Test Slide

1
Test Slide

• Text works.
• Graphics work.

2
Self-MappingMobile Robot

• Senior Capstone Project
• Department of Electrical
• and Computer Engineering
• Bradley University
• Advisor Dr. A. Malinowski
• Presented by Stephanie Luft
• 27 April 2006

3
Presentation Outline

• Project Overview
• System Block Diagram
• Functional Description
• Design Process
• Conclusion
• Questions

4
Presentation Outline

• Project Overview
• System Block Diagram
• Functional Description
• Design Process
• Conclusion
• Questions

5
Project Overview

• Objective To develop a robot that will
• Map an area of its environment
• Locate itself within the map
• Orient itself within the environment

6
Applications

• Military Robot (PackBot)
• Household Robot (Roomba)
• Moon or Mars Rover

7
Review of Previous Work

• GuideBot Capstone Project 2005
  by John Hathway and Dan Leach
• Laser Meter From Drexel University
• Thesis Concurrent Map Building and
  Self-Localization for Mobile Robot Navigation

8
Presentation Outline

• Project Overview
• System Block Diagram
• Functional Description
• Design Process
• Conclusion
• Questions

9
System Block Diagram
10
MapBot
11
MapBot
User Commands
Robot Platform
12
MapBot
Laser Distance Meter
13
MapBot
Map, Robot Location
Audio Warning
Robot Movement
14
Presentation Outline

• Project Overview
• System Block Diagram
• Functional Description
• Design Process
• Conclusion
• Questions

15
Software Functionality

• MATLAB
• Server/C
• Functional Modes
• Mapping
• Maneuvering

16
Software Mapping Mode

• Plot environment and locate robot
• Distance sensing
• Immediate response
• Data transmission
• Plotting obstacles and robot location
• Self-locating
• Previous map identification
• Navigation and maneuvering

17
Software Mapping Mode

• Plot environment and locate robot
• Distance sensing
• Immediate response
• Data transmission
• Plotting obstacles and robot location
• Self-locating
• Previous map identification
• Navigation and maneuvering

18
Software Maneuvering Mode

• Allow user to control robot
• User interface
• Data transmission
• Immediate response
• Navigation and maneuvering
• Distance sensing
• Self-locating
• Plotting robot location only

19
Software Maneuvering Mode

• Allow user to control robot
• User interface
• Data transmission
• Immediate response
• Navigation and maneuvering
• Distance sensing
• Self-locating
• Plotting robot location only

20
Presentation Outline

• Project Overview
• System Description
• Design Process
• Conclusion
• Questions

21
Main Program Flowchart
Start
Take Mapping Readings with Laser Distance Meter
Plot Obstacle and Robot Location Probabilities
on Map
Compare Current Map to Previous Maps
Maneuver as Needed
22
Design Theory Distance Meter
Target
Camera
Focal Plane
pfc
q
h
LASER
D
23
Design Theory Distance Meter

• Guiding Equation D h
• 
  tan(pfcmb)
• D distance in meters
• h distance (meters) between laser and center of
  lens
• pfc pixels from center of image
• m calibration coefficient
• b calibration offset

24
Design Theory Distance Meter

• Results of calibration data
• h 0.189 m
• m 0.00113923
• b -0.0324705
• Final Equation D 0.189
• 
  pfc.00113923 0.0324705

25
Design Theory Mapping

• Probabilistic Algorithm
• Pixel value range 0 to 1 (white to black)
• Initial pixel value .25 (light gray)
• Obstacle .25
• Empty space -.25

.25 X .25 .25 .25 .25
.25 .25 .25 .25 .25 .25
.25 .25 .25 .25 .25 W
.25 .25 .25 .25 .25 A
.25 .25 .25 .25 .25 L
R .25 .25 .25 .25 L
26
Design Theory Mapping

• Probabilistic Algorithm
• Pixel value range 0 to 1 (white to black)
• Initial pixel value .25 (light gray)
• Obstacle .25
• Empty space -.25

.25 X .25 .25 .25 .25
.25 .25 .25 .25 .25 .25
.25 .25 .25 .25 .25 W
.25 .25 .25 .25 .25 A
.25 .25 .25 .25 .25 L
R .25 .25 .25 .25 L
27
Design Theory Mapping

• Probabilistic Algorithm
• Pixel value range 0 to 1 (white to black)
• Initial pixel value .25 (light gray)
• Obstacle .25
• Empty space -.25

.25 X .25 .25 .25 .5
.25 0 .25 .25 0 .25
.25 0 .25 0 .25 W
0 .25 0 .25 .25 A
0 0 .25 .25 .25 L
R .25 .25 .25 .25 L
28
Design Theory Mapping

• Probabilistic Algorithm
• Pixel value range 0 to 1 (white to black)
• Initial pixel value .25 (light gray)
• Obstacle .25
• Empty space -.25

.25 -- .25 .25 .25 .5
.25 0 .25 .25 0 .25
.25 0 .25 0 .25 W
0 .25 0 .25 .25 A
0 0 .25 .25 .25 L
R .25 .25 .25 .25 L
29
Design Theory Mapping

• Probabilistic Algorithm
• Pixel value range 0 to 1 (white to black)
• Initial pixel value .25 (light gray)
• Obstacle .25
• Empty space -.25

.25 .25 .25 .25 .25 .75
.25 0 .25 .25 0 .25
.25 0 .25 0 .25 W
0 .25 0 .25 .25 A
0 0 .25 .25 .25 L
R .25 .25 .25 .25 L
30
Design Simulation Mapping
Shape of room

• Simulating the mapping algorithm using imaginary
  data
• This demonstrates that the mapping algorithm
  works for a simple case

31
Design Testing Mapping

• Actual Map taken at the Student Expo
• Mirror Effect
• Outliers

Actual Shape of Environment
32
Design Testing Mapping

• Actual Map taken at the Student Expo
• Mirror Effect
• Outliers

Actual Shape of Environment
33
Presentation Outline

• Project Overview
• System Description
• Design Process
• Conclusion
• Questions

34
Results Laser Distance Meter

• Laser Distance Meter Testing
• Good mid-distance fit
• Average 2 error
• Higher errors at ends
• 12 at edges of range
• Absolute Maximum Range 0.5 to 175 meters

35
Results Laser Distance Meter
Measured Data Calculated Data
36
Results

• Mapping
• Multiple mappings from same location
• PTU Control
• Complete (Thank you, Dr. Malinowski!)
• Robot Movement

37
Future Work

• Web Control for Remote User
• Navigation
• Enhanced Capabilities
• Bright Sunlight
• Stairs
• Greater Distance Meter Accuracy
• Power Conservation

38
Results Complete Mapping Sequence

• Second Round Map Taken in the EE Student Lounge

Actual Shape of Environment
39
Self-MappingMobile Robot

• Department of Electrical
• and Computer Engineering
• Bradley University
• Advisor Dr. A. Malinowski
• Presented by Stephanie Luft
• 27 April 2006

Questions? sluft_at_bradley.edu
40
Self-MappingMobile Robot

• Website
• http//cegt201.bradley.edu/projects/proj2006/mapbo
  t

41
Review of Previous Work

• GuideBot Capstone Project 2005
• John Hathway and Daniel Leach
• Laser Meter From Drexel University
• http//www.pages.drexel.edu/twd25/webcam_laser_ra
  nger.html
• Mapping From Dartmouth University, 1999
• http//www.jonh.net/7Ejonh/robots/mapping/submitt
  ed-paper.html
• Thesis Concurrent Map Building and
  Self-Localization for Mobile Robot Navigation
• Thomas Duckett, Manchester, United Kingdom

42
Software Flowchart Basic Mapping, Part 1
Start Distance Meter
Capture Image from Webcam
Calculate Distance to Obstacle
Place Distance and f into Matrices for Mapping
Repeat for 320
43
Software Flowchart Basic Mapping, Part 2
Start Mapping
Initialize Variables and Create or Retrieve
Initial Map
Calculate Coordinates of Obstacles
Grow the map to accommodate new obstacles,
retaining previous map details
Plot obstacles on the map
Compare to Previous Maps and Adjust as Necessary