Presenter: Alixandra Demers
Advisor(s): Dr. George F. List, P.E.
Author(s): Alixandra Demers
Graduate Program: Civil, Construction, and Environmental Engineering
Title: Capital District Advanced Traveler Information System Field Experiment
Abstract: Sitting in traffic is not why we drive, yet it is an unfortunate side effect when many of us decide to travel at the same time or to the same place. For years, transportation engineers have assisted travelersí wayfinding by use of static and dynamic (changeable) signage to indicate the best routes. More recently, assistance has been provided on websites (static: Mapquest, dynamic: Traffic.com), 511 phone systems, and through in-vehicle satellite navigation devices (such as CoPilot and Navteq, both static). Therefore, with the recent advances in Global Positioning System (GPS), wireless communication, and route guidance system technologies our research team went the next step to help travelers, creating an Advanced Traveler Information System (ATIS) in which the vehicles both acted as probes that collected data, and information receivers engaged in real-time route guidance (1st time in the world). The team designed a 3-month ATIS field experiment in which 200 route guidance-equipped vehicles shared information about network travel times so they could engage in real-time path-choice decision making. The key was recognizing, then routing drivers around, congestion pockets. Modifying a version of CoPilot, raw link travel times were collected on a server, smoothed with a travel time estimation algorithm the team wrote, then posted for distribution to the probes for recalculating their fastest routes. Besides collecting objective data via the server, pre- and post-experiment surveys were administered for perception insights. Preliminary results show that (a) technologically-inclined drivers are willing to use the device, (b) some drivers did experience improved travel times, (c) driver compliance was approximately 62 percent, and (d) traffic flow characteristics such as turning time distributions are as expected. Our analysis is ongoing; the current focus is (1) an in-depth assessment of driver compliance and trust, and (2) study of the link travel time characteristics.