1. Calling all cars: cell phone networks and the future of traffic Presentation by Scott Corey Article written by Haomiao Huang

2. The Future of Cars  Self-driving cars?  Boosting the brainpower of the environment cars drive in  Traffic monitoring has been revolutionized

3. An intelligent highway  Reducing the effect of traffic jams and accidents  Traffic control schemes to react to real time data  Aid in planning for the future

4. Sensors  Monitor traffic  Parking availability  Air pollution  Have traditionally been static sensors  Inductive Loop Detectors  Traffic Cameras  RFID tags

5. Problems  Expensive to deploy, operate, repair  Placed only at key locations  Mobile sensors are a necessity

6. Mobile Phones  Equipped with GPS and Internet access  Smartphones enable more widespread source of data  Worldwide, there are more cell phones in use than toothbrushes

7. Mobile Millennium  One of the first large-scale phone-based traffic monitoring projects in the US  Run by Nokia, NAVTEQ, and UC Berkeley

8. Gathering data, but privately  User privacy is key for user acceptance  Two main needs:  Preventing the path of a vehicle to be reconstructed  Separating the identification of the phone from the data

9. Anonymity  Data from phones is tagged with user information  The data packet is encrypted at transmission  Proxy server cannot decrypt packet, but can strip identifying information  Sent to traffic servers after information stripped

10. Reconstructing paths  Uses virtual trip lines instead of constant reporting  VTL spacing varies based on speed to maximize number of cars  Randomizing measurements

12. Making sense of it all  UC Berkeley tasked to fuse all the data together  GPS from phones  GPS data from dedicated vehicles  Static sensors  Given all of the measurements being gathered and a stretch of road of interest, what is the best estimate of the number of cars on that road, and how fast they're going?

13. Combining data with maps  GPS tracks are useless alone – need to combine with maps to know what road network you are monitoring  Measurements have to use machine- learning methods to correct for people walking with phones, parked cars

14. The flow of traffic  Tracking thousands of cars individually is difficult and expensive  Traffic researchers treat movement of cars as liquid flowing through tubes

15. Fluid Dynamics  Requires initial conditions and rate of cars entering/leaving roadway  Fluid dynamics model works well with fixed sensors  Cameras can determine initial conditions  Sensors attached to on and off ramps

16. Disruptions  Drivers are not perfect  Accidents  Unnecessary slow-downs  Adding GPS dramatically increases the versatility of the fluid model  GPS incorporated as internal conditions for the flow to satisfy

17. Mobile Century  Proof of concept test  100 cars with mobile phones mixed into traffic  Ran for 10 hours with 150 student drivers  Despite accounting for 2-5% of cars on the highway, speed and density of cars measured at a high resolution  Accident was detected and reported in less than a minute

18. Till all are one  Concepts and technology are now widespread  Mobile sensors used to identify potholes in roads  Connections to vehicle sensors  Mobile sensing is the future