Autonomous Mobile Robots, Chapter 4 © R. Siegwart, I. Nourbakhsh with Skubic augmentations Perception l Sensors l Uncertainty l Features 4 PerceptionMotion.

Autonomous Mobile Robots, Chapter 4

© R. Siegwart, I. Nourbakhsh with Skubic augmentations

Perception

Sensors Uncertainty Features

4

Perception Motion Control

Cognition

Real WorldEnvironment

Localization

PathEnvironment ModelLocal Map

"Position" Global Map



Example HelpMate, Transition Research Corp.

4.1



Example B21, Real World Interface

4.1



Example Robart II, H.R. Everett

4.1



Savannah, River Site Nuclear Surveillance Robot

4.1



BibaBot, BlueBotics SA, Switzerland

Pan-Tilt Camera

Omnidirectional Camera

IMUInertial Measurement Unit

Sonar Sensors

Laser Range Scanner

Bumper

Emergency Stop Button

Wheel Encoders

4.1



Our new robot: Killianunder development

gripper with sensors: gripper with sensors: IR rangefindersIR rangefinders

strain gaugestrain gauge

top sonar ringtop sonar ring

bottom sonar ringbottom sonar ring

laser range-finderlaser range-finder

stereo visionstereo vision

laptop brainlaptop brain



General Classification (Table 4.1)

4.1.1



General Classification (Table 4.1, cont.)

4.1.1



Sensor Terminology

Sensitivity Dynamic Range Resolution Bandwidth Linearity Error Accuracy Precision Systematic Errors Random Errors



Active Ranging Sensors : Ultrasonic sensor

4.1.6

transmit a packet of (ultrasonic) pressure waves distance d of the echoing object can be calculated based on the

propagation speed of sound c and the time of flight t.

The speed of sound c (340 m/s) in air is given by

where

: ration of specific heats

R: gas constant

T: temperature in degree Kelvin

TRc ..

2

* tcd

4.1.6



Ultrasonic Sensor (time of flight, sound)

Transmitted sound

Analog echo signal

Threshold

Digital echo signal

Integrated time

Output signal

integrator Time of flight (sensor output)

threshold

Wave packet

Effective range: typically 12 cm to 5 m

4.1.6



Ultrasonic Sensor (time of flight, sound)

typically a frequency: 40 - 180 kHz generation of sound wave: piezo transducer

transmitter and receiver separated or not separated sound beam propagates in a cone like manner

opening angles around 20 to 40 degrees regions of constant depth segments of an arc (sphere for 3D)

Typical intensity distribution of a ultrasonic sensor

-30°

-60°

0°

30°

60°

Amplitude [dB]

measurement cone

4.1.6



SRF10 sensor

Range: 3 cm to 6 m See also www.acroname.com



SRF10 Characteristics

SONAR VALUES

0

200

400

600

800

1000

1200

0 20 40 60 80 100 120 140 160

Distance measured (in cm)

sen

sor

read

ing group1

group2

group3

group4

group5

group6



SRF10 Characteristics (previous years)

Sonar Rangefinder

0

500

1000

1500

2000

2500

3000

0 50 100 150 200 250 300

Range (cm)

Sen

sor

Rea

din

g

Person

Cardboard

Metal

Wall

Legos



Ultrasonic Sensor Problems

Soft surfaces that absorb most of the sound energy Undesired from non-perpendicular surfaces

Specular reflection

Foreshortening

Cross-talk between sensors

4.1.6

What if the robot is moving or the sensor is moving (on a servo motor)? What if another robot with the same sensor is nearby?



Optical Triangulation (1D)

Principle of 1D triangulation.

distance is proportional to 1/x

Target

D

L

Laser / Collimated beam

Transmitted Beam

Reflected Beam

P

Position-Sensitive Device (PSD)or Linear Camera

x

Lens

x

LfD

x

LfD

4.1.6

http://www.acroname.com/robotics/parts/SharpGP2D12-15.pdf



Sharp Optical Rangefinder (aka ET sensor)

ET sensor values

0

20

40

60

80

100

120

140

160

0 20 40 60 80 100 120 140 160

Distance (in cm)

sen

sor

Rea

din

g group4

group6

group1

group2

group3

group5



Sharp Optical Rangefinder (previous years)

IR Optical Rangefinder

0

20

40

60

80

100

120

140

160

0 10 20 30 40 50 60 70 80

Range (cm)

Sen

sor

Rea

din

g Black

Blue

Red

Green



IR Sensor (aka Top Hat sensor)

TopHat sensor

150

170

190

210

230

250

270

0 1 2 3 4 5 6 7

Group #

Sen

sor

Rea

din

g

Cardboard

Metal

Black Surface

White Paper

Wall

Used for:Used for:Line followingLine followingBarcode readerBarcode readerEncoderEncoder



Ground-Based Active and Passive Beacons

Elegant way to solve the localization problem in mobile robotics Beacons are signaling guiding devices with a precisely known position Beacon base navigation is used since the humans started to travel

Natural beacons (landmarks) like stars, mountains or the sun Artificial beacons like lighthouses

The recently introduced Global Positioning System (GPS) revolutionized modern navigation technology

Already one of the key sensors for outdoor mobile robotics For indoor robots GPS is not applicable,

Major drawback with the use of beacons in indoor: Beacons require changes in the environment

-> costly. Limit flexibility and adaptability to changing

environments.

4.1.5



Global Positioning System (GPS)

Developed for military use Recently it became accessible for commercial applications 24 satellites (including three spares) orbiting the earth every 12 hours at a

height of 20.190 km. Four satellites are located in each of six planes inclined 55 degrees with respect

to the plane of the earth’s equators Location of any GPS receiver is determined through a time of flight

measurement

Technical challenges: Time synchronization between the individual satellites and the GPS receiver Real time update of the exact location of the satellites Precise measurement of the time of flight Interferences with other signals

4.1.5



Global Positioning System (GPS)

4.1.5

Satellites synchronize transmissions of Satellites synchronize transmissions of location & current timelocation & current time

GPS receiver is GPS receiver is passivepassive

4 satellites provide (x,y,z) and 4 satellites provide (x,y,z) and time correctiontime correction



Laser Range Sensor (time of flight, electromagnetic) (1)

Transmitted and received beams coaxial Transmitter illuminates a target with a collimated beam Receiver detects the time needed for round-trip A mechanical mechanism with a mirror sweeps

2 or 3D measurement

PhaseMeasurement

Target

D

L

Transmitter

Transmitted BeamReflected Beam

P

4.1.6




Time of flight measurement Pulsed laser

measurement of elapsed time directly resolving picoseconds

Beat frequency between a frequency modulated continuous wave and its received reflection

Phase shift measurement to produce range estimation technically easier than the above two methods.

4.1.6




Phase-Shift Measurement

Wherec: is the speed of light; f is the modulating frequency; D’ is the total distance covered by the emitted light

for f = 5 Mhz (as in the A.T&T. sensor), = 60 meters

PhaseMeasurement

Target

D

L

Transmitter


P

2

2 LDLD = c/f

4.1.6




Distance D, between the beam splitter and the target

where : phase difference between the transmitted and reflected light beams

Theoretically ambiguous range estimates since for example if = 60 meters, a target at a range of 5 meters = target at

65 meters

4D


0

Phase [m]

Amplitude [V]

(2.33)

4.1.6




Confidence in the range (phase estimate) is inversely proportional to the square of the received signal amplitude.

Hence dark, distant objects will not produce such good range estimated as closer brighter objects …

4.1.6



Laser Range Sensor (time of flight, electromagnetic)

Typical range image of a 2D laser range sensor with a rotating mirror. The length of the lines through the measurement points indicate the uncertainties.

4.1.6



Vision-based Sensors: Sensing

Visual Range Sensors Depth from focus Stereo vision

Motion and Optical Flow

Color Tracking Sensors

4.1.8



Vision-based Sensors: Hardware

CCD (light-sensitive, discharging capacitors of 5 to 25 micron)

CMOS (Complementary Metal Oxide Semiconductor technology)

2048 x 2048 CCD array

Cannon IXUS 300

Sony DFW-X700

Orangemicro iBOT Firewire

4.1.8



Color Tracking Sensors

Motion estimation of ball and robot for soccer playing using color tracking

4.1.8



Robot Formations using Color Tracking



Image representation

(1,1)(1,1)

(640,480)(640,480)

R = (R = (255255,,00,,00))

G = (G = (00,,255255,,00))

B = (B = (00,,00,,255255))

Yellow = (Yellow = (255255,,255255,,00))

Magenta = (Magenta = (255255,,00,,255255))

Cyan = (Cyan = (00,,255255,,255255))

White = (White = (255255,,255255,,255255))



Image Representation

YCrCb YCrCb illumination data stored in a separate channelillumination data stored in a separate channel(may be more resistant to illumination changes)(may be more resistant to illumination changes)

R-G-B channels map to Cr-Y-CbR-G-B channels map to Cr-Y-CbwherewhereY = 0.59G + 0.31R + 0.11B (illumination)Y = 0.59G + 0.31R + 0.11B (illumination)Cr = R-Y Cr = R-Y Cb = B-YCb = B-Y



CMU cam

Ubicom SX28 microcontroller with 136 byes SRAM 8-bit RGB or YCrCb Max resolution: 352 x 288 pixels Resolution is limited to 80 horizontal pixels x 143 vertical pixels

because of the line by every other line processing.

(1,1)(1,1)

(352,288)(352,288) (80,143)(80,143)



CMU cam Operation

init_camera() auto-gain – adjusts the brightness level of the image white balance adjusts the gains of the color channels to accommodate

for non-pure white ambient light clamp_camera_yuv()

point the camera at a white surface under your typical lighting conditions and wait about 15 seconds

trackRaw(rmin, rmax, gmin, gmax, bmin, bmax)

GUI interface for capturing images and checking colors



CMU cam Tracking

Global variablesGlobal variables•track_size … in pixelstrack_size … in pixels•track_xtrack_x•track_ytrack_y•track_area … area of the bounding boxtrack_area … area of the bounding box•track_confidencetrack_confidence

(1,1)(1,1)

(80,143)(80,143)



CMU cam – Better tracking

Auto-gain Adjusts the brightness level of the image

White balance Adjusts the color gains on a frame by frame basis Aims for an average color of gray Works great until a solid color fills the image

One strategy – use CrYCb Aim at the desired target and look at a dumped frame (in GUI)Aim at the desired target and look at a dumped frame (in GUI) Set the Cr and Cb bounds from the frame dumpSet the Cr and Cb bounds from the frame dump Set a very relaxed Y (illumination)Set a very relaxed Y (illumination)



Adaptive Human-Motion Tracking

4.1.8

Autonomous Mobile Robots, Chapter 4 © R. Siegwart, I. Nourbakhsh with Skubic augmentations Perception l Sensors l Uncertainty l Features 4 PerceptionMotion.

Documents

autonomous mobile robots

skubic augmentations

skubic augmentations

sound beam

generation of sound

propagation speed of

dynamic range

real world interface