1 / 23

Philip L. Winters, Center for Urban Transportation Research at USF

Philip L. Winters, Center for Urban Transportation Research at USF. Problem. Past vehicle-based GPS tracking give low-resolution view of daily travel behavior Are these GPS fixes: Points-of-interest? Stops in traffic? Difficult to extract info: Distance traveled Origin-Destination pairs

raine
Download Presentation

Philip L. Winters, Center for Urban Transportation Research at USF

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Philip L. Winters, Center for Urban Transportation Research at USF

  2. Problem • Past vehicle-based GPS tracking give low-resolution view of daily travel behavior • Are these GPS fixes: • Points-of-interest? • Stops in traffic? • Difficult to extract info: • Distance traveled • Origin-Destination pairs • Misses non-vehicle trips

  3. Innovation • USF’s TRAC-IT can capture “high-definition” view of travel behavior • Much easier to determine: • Path, distance traveled • Origin-Destination pairs • Avg. speeds • Can capture transit/bike/walk trips Sprint CDMA EV-DO Rev. A network

  4. New Problem • We can record GPS fixes as frequently as once per second and send to our server • However, frequent GPS fixes come at great cost to: • battery energy • data transfer over network • Both battery life and cell network data transfer are very limited resources

  5. One-day Requirement Sprint CDMA EV-DO Rev. A network Sprint CDMA EV-DO Rev. A network

  6. Let’s decrease the GPS recalculation rate when stationary!

  7. What is “Stationary”?Detecting User Movement 4 second GPS sampling 5 minute GPS sampling • GPS noise causes uncertainty in states • Many false transitions waste battery energy

  8. 4 second GPS sampling 5 minute GPS sampling Auto-Sleep to Reduce Energy Consumption Dynamicallychange the GPS sampling interval on the phone US Patent 8,036,679 October 11, 2011

  9. Evaluation – Summary of 30 tests • Approx. 88% mean accuracy in state tracking • Avg. doubling of battery life (based on TRAC-IT tests)

  10. Using TRAC-IT to Assess Variable Pricing Impacts on Carshare User Behavior

  11. Case Study - Carsharing Summary Results • Provided flip-phones for test and control subjects • Carried phone for all trips (passive data collection) • Varied hourly price in peak to shift time of rentals • Provided daily summary and map of trips via email • Collected data for two 3-week data collection periods; data instantly transmitted to us

  12. Lessons Learned Pluses Minuses Need to carry a second phone/charger Providing cell phones and data plans More work needed to differentiate “points of interest” from stuck in traffic when passively collecting data A current research focus • Providing phone with data only capabilities rather than software reduced need to test on multiple platforms and provided additional privacy protection • Continuous tracking while moving without running out of battery energy • Passive collection with free-text self-validation worked well with extended period of data collection • Phone instantly provides data to identify problems quickly • Virtually limitless length of field deployment

  13. Contact Info Philip L. Winters Director, TDM Program Center for Urban Transportation Research University of South Florida winters@cutr.usf.edu 813.974.9811

  14. “Asleep” Sanyo Pro 200 Sprint CDMA EV-DO Rev. A network “Awake”

  15. Evaluation – Daily Tracking Impact of GPS Auto-Sleep on Battery Life Sanyo Pro 200 Sprint CDMA EV-DO Rev. A network

  16. Patent issued on Oct. 11, 2011 US Patent # 8,036,679 Optimizing Performance of Location-Aware Applications using State Machines

  17. Utility & Commercialization • We have been tracking high-def travel behavior of over 30 participants over 9 months • Using TRAC-IT mobile app w/ GPS Auto-Sleep • USDOT-funded Value Pricing project • DAJUTA, a Florida-based company, has non-exclusively licensed the technology from USF • Other companies are also expressing interest • GPS Auto-Sleep is one module in “Location-Aware Information Systems Client (LAISYC)” framework • 15 patents pending on other modules

  18. New Idea • What if we could dynamically change the GPS sampling interval on the phone? • Use four second sampling interval when moving • Use five minute interval when stopped • Challenges: • Can we create mobile apps that do this? • Would this be enough to make a difference in battery life? • How do we handle GPS noise when trying to detect movement?

  19. Detecting User Movement 4 second GPS sampling 5 minute GPS sampling • What if we represent this binary, or two-state, problem more like a continuum?

  20. Gradually change GPS interval from “awake” to “asleep” based on certainty in user’s movement With some data (e.g., very high speeds) snap back to “awake” New invention – GPS Auto-Sleep “AWAKE” “ASLEEP” 22

  21. Sanyo Pro 200 Sprint CDMA EV-DO Rev. A network

More Related