Can transportation agencies call on smartphones for traffic data?
- By Patrick Marshall
- Feb 10, 2014
This is the first of a three-part series on the Department of Transportation’s plan to use technologies to improve highway safety. Read part two and part three.
The Transportation Department’s National Highway Traffic Safety Administration announced this month it would begin to take the steps necessary to enable new vehicle-to-vehicle (V2V) technologies in light cars and trucks, bringing the dream of systematic highway safety closer to reality.
The new V2V technologies would help vehicles avoid crashes by allowing them to “talk” to one another to exchange safety data, including vehicle speed and positioning information, through a combination of roadside data-gathering sensors and in-vehicle devices.
Yet even before the agency completed its Connected Vehicle safety pilot last November, transportation managers were beginning to see the initial stages of a potential revolution in traffic safety technologies: the integration of crowdsourced data from smartphones.
Connected Vehicle project
The federal Connected Vehicle project, which began with a test deployment in Michigan in August 2011, relies on V2V and vehicle-to-infrastructure (V2I) communications as well as dedicated short-range communications (DSRC) devices.
The safety pilot, which was conducted in Ann Arbor, Mich., involved 3,000 vehicles as well as 29 roadside sensor/communications devices that covered 73 lane-miles of roads.
Of the 3,000 vehicles, 64 were fully equipped with embedded devices that communicated data on speed, acceleration, deceleration, yaw rate, turns, wiper activity, braking and other conditions and activities. Another 300 vehicles were equipped with one or more after-market devices, such as GPS units. All of the 3,000 vehicles were equipped with beacons – called "vehicle awareness devices," or VADs – that emit a signal 10 times per second.
The safety applications envisioned in the program would not take control of a vehicle’s steering and brakes. Instead, they are designed to use the technology’s 360-degree situational awareness to help drivers prevent an accident. In a case such as when a driver is trying to decide if it’s safe to pass on a two-lane road, V2V communications can help detect threats hundreds of yards away, officials said.
“V2V crash avoidance technology has game-changing potential to significantly reduce the number of crashes, injuries and deaths on our nation's roads,” said NHTSA Acting Administrator David Friedman as the agency wound up a one-year “model deployment” of its Connected Vehicle project last August.
By all accounts, the safety pilot has been a success, though the NHTSA has released few details of the pilot. The agency said it is currently finalizing its analysis of the data gathered as part of its year-long pilot program and will publish a research report on V2V communication technology for public comment in the coming weeks.
"There is a scarcity of data because everyone is holding their cards close to the vest," said Leo McCloskey, senior vice president for technical programs at the Intelligent Transportation Society of America (ITSA), a non-profit public-private-academic research and advisory organization. "But no one is even hinting that there are any challenges in the architecture. It is working exactly as defined."
Traffic data from smartphones?
However, some industry observers note that the federal Connected Vehicle program may already be out of date in that it doesn't include the most rapidly growing source of all traffic data – smartphones.
According to Nancy Wilochka, public affairs officer at DOT's Research and Innovative Technology Administration, "We are just beginning research into the crowdsourcing area, so at this point we don’t have information to share."
Crowdsourced smartphone data could solve one of the major challenges of connected vehicle efforts – the expense of sensors. "Sensors are really expensive to deploy," said Ofer Avni, CEO of Cellint, a provider of cell-based traffic data. "After billions of dollars invested worldwide, only a very small fraction of roads are covered, less than 1 percent. So we look for alternatives."
"I believe it's a revolution," said Pushkin Kachroo, professor of transportation, electrical and computer engineering at the University of Nevada, Las Vegas. "Before, I used to build gadgets [to collect data]. Now we have a smartphone, so compact and with all the technology already in there. It has GPS, it has an accelerometer, it has a magnetic compass, it has a thermometer – it has all of that. And then to make it the best deal, it connects with wireless and through Bluetooth."
To top things off, says Kachroo, "There are software development kits so anyone can program for them. To me it is absolutely a revolution. Many applications are coming. It's just the beginning."
Bluetooth monitoring systems
Even before researchers started creating traffic applications for smartphones, developers of traffic management systems began taking advantage of the growing presence of smartphones to track traffic by sensing the Bluetooth signals emitted by the devices.
"You put an antenna at an intersection or along a freeway, you identify and re-identify these [signals] on a unique Bluetooth Mac address, and from that you can map trips that are being made and get travel times in real time," said Shawn Turner, head of the mobility division of Texas A&M's Transportation Institute.
TTI developed just such a system for the Texas Department of Transportation and the city of Houston. Visitors to the Houston TranStar website can view a map displaying in near real time current speed on major highways in the region.
One problem with Bluetooth monitoring systems is that, like the Connected Vehicle program, they require the deployment of equipment along roadsides. "Road sensors work very well," said Jim Bak, public relations manager at Inrix, a provider of traffic data. "But you are limited to where governments attach funding and the wherewithal to install the sensors. Then the challenge is how do you scale that if you want to cover every road? It just doesn't scale. It's cost prohibitive."
Turner agrees. And he thinks the future of traffic management is to be found in eliminating the roadside sensors. "A whole private industry has developed around crowdsourcing data from smartphones," he said. "In my opinion that is going to be more prevalent than what we're doing."
– Michael Cipriano contributed to this story.
Next: Sourcing the smartphone crowd