2011 gis in transit - cell phones and gis - lessons learned from developing transit navigation...
DESCRIPTION
Discussion of various lessons learned while developing the first real-time public transit navigation app to help transit riders with intellectual disabilities.TRANSCRIPT
Patents Pending USF 2009
Sean J. BarbeauResearch AssociateCenter for Urban Transportation Research
(CUTR)University of South Florida (USF)
Mark SheppardTravel TrainerHillsborough Area Regional Transit (HART)
1
Cell Phones and GIS: Lessons Learned from
Developing Transit Navigation Software
Work to HomeWork to Home
Back
Distance to Final Stop:5.6 miles
5 DOWNTOWNNext Bus:
9:58 minutes
Patents Pending USF 2009Patents Pending USF 2009
Overview
Role of travel training
Travel Assistance Device
Challenges and Solutions
Next steps
04/13/2311/17/2008
2
Patents Pending USF 2009Patents Pending USF 2009
Project Partners3
Sean Barbeau, Phil Winters, and Nevine Georggi, CUTR, USF
Miguel Labrador and Rafael Perez, Computer Science & Engineering, USF
Mark Sheppard, HART Travel TrainerGigi Gonzalez, Special Education Facilitator
for STAGES program at USFAmy Datz, FDOT Project ManagerHarvey Berlin, TRB IDEA Project Manager
11/17/2008
Patents Pending USF 2009Patents Pending USF 2009
Travel Training 4
. . . the Challenges . . . the Victories
11/17/2008
Patents Pending USF 2009Patents Pending USF 2009
The Challenges . . . 5
Transportation is essential for independent living
Individuals with mental/cognitive disabilities (14.2M Americans, 6.9% of pop.)¹ often have problems with quick actions required by transit
Paratransit is expensive $27.90 per unlinked passenger trip (bus) versus $3.20 per unlinked passenger trip (demand responsive) and can be restrictive to riders
Travel training helps reduce learning curve for fixed route transit1 - National Institute on Disability and Rehabilitation Research. “Survey of Income and Program Participation
(SIPP)”, 1997.
11/17/2008
Patents Pending USF 2009Patents Pending USF 2009
A Solution . . .TAD6
Develop first navigation software for public transportation using GPS-enabled mobile phones Alert user when to get off the bus with audio, visual,
and tactile prompts Target simplicity, with cognitively disabled in mind
11/17/2008
Patents Pending USF 2009Patents Pending USF 2009
TAD Web Page – Create Trips7
11/17/2008
Patents Pending USF 2009Patents Pending USF 2009
Transit Rider Selects Trip That Was Planned On Website
TADTAD
Cancel Select
Select Trip
(1) Home to Work(2) Work to Home(3) Home to Movie
TAD Cell Phone Application
Patents Pending USF 2009Patents Pending USF 2009
On Bus…
Then the user hears: “Get Ready!”
Work to HomeWork to Home
Back
Distance to Final Stop:5.6 miles
5 DOWNTOWN
TAD Cell Phone Application
Patents Pending USF 2009Patents Pending USF 2009
Then: “Pull the Cord Now!”
TADTAD
OK
Pull the Cord Now!
(+Sound and Vibration)
TAD Cell Phone Application
Patents Pending USF 2009Patents Pending USF 2009
Field Tests11
Over 50 test runs in Tampa: Over 38 by testing team 12 trips with riders with special needs
Overall, proof-of-concept was successfulMost consistent issue in tests:
Error in bus stop locations Problems with close proximity of stops
11/17/2008
Alert Timing
Number of Occurrences
Ideal 39
Early 2
Late 6
None 4
Patents Pending USF 2009
12
Challenge: GPS is very dynamic…
11/17/2008
Patents Pending USF 2009
Challenge: Updating Route Information
Google Transit provides free trip planning tool to agencies
Agency has incentive to post schedule updates to a webpage so Google can update their system
TAD system can grab the same updates and use them!
This feature also allows adding new agencies to TAD with the click of a button!
13
11/17/2008
Patents Pending USF 2009Patents Pending USF 2009
Challenge: Bus Stop Accuracy14
11/17/2008
• Errors in bus stop inventory location vs. true geographic location
Patents Pending USF 2009Patents Pending USF 2009
Lessons Learned15
First Bus Stop Detection algorithm, based on radius surrounding final stop
11/17/2008
2nd-to-last Stop
Destination Stop
Alert Radius
Patents Pending USF 2009Patents Pending USF 2009
Lessons Learned16
To avoid delivering alert late, radius should be increased
11/17/2008
2nd-to-last Stop
Destination Stop
Patents Pending USF 2009Patents Pending USF 2009
Lessons Learned17
With GPS, uncertainty of true position exists
11/17/2008
2nd-to-last Stop
Destination Stop
Uncertainty Radius
Possible True Position
Calculated Position
Patents Pending USF 2009Patents Pending USF 2009
Lessons Learned18
Problems arise when alert radius and uncertainty overlap when bus stops are close
11/17/2008
2nd-to-last Stop
Destination Stop
Patents Pending USF 2009Patents Pending USF 2009
Lessons Learned19
When alert radius is big, can cause unpredictable results for certain route configurations
11/17/2008
2nd-to-last Stop
Destination Stop
XEarly Alert
Patents Pending USF 2009Patents Pending USF 2009
Lessons Learned20
New Algorithm based on circles surorunding 2nd-to-last stop
11/17/2008
2nd-to-last Stop
Destination Stop
Patents Pending USF 2009Patents Pending USF 2009
Detects user passing 2nd-to-last stop
Lessons Learned21
11/17/2008
2nd-to-last Stop
Destination Stop
Trigger Alert!
Patents Pending USF 2009Patents Pending USF 2009
Lessons Learned
22
Create Bus Stop Management Tool
11/17/2008
Patents Pending USF 2009Patents Pending USF 2009
Lessons Learned23
Human Factors – Investigate Bluetooth Headset Privacy for Individual Easier to hear over noise on bus
11/17/2008
Patents Pending USF 2009Patents Pending USF 2009
TAD with AVL24
Ideas Deserving Exploratory Analysis (IDEA) project to link TAD with AVL system Provides new services based on real-time bus
location
11/17/2008
Work to HomeWork to Home
Back
Distance to Final Stop:5.6 miles
5 DOWNTOWNNext Bus:
9:58 minutes
Patents Pending USF 2009Patents Pending USF 2009
Next Steps25
Testing with HART in Tampa Bay area has been successful
Expanding TAD deployment opportunities with other transit agencies
11/17/2008
Patents Pending USF 2009Patents Pending USF 2009
New TAD Partners
Partners will be asked to: Format their data into Google Transit format
(HASTUS and Trapeze both provide export tool)
Test TAD by their travel trainer as part of daily trips with trainees
Correct errors in any bus stop locations through TAD webpage
Provide Feedback to TAD research team
Patents Pending USF 2009Patents Pending USF 2009
New TAD Partners
CUTR will: Provide adequate TAD
orientation/demo/training sessions Help transit agency with TAD integration Monitor TAD and provide support during tests Resolve any TAD technical issues as they occur
Patents Pending USF 2009
SEAN BARBEAURESEARCH ASSOCIATE(813) [email protected]
MARK SHEPPARDTRAVEL TRAINER(813) [email protected]
28
Questions?
11/17/2008
Please contact Sean Barbeau if you’re interested in TAD Deployment in your city!