Aerial Robotics

Sensors

C. Papachristos

Robotic Workers (RoboWork) Lab

University of Nevada, Reno

CS-491/691

Sensors

Sensors:

​

• Providing the ability to estimate the state of the robot – environment setup

​

• Localization – Self-localize and estimate own pose w.r.t. a map of the environment
• Mapping – Build a map representation of the environment

​

• Ego-state / Configuration estimation – Estimate own configuration through measurements
• External measurement – Derive measurements by processing received signals from the environment

​

• Remote External – can also rely on external systems (e.g. Global Positioning System, Ultra-WideBand, etc.) which are engineered solutions tailored to assist localization / state-estimation
• Non-autonomous solutions

​

CS491/691 C. Papachristos

Sensors

Sensor Classifications:

​

Proprioceptive sensors

• Measure values internal-to the robot
• Angular rate, heading

​

Exteroceptive sensors

• Information pertaining to the robot environment
• Distances to objects, extraction of features from the environment

​

Passive sensors

• Measure energy coming from a signal of the environment
• Highly influenced by the environment

​

Active sensors

• Emit own signal and measure environment’s response
• Better performance, but some influence the environment
• Not always easily applicable concept

​

​

CS491/691 C. Papachristos

Uncertainty

Sensor-based Measurement(s) inherently carry Uncertainty

​

Uncertainty representation / quantification is critical

Uncertainty combination and propagation is often required

​

Systematic errors

• Caused by factors or processes that can in theory�be modeled and, thus, calibrated for
• E.g. misalignment of a 3-axes accelerometer

​

Random errors

• Cannot be pinned down to compensation terms,�can only be described in probabilistic terms
• E.g. measurement noise of electro-optical sensors

​

​

CS491/691 C. Papachristos

Robot Sensors

Sensors typically employed for robotic operations including navigation / manipulation etc. :

• Encoders
• Absolute / Incremental measurement, Electrical / Electro-optical principles
• Wheel odometry, Joint positioning, etc.
• Tactile sensors / bumpers
• Detection of physical contact, security switches
• Inertial Sensors
• Accelerometers, Gyroscopes
• Orientation and acceleration of the rigid body
• Magnetometers (digital compass)
• Pressure Sensors
• Barometric pressure for altitude sensing
• Airspeed measurement
• Global Positioning System (GPS)
• Global localization and navigation
• Camera(s)
• Texture information, motion estimation, scene interpretation
• Time-of-Flight / Range sensors
• Laser-based, Structured light, Sonar
• Obstacle avoidance, motion estimation, scene interpretation

CS491/691 C. Papachristos

Encoders

Encoders:

​

Applications:

• Measure shaft position with mounted revolute joints
• Servomotors
• Arms
• Measure position or speed of the wheels or steering
• Integrate wheel motion to estimate odometry
• Optical encoders are proprioceptive sensors
• Typical resolutions: 64 - 2048 increments per revolution
• For higher resolution: interpolation

​

Working principle of optical encoders:

• Regular: Counts the number of transitions but cannot tell the direction of motion
• Quadrature: Uses two sensors in quadrature-phase shift; the ordering of each wave produces a rising edge first tells the direction of motion
• Additionally, resolution is 4x
• A single slot in the outer track generates a reference pulse per revolution

CS491/691 C. Papachristos

Encoders

Encoders:

​

Main principle:

• The four fields on the scanning reticle are shifted in phase relative to each other by one quarter of the grating period, which equals 360°/(number of lines)

​

​

​

​

​

• This configuration allows the detection of a change in direction

​

​

​

​

​

• Easy to interface with a micro-controller
• e.g. wire quadrature channel outputs to interrupt pins

CS491/691 C. Papachristos

Inertial Sensors

Construction schematic of ADXL-203 dual axis accelerometer

CS491/691 C. Papachristos

Inertial Sensors

CS491/691 C. Papachristos

Inertial Sensors

Accelerometer – Simplified Model:

​

For the cases where the vehicle acceleration is constant, then the steady-state output of the accelerometer is also constant, therefore indicating the existence and value of the acceleration

​

​

The undamped natural frequency and the damping ratio of the accelerometer are:

​

​

​

​

​

​

​

CS491/691 C. Papachristos

Inertial Sensors

CS491/691 C. Papachristos

Inertial Sensors

Gyroscope:

​

Devices that measure rate of change in orientation

​

Mechanically a gyroscope is a spinning wheel where the axis of rotation is free to assume any possible orientation by being mounted on a series of gimbals

• When rotating, the orientation of this axis remains unaffected by the tilting/rotation of the mounting frame, according to the conservation of angular momentum
• This is employed to measure the mounting’s orientation change

​

• In robotics & automation we typically use MEMS gyroscopes or solid-state ring lasers, and fibre-optic gyroscopes

MEMS gyroscopes rely on the effects of Coriolis acceleration

• Remember: Coriolis acceleration is an “apparent” acceleration arising in a rotating frame of reference, and is proportional to that rate of rotation
• Employed to measure the frame’s orientation rate of change

​

Lumped structural model of MEMS vibratory gyroscope

CS491/691 C. Papachristos

Inertial Sensors

https://en.wikipedia.org/wiki/Coriolis_force#/media/File:Corioliskraftanimation.gif

CS491/691 C. Papachristos

Inertial Sensors

Gyroscope:

​

Quartz Dipole – Tuning Fork:

• 2 masses forced to oscillate (constant motion in opposite directions)
• The masses –constructed with crystalline quartz– are kept moving by oscillating via electrostatic actuation

​

• When angular velocity is applied, Coriolis force on each is in opposite direction
• When linear acceleration is applied, the forces are in the same direction

​

• Pickup tines for sensing – We have to amplify and demodulate the pickup signal

​

• Drive

​

• Sense

​

​

Input:

Drive Tines

Output:

Pickup Tines

CS491/691 C. Papachristos

Inertial Sensors

Gyroscope:

​

Mechanically suspended Silicon-On Insulator MEMS:

• 1 mass forced to oscillate (constantly back & forth)
• Drive circuitry responsible for actuation

​

• When angular velocity is applied, Coriolis force deflects mass in perpendicular direction w.r.t. phase of its motion cycle

​

​

​

​

​

​

​

​

​

• Sensing relies on differential capacitance measuring displacement

CS491/691 C. Papachristos

Inertial Sensors

Gyroscope:

​

Mechanically suspended Silicon-On Insulator MEMS:

• Example: A pair of 1-mass assemblies
• Oscillating in phase opposition to detect false measurements

​

​

​

​

​

​

​

​

CS491/691 C. Papachristos

Inertial Sensors

Gyroscope:

​

Bias effects:

We may not simply integrate Gyroscope measurements to derive orientation estimates due to measurement bias

​

Biases are caused by:

• Drive excitation feedthrough
• Output electronics offsets
• Bearing torques

​

Biases are present in three forms:

• Fixed Bias
• Bias Stability, from one turn-on to another (thermal)
• Bias Drift, usually modeled as a Random Walk process

Input:

Drive Tines

Output:

Pickup Tines

CS491/691 C. Papachristos

Inertial Sensors

Input:

Drive Tines

Output:

Pickup Tines

CS491/691 C. Papachristos

Magnetic Sensors

Magnetometer:

​

Measures strength and direction of the local magnetic field

​

• The magnetic field measured will be a combination of both the earth's magnetic field and any magnetic field created by nearby objects
• The magnetic field is measured in the sensor reference frame

​

The earth's magnetic field is a self-sustaining magnetic field that resembles a magnetic dipole with one end near the Earth’s geographic North Pole and the other near the Earth’s geographic South Pole

• The strength of this magnetic field varies across the Earth with strengths as low as 0.3 Gauss in South America to over 0.6 Gauss in Northern Canada

​

CS491/691 C. Papachristos

Magnetic Sensors

Magnetic Declination Map:

​

​

World Magnetic Model, National Geophysical Data Center

CS491/691 C. Papachristos

Magnetic Sensors

Magnetic Inclination Map:

​

​

Inclination: Compass needle angle w.r.t. horizontal at any point on the Earth's surface. Positive: downward, into the Earth

CS491/691 C. Papachristos

Pressure Sensors

Pressure Sensor:

​

Measures pressure of gases or liquids

• Pressure is an expression of the force required to stop a fluid from expanding, usually stated in terms of force per unit area
• A pressure sensor usually acts as a transducer; it generates a signal as a function of the pressure imposed

​

Pressure sensing:

Measuring pressure itself, expressed as a force per unit area

• Useful in weather instrumentation, aircraft, automobiles, and any other machinery that has pressure functionality implemented

​

Altitude sensing:

Correlating atmospheric pressure to altitude

• In aircraft, rockets, satellites, weather balloons, and many other applications

CS491/691 C. Papachristos

Pressure Sensors

Strain Sensing

Diaphragm

Transducer

CS491/691 C. Papachristos

Pressure Sensors

Pitot-Tube Pressure Sensor:

​

Altitude Measurement: Usually assume that the density is constant (valid for small altitude variations):

​

​

Airspeed Measurement:

From Bernoulli’s equation:

​

​

Pitot-static Pressure Sensor measures Dynamic Pressure:

Transducer

CS491/691 C. Papachristos

(Bias & White Noise errors)

(Bias & White Noise errors)

Inertial Measurement Unit

Integrated Inertial Measurement Unit (IMU):

​

Uses gyroscopes and accelerometers to estimate the relative pose (position and orientation), velocity and acceleration of a moving vehicle w.r.t. an Inertial Frame

• To estimate the rigid-body motion, the gravity acceleration vector must be subtracted, and the initial velocity has to be known
• After long periods of operation, estimation drifts eventually occur
• Require external “correcting” reference to cancel it

CS491/691 C. Papachristos

Global Positioning System

Global Positioning System (GPS):

​

24 Satellites orbiting the Earth

• And some spares – total of 30 as of 9/9/2020
• 24 operational 95% of the time

​

• Altitude at 20,180 km

​

• Any point on the Earth’s surface can be “seen” by at least 4 satellites at all times

​

• Time-of-Flight of radio signal from 4 satellites to a receiver

​

• 4 range measurements needed to account for clock offset error
• 4 nonlinear equations in 4 unknown results:
• Latitude
• Longitude
• Altitude
• Receiver clock time offset

CS491/691 C. Papachristos

Global Positioning System

CS491/691 C. Papachristos

Global Positioning System

Global Positioning System (GPS):

​

Time-of-Flight of the radio signal from satellite to receiver used�to calculate pseudo-range measurement

​

• Numerous sources of error in Time-of-Flight measurement:
• Ephemeris Data – errors in satellite time and (orbital) location
• GPS atomic clock error accuracy is 1 ns per tick
• Atmosphere refraction
• Ionosphere – upper atmosphere, free electrons slow transmission
• Troposphere – lower atmosphere, weather (temperature and density) effects
• Multipath Reception – signals not following direct path

​

• Small timing errors can result in large position deviations:
• 10ns timing error leads to 3m pseudo-range error

​

• Relativity: Time Dilation, Gravitational Frequency Shift, Eccentricity
• Time on GPS satellite frame, travelling at a distance w.r.t the Earth’s inertial frame –thus at a “high” orbital velocity–, is slowed down
• Relativistic time correction computed based on satellite orbital data (transmitted with GPS signal)

CS491/691 C. Papachristos

Global Positioning System

GPS Error Characterization:

​

Cumulative effect of GPS pseudorange errors is described by the User-Equivalent Range Error (UERE)

​

• UERE has two components:
• Bias
• Random

CS491/691 C. Papachristos

Global Positioning System

GPS Error Characterization:

​

Effect of satellite geometry on position calculation is expressed by Dilution of Precision (DOP)

• Satellites closer together leads to high DOP
• Satellites further apart leads to low DOP
• DOP varies with time

​

Horizontal DOP (HDOP) is smaller than Vertical DOP (VDOP):

• Nominal HDOP = 1.3
• Nominal VDOP = 1.8

​

CS491/691 C. Papachristos

Global Positioning System

GPS Error Characterization:

​

Standard deviation of RMS error in the North-East plane:

​

​

​

​

Standard deviation of RMS altitude error:

​

​

​

​

• An ellipsoidal error model
• Altitude VDOP is higher, making its RMS error standard deviation worse

​

CS491/691 C. Papachristos

Camera(s)

Camera:

​

Electro-Optical imaging sensor

• Receptor element array (Pixels) sensitive to radiation
• Visible light spectrum radiation – “Visible light” cameras
• Near Infra Red radiation – “Night vision” cameras
• Short/Mid/Long-Wave Infra Red radiation – “Thermal” cameras
• Ultra-violet radiation – “UV” cameras
• Gamma radiation – “Nuclear” Ionizing Radiation cameras
• etc…

​

• Radiation focusing element – Lens
• Not always translucent (e.g. Germanium)

​

• “Intrinsic” camera parameters:
• Size of pixels
• Focal length (also “Extrinsic”)
• Distortion model (also “Extrinsic”)

CS491/691 C. Papachristos

Camera(s)

Physical Model

Virtual Model

(Projection Plane in front of Pinhole)

CS491/691 C. Papachristos

Note: From triangle similarity

Camera(s)

K P

Composed as a�sequence of Scaling, Shearing, Translation

CS491/691 C. Papachristos

Camera(s)

True Pinhole

Lens (Fisheye)

Aperture

Aperture

CS491/691 C. Papachristos

Range Sensor(s)

Sonar:

​

​

Laser Range-Finder:

​

​

​

Time-of-Flight Camera:

​

​

​

​

Structured Light:

​

​

​

CS491/691 C. Papachristos

Range Sensor(s)

Time-of-Flight Ranging:

​

Important points:

• Propagation speed of sound: 0.3 m/ms
• Propagation speed of electromagnetic signals: 0.3 m/ns (1 million times faster)
• 3 meters

Equivalent to 10 ms for an ultrasonic system

• Equivalent to only 10 ns for a laser range sensor

​

Measuring time-of-flight with electromagnetic signals is not an easy task

• Laser range sensors expensive and delicate

​

Quality of time-of-flight range sensors mainly depends on:

• Inaccuracies in the time of fight measurement (laser range sensors)
• Opening angle of transmitted beam (especially ultrasonic range sensors)
• Interaction with the target (surface, specular reflections)
• Variation of propagation speed (sound)
• Speed of mobile robot and target (if not at stand still)

CS491/691 C. Papachristos

Range Sensor(s)

Ultrasonic Range Sensor (Sonar):

​

Ultrasonic pulse is generated by a piezoelectric emitter, reflected by an object in its path, and sensed by a piezo-electric receiver

• Based on the speed of sound in air and the elapsed time from emission to reception, the distance between the sensor and the object is calculated

​

• Precision influenced by angle to object
• Soft surfaces can absorb most of the sound energy
• Surfaces that are far from perpendicular to the direction of sound - specular reflections

​

• Useful in ranges from several cm to several meters
• Relatively inexpensive
• Applications
• Distance measurement to (works with transparent surfaces)
• Collision detection

CS491/691 C. Papachristos

Range Sensor(s)

Laser Range Sensor – LIght Detection And Ranging (LIDAR):

​

Use Pulse measurement time-shift or Phase-shift measurement to derive Time-of-Flight

CS491/691 C. Papachristos

Range Sensor(s)

Laser Range Sensor – LIght Detection And Ranging (LIDAR):

​

Laser Triangulation – Use principles of Projective geometry

CS491/691 C. Papachristos

Range Sensor(s)

Structured Light Ranging:

​

Eliminates the correspondence problem by actively projecting structured light onto the scene

​

• Slits of light or emit collimated light (IR, Laser)
• Light picked up by properly “attuned” camera

​

• Range determined via the assumed light geometry:

​

​

​

​

​

​

• Suffer in environments with intense own light content
• E.g. IR projection based devices in settings with intense sunlight presence

IR Projector

IR Camera

RGB Camera

CS491/691 C. Papachristos

Time for Questions !

CS-491/691

CS491/691 C. Papachristos