Robotic Agents (CMPSC 311) Sensors: Perception Janyl Jumadinova - - PowerPoint PPT Presentation

Robotic Agents (CMPSC 311)

Sensors: Perception Janyl Jumadinova September 5, 2019

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 1 / 16

Mobile Robots

Robot = sensors + actuators. Actuators make the mobility possible. Sensors are the key components for perceiving the environment.

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 2 / 16

UK EPSRC’s Principles of Robotics

1 Robots are multi-use tools. Robots should not be designed solely or

primarily to kill or harm humans, except in the interests of national security.

2 Humans, not robots, are responsible agents. Robots should be

designed and operated as far as is practicable to comply with existing laws, fundamental rights and freedoms, including privacy.

3 Robots are products. They should be designed using processes which

assure their safety and security.

4 Robots are manufactured artifacts. They should not be designed in a

deceptive way to exploit vulnerable users; instead their machine nature should be transparent.

5 The person with legal responsibility for a robot should be attributed. Bryson, Connection Science 2017 Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 3 / 16

The Ethics in Robots

Are ethical robots possible or even desirable? https://forms.gle/xYrwrJ631rsm3Sju8

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 4 / 16

The Ethics in Robots

Are ethical robots possible or even desirable? https://forms.gle/xYrwrJ631rsm3Sju8 How Might an Ethical Robot be Compromised? Through Perception?

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 4 / 16

Perception is hard!

Understanding = raw data + (probabilistic) models + context Intelligent systems interpret: raw data according to probabilistic models and use contextual information that gives meaning to the data.

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 5 / 16

Perception is hard!

• S. Pinker. The Language Instinct. New York: Harper Perennial Modern Classics, 1994

“In robotics, the easy problems are hard and the hard problems are easy.”

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 6 / 16

Perception for Mobile Robots

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 7 / 16

Sensors

Optical encoders Heading sensors: Compass, Gyroscopes Accelerometer IMU (Inertial Measurement Unit) GPS Range sensors: Sonar, Laser, Structured light Vision

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 8 / 16

Sensor Modality Sensors which measure same form of energy and process it in similar ways. “Modality” refers to the raw input used by the sensors.

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 9 / 16

Sensor Modality Sensors which measure same form of energy and process it in similar ways. “Modality” refers to the raw input used by the sensors. Different modalities: Sound Pressure Temperature Light (Visible light, Infrared light, X-rays, Etc.)

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 9 / 16

Classification of Sensors: What

Proprioceptive sensors measure values internally to the system (robot),

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 10 / 16

Classification of Sensors: What

Proprioceptive sensors measure values internally to the system (robot), e.g. motor speed, wheel load, heading of the robot, battery status.

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 10 / 16

Classification of Sensors: What

Proprioceptive sensors measure values internally to the system (robot), e.g. motor speed, wheel load, heading of the robot, battery status. Exteroceptive sensors information from the robots environment,

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 10 / 16

Classification of Sensors: What

Proprioceptive sensors measure values internally to the system (robot), e.g. motor speed, wheel load, heading of the robot, battery status. Exteroceptive sensors information from the robots environment, e.g., distances to objects, intensity of the ambient light, unique features.

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 10 / 16

Classification of Sensors: How

Passive sensors energy coming from the environment.

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 11 / 16

Classification of Sensors: How

Passive sensors energy coming from the environment. Active sensors

• emit their proper energy and measure the reaction;
• better performance, but some influence on environment.

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 11 / 16

General Classification

General classification Sensor PC or EC A or P Tactile sensors contact switches, bumpers EC P (detection of physical Optical barriers EC A contact or closeness) Noncontact proximity sens EC A Wheel/motor sensors Brush encoders PC P (wheel/motor speed Optical encoders PC A and position) Magnetic encoders, ... PC A

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 12 / 16

Characterizing Sensor Performance

Basic sensor response ratings: Range

• lower and upper limits

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 13 / 16

Characterizing Sensor Performance

Basic sensor response ratings: Range

• lower and upper limits

Resolution

• minimum difference between two values

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 13 / 16

Characterizing Sensor Performance

Basic sensor response ratings: Range

• lower and upper limits

Resolution

• minimum difference between two values

Linearity

• variation of output signal as function of the input signal

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 13 / 16

Characterizing Sensor Performance

Basic sensor response ratings: Range

• lower and upper limits

Resolution

• minimum difference between two values

Linearity

• variation of output signal as function of the input signal

Bandwidth or Frequency

• the speed with which a sensor can provide a stream of readings
• usually there is an upper limit depending on the sensor and the

sampling rate

• lower limit is also possible, e.g. acceleration sensor
• have to also consider signal delay

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 13 / 16

In Situ Sensor Performance

Sensitivity

• ratio of output change to input change
• however, in real world environment, the sensor has very often high

sensitivity to other environmental changes, e.g. illumination

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 14 / 16

In Situ Sensor Performance

Sensitivity

• ratio of output change to input change
• however, in real world environment, the sensor has very often high

sensitivity to other environmental changes, e.g. illumination Cross-sensitivity

• sensitivity to environmental parameters that are orthogonal to the

target parameters

• influence of other active sensors

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 14 / 16

In Situ Sensor Performance

Sensitivity

• ratio of output change to input change
• however, in real world environment, the sensor has very often high

sensitivity to other environmental changes, e.g. illumination Cross-sensitivity

• sensitivity to environmental parameters that are orthogonal to the

target parameters

• influence of other active sensors

Error / Accuracy

• difference between the sensor’s output and the true value

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 14 / 16

In Situ Sensor Performance

Systematic error → deterministic errors

• caused by factors that can (in theory) be modeled → prediction
• e.g. calibration of a laser sensor or of the distortion cause by the
• ptic of a camera

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 15 / 16

In Situ Sensor Performance

Systematic error → deterministic errors

• caused by factors that can (in theory) be modeled → prediction
• e.g. calibration of a laser sensor or of the distortion cause by the
• ptic of a camera

Random error → non-deterministic

• no prediction possible
• however, they can be described probabilistically
• e.g. Hue instability of camera, black level noise of camera

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 15 / 16

In Situ Sensor Performance

Systematic error → deterministic errors

• caused by factors that can (in theory) be modeled → prediction
• e.g. calibration of a laser sensor or of the distortion cause by the
• ptic of a camera

Random error → non-deterministic

• no prediction possible
• however, they can be described probabilistically
• e.g. Hue instability of camera, black level noise of camera

Precision

• reproducibility of sensor results

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 15 / 16

EV3 Sensors

Sensor Framework: https://sourceforge.net/p/lejos/wiki/Sensor%20Framework/

Janyl Jumadinova Robotic Agents (CMPSC 311) September 5, 2019 16 / 16