus:
• Interpreting scenes by recognizing and understanding the objects within them.
• Utilizing knowledge about specific objects and their relationships.
• Emphasizes understanding the meaning and use of information in the image,
rather than just extracting physical properties.
• Key Activities:
• Object Recognition: Identifying and classifying objects based on their shape,
size, and context.
• Scene Interpretation: Understanding the relationships between objects, their functions, and the overall context of the
scene.
• Decision Making: Determining object boundaries, identifying occluded objects, and distinguishing object structures from
surface textures and shadows.
• Characteristics:
• Object-Oriented: Focused on recognizing and understanding specific objects.
• Top-Down Processing: Influenced by prior knowledge and expectations about the objects and scene.
• Selective Attention: Processes specific regions
Department
of theofimage
Electronics and Instrumentation
based on interest or Engineering,
task demands. 74
B.M.S.C.E., B'luru