Where are you - PowerPoint PPT Presentation

1 / 31
About This Presentation
Title:

Where are you

Description:

University of Wyoming, Computer Science Department, Laramie, Wyoming, ... William M. Spears, Jerry C. ... Oscilloscope snapshot. RF arrival time. Sound ... – PowerPoint PPT presentation

Number of Views:74
Avg rating:3.0/5.0
Slides: 32
Provided by: swarmro
Category:

less

Transcript and Presenter's Notes

Title: Where are you


1
University of WyomingDepartment of Computer
ScienceDistributed Robotics LaboratoryLaramie,
Wyoming, USA
  • Where are you?
  • Paul M. Maxim
  • William M. Spears, Jerry C. Hamann, Thomas
    Kunkel,
  • Rodney Heil, Dimitri V. Zarzhitsky, Diana F.
    Spears, Christer Karlsson

Project sponsored in part bythe Joint Robotics
Program and NSF
2
Outline
Standalone Trilateration Module
Highly portable I2C based modular concept
Accurate without s/w noise filtering
Fast improved equations
Coupled with Communication
Allows asynchronous movement
Tackle a wide range of problems
3
Problem
  • Ability for robots to interact with each other
  • chemical/biological plume tracing,
  • distributed sensing grids,
  • obstacle avoidance,
  • map generation,
  • mine clearing.

Old UW-DRL Robot Design
4
Solution
  • Swarm robot technology equipped with a
    localization system
  • Use trilateration method for localization system

New UW-DRL Robot Design
5
Trilateration
  • Uses one RF transceiver and three acoustic
    transducers to determine relative position of
    other robots

Thunder and lightning
6
Related work
  • Robots serve as beacons
  • Millibots
  • Navarro-Serment et al. - 2002,
  • Shapebugs
  • Cheng et al. 2005
  • Inline acoustic transducers MacArthur - 2003

Prior work does collaborative trilateration
Our localization system inexpensive, fast,
accurate, and self-contained.
7
Outline
Standalone Trilateration Module
Highly portable I2C based modular concept
Accurate without s/w noise filtering
Fast improved equations
Coupled with Communication
Allows asynchronous movement
Tackle a wide range of problems
8
Trilateration methods
d
d
Equilateral triangle pattern
XY coordinate system pattern
9
Equilateral triangle pattern
trilateration equations
our goal k 1
d
10
XY coordinate system pattern
well-conditioned matrix
d
11
Outline
Standalone Trilateration Module
Highly portable I2C based modular concept
Accurate without s/w noise filtering
Fast improved equations
Coupled with Communication
Allows asynchronous movement
Tackle a wide range of problems
12
Synchronization protocol
  • every 300 ms one robot is transmitting
  • additional sensor information is transmitted

Robot 3 receives
13
Outline
Standalone Trilateration Module
Highly portable I2C based modular concept
Accurate without s/w noise filtering
Fast improved equations
Coupled with Communication
Allows asynchronous movement
Tackle a wide range of problems
14
The Essence of Trilateration
  • Recall - must be able to detect an RF pulse and
    three acoustic signals.

15
Oscilloscope snapshot
RF arrival time
Sound arrival time
C
B
A
16
Trilateration challenges
  • High sensitivity
  • (signal amplification signal filtering)
  • High accuracy
  • (requires hardware and software solution)
  • Solution need to build new sonic range finder.

17
Experimental XSRF schematic
18
X Sonic Range Finder - XSRF
19
Complete Trilateration Module
  • Three XSRF sensor boards,
  • Three ultrasonic transducers,
  • Three parabolic cones,
  • Wireless RF transceiver.

20
Channelling acoustic energyinto a plane
Acoustic transducers and parabolic cones
21
Trilateration module
A
C
B
22
Previous Accuracy Resultsaverage error 4.8
inches (12.2 cm)
23
Current Accuracy Resultsaverage error 0.6
inches (1.5 cm)
24
The Maxelbot
Trilateration Module
25
HCS12 Controls Trilat
Digital Compass
HCS12 Controls Robot
26
(No Transcript)
27
Outline
Standalone Trilateration Module
Highly portable I2C based modular concept
Accurate without s/w noise filtering
Fast improved equations
Coupled with Communication
Allows asynchronous movement
Tackle a wide range of problems
28
Chemical Plume Tracing
  • Objective Identify a chemical emission and trace
    the chemical plume to its source emitter.
  • Requires at least two chemical sensors to find
    the chemical gradient.
  • Significantly better results when use multiple
    collaborating robots.

29
Experiments and demonstrations
  • three Maxelbots pulling a box,gt
  • three Maxelbots using AP to self organize and
    move towards a goal,gt
  • three Maxelbots using AP to self-repair and move
    towards a goal,gt
  • three Maxelbot CPT test run.gt

30
Summary
Standalone Trilateration Module
Highly portable I2C based modular concept
Accurate without s/w noise filtering
Fast improved equations
Coupled with Communication
Allows asynchronous movement
Tackle a wide range of problems
31
  • Questions
Write a Comment
User Comments (0)
About PowerShow.com