presentation
Download
Report
Transcript presentation
GPS
Service Animal Collar
By
Chris Stoddard, Harrison Rose & Richard Lew
•
•
•
•
•
•
Overview
Introduction
Features
Design
Testing and Verifications
Future Work
Q&A
Introduction
• ECE 498: JJM Global Positioning
Systems
• Inspired by a guest speaker from
ANSC 250
Objective
To provide visually impaired
people a fast, safe, and reliable
method for exploring new points
of interest
Features
•
•
•
•
•
•
•
Turn-By-Turn Directions
Up to 5 locations can be recorded and stored
Location accuracy within 5 meters
Backtracking for all routes
Two vibrating motors for executing routes
2-hour battery life
Adjustable collar
Reviewing Our Original Design
Figure 1: Original Block Diagram
Curb Detection
• Originally planned as feature to avoid users from
leaving the sidewalk
• Solution was to use ultrasonic range finder to detect
curb edges
• After reviewing specifics, we decided that curb
detection was not achievable
-1
-1
• tan (X/Y) = 15 Y = 2 [ft]/ tan (15) = 7.4641 [ft]
• 65 [inches]/2.54 [cm/inch] = 25.59 [inches]
25.59*tan(15) = 6.8585 [inches] in front of the sensor.
Ethical Issues
• Accuracy and Precision
• Reliable Battery life
• Safe Hardware for Animal
Interaction
Determining Parameters
GPS:
1)
Ephemeris data: Errors in the transmitted location of the satellite (3 [m] on average)
2)
Satellite clock: Errors in the transmitted clock of the satellite (2 [m] on average)
3)
Ionosphere: Errors in the corrections of pseudorange due to ionospheric effects (5 [m] on
average)
4)
Troposphere: Errors in the corrections of pseudorange due to tropospheric effects (1 [m] on
average)
5)
Multipath: Errors caused by reflected signals entering the receiver antenna (1 [m] on average)
6)
Receiver: Errors associated with the receiver such as noise, software, and biases. (1 [m] on
average)
Average Errors in GPS Navigation Solution:
Average Error = [Ephemeris Error + Satellite Clock Error + Ionosphere Error + Troposphere Error +
Multipath Error + Receiver Error]
(Equation 1)
=
3 [m] + 2 [m] + 5 [m] + 1 [m] + 1 [m] + 1[m] = 13 [m]
Converting Coordinates
• Usually done in X, Y, Z
• Read in Lattitude, Longitude,
Altitude but needs to measure in
meters, not degrees
• 0.0026 minutes = 5 meters
Motor System
• Two motors
• Two different vibrations
– Turn (Left, Right)
– Stop (Both)
Original Turning System
• Originally, turns were going to be pre-computed after
the route was originally recorded
• However, we decided that this method was
cumbersome and a real-time decision would be better
Figure 2 and 3: X and Y coordinate Turn Errors with Noise
New Turning System
• Use the current heading reported by the GPS
• Calculate the next heading using the next two
waypoints
• Using these two vectors, calculate the angular
difference and use this to decide the turn direction
Route Recording
• To record a route, users press a button at each
waypoint
• Each of these locations is stored sequentially in
memory
• Supporting five routes, the memory holds five lists of
waypoints
Route Recording MatLab
Simulation
Figure 4: Screen Capture of Route Record
Backtracking
• Backtracking supports navigating an entire route
backwards, or returning from a route before
completion
• Order of waypoints pulled from memory controlled by
backtracking
Hardware Overview
Figure 5: Final Block Diagram
PCB Layout
Figure 6: PCB Schematic
Board Schematic
Figure 7: Board Schematic
Motor Schematic
Figure 8: Motor Schematic
Testing
•
•
•
•
•
MatLab Programs
iPhone GPS Application
Motor/Micro Interfacing
GPS/Micro Interfacing
Field Position Testing
Success: Micro and GPS Interact
Figure 9: Instamapper iPhone App Map
Debugging the Interface
Verifications
• Powering motor by itself
• Measuring output from micro
Challenge: Motors and
Micro Do Not Interact
• Change transistor
• Change base resistor
• Inverting output from micro
Success: Motors and
Micro finally Work
• Used the same power supply for
micro and motors
The Final Product
Testing
st
1
Test 1
Test 2
Test 3
Test 4
Test 5
Test 6
nd
2
rd
3
th
4
th
5
th
6
point point point point point point
2m
8m
2m
N/A
9m 5.7m
2m
N/A
1m 5.3m
4m 2.2m
4m
8m 6.7m 0.7m
2m
5m
3m
1m
2.1m 3.7m 0.8m 6.3m 4.7m 2.5m
Chart 1: Turn Error Tests
Average
4m
5.57m
3.125m
4.85m
2.75m
3.35m
Future Work
•
•
•
•
Increase precision and accuracy
Course Correction
Load routes from database
Reliable curb detection
Questions
Resources
“Blind people safety project”. Faith Degirmenci, Abdessettar Ibourki, Morad Oumina. Proposal, Spring 2006
GPS Instampper for Apple iPhone <http://www.instamapper.com>
“GPS Tracking Device with DGPS”. Alex Stezskal, Algirdas Navickas, Vivek Thyagarajan. Proposal, Spring 2010
Misra, Pratap, and Per Enge. Global Positioning System: Signals, Measurements, and Performance. Lincoln, MA: Ganga-Jamuna, 2011. Print.
"Sam Wormley's GPS Errors & Estimating Your Receiver's Accuracy." Educational Observatory Institute. Web. 08 Feb. 2012. <http://eduobservatory.org/gps/gps_accuracy.html>.
Skytraq datasheet for Skytraq Venus634FLPx. <http://www.sparkfun.com/datasheets/GPS/Modules/Skytraq-Venus634FLPx_DS_v051.pdf>
"Switch Debouncing (Bounce-free Switches) Using NAND Gates - Www.ECELab.com." Electronics and Communications Engineering. Web.
21 Feb. 2012. <http://www.ecelab.com/switch-debounce-NAND.htm>.
TI datasheet for MSP430F2274. < http://www.ti.com/product/msp430f2274>.
TI datasheet for SN7400N. <http://www.ti.com/lit/ds/symlink/sn7400.pdf>.
Special Thanks To…
•
•
•
•
Tom Galvin
Professor Scott Carney
Professor Jonathan Makela
Mark & Scott from the parts shop