Berkeley Engineering Home
Volume 5, Issue 5
May 2005



In This Issue
Spare Parts for the Human Body Shop

Trafficking in Roadway Sensors

Bioengineers Battle Stowaways at Sea

Cool Alumni

Dean's Digest

Archives 2005
2004
2003
2002
2001

Lab Notes, Research from the College of Engineering

Trafficking in Roadway Sensors
by David Pescovitz

Printer-friendly versionPrinter-friendly version
Varaiya

Professor Pravin Varaiya is a researcher with the California Partners for Advanced Transit and Highways.

We drive over them constantly but may have never noticed the octagonal shapes in the freeway. Buried below the surface of many roadways every third of a mile are wire sensors called loop detectors that, when they're working, count how many cars pass over them and the average time a car is on top of the loop. California alone has more than 25,000 loop detectors. In recent years, UC Berkeley researchers used the loop detectors as a traffic monitoring system to make commuters aware of congestion hotspots before they get on the road. The next step, says professor Pravin Varaiya, is to replace the troublesome and costly loop detector with wireless studs that not only count the vehicles but also distinguish between cars, SUVs, and trucks as they zoom past.

"Throughout the world, 90 percent of traffic sensing is done using loop detectors," says Varaiya, a professor of electrical engineering and computer science and researcher with the Center for Information Technology Research in the Interest of Society (CITRIS). "To install each one, you have to saw into the pavement and connect all the wires. Many loop detectors are broken and aren't repaired until roads are repaved."

Stud sensor

Each wireless stud contains a sensor, microprocessor, radio, and battery.

Varaiya became intimately familiar with loop detectors while designing the freeway Performance Measurement System (PeMS) in collaboration with the California Department of Transportation (Caltrans) and California Partners for Advanced Transit and Highways (PATH). The system collects data from the loop detectors and translates it into "news you can use" in the form of real-time traffic reports for commuters. PeMS also provides traffic engineers with historical data to alter operational decisions regarding metering lights or highway closures due to construction. Eventually, strategically-placed electronic message boards informed by PeMS could display constantly-updated travel tips such as: "You're going to spend six minutes on this ramp. Uou should proceed to the next exit instead."

Of course, a system like PeMS is only as accurate as its sensors. Inspired by Berkeley 's wireless sensor innovations like Smart Dust, Varaiya and his colleagues built a new wireless road sensor that's far more accurate and much less expensive than purchasing and installing a new loop detector. Approximately the diameter of a saucer for a teacup, the sensor nodes resemble reflector studs that dot roads. Like the reflectors, the sensors can be glued into place without sawing into the asphalt. Each node contains a small magnetic sensor, computer processor, short-range wireless radio, and battery.

base station

A wireless base station receives data from nearby traffic sensors and transmits via the cellular network to traffic management headquarters. [View larger image]

"As a vehicle's ferrous structure goes over the sensor, the sensor detects a localized shift in the earth's magnetic field in three dimensions," Varaiya explains. "Buses have different distribution of ferrous material than cars or trucks. So you can use the signature of that distribution to determine the kind of vehicle passing over the sensor."

Data is then transmitted to a base station mounted roadside in range of a handful of the nodes. The base station acts as a midway point for the data, ultimately delivering it via a cellular network to a central traffic management office. It's a two-way channel, enabling the sensor network to be programmed remotely as new software is developed.

Readying a prototype device for the road was no small feat, Varaiya says. From an electrical engineering perspective, it required a very efficient wireless communication system that would run on a single battery for seven years, comparable to the estimated lifetime of loop detectors. Finally, the casing had to be strong enough so "trucks could drive repeatedly over it without cracking it."

Once their design was proven out on real roads in Berkeley, Varaiya and two other Berkeley engineering alums spun out a company, Sensys Networks, to bring the technology to market. Back at the University, Varaiya is exploring variations on the sensor stud design and other real-world applications.

For example, outfitting the studs with tiny accelerometers, devices that sense direction and speed of motion, could measure the vibration of the pavement as a vehicle passes over it. That vibration correlates to the load of the vehicle, Varaiya says. Unlike existing roadside weighing stations, the distributed sensors wouldn't be used to tax trucks based on their load, at least initially. Still, they could be cheaply installed in many locations to gather data that's extremely useful in aggregate.

"When Caltrans considers how fast pavements are wearing, they'd like to know the weight of vehicles and how much load trucks are carrying," Varaiya says. "The studs could be placed right on the road at a cost of a couple thousand dollars each compared to several hundred thousand to build a roadside weigh station."

Your Turn

Do you have a comment or question regarding this research?

We want to hear from you...

Recently, Varaiya has considered how the sensor studs may help alleviate the other major headache associated with driving: parking. If simple wireless stud sensors could be produced for just a few hundred dollars each, it would be practical to install them in parking garages to keep tabs on each space. The researchers are just now embarking on experiments of this sort in the campus's student union parking structure. Meanwhile, Emeryville-based start-up ParkingCarma, who has partnered with PATH to test "Smart Parking" systems, is also looking at the Sensys Networks technology for future deployments.

"Parking lots could accurately advertise in real time on a display board how many spaces are open or, of course, say 'sorry, don't even bother, no spots available,'" Varaiya says.


Related Sites

Pravin Varaiya's home page

Freeway Performance Measurement System (PeMS)

Sensor Networks for Measuring Traffic

"On the Road to Smarter Highways" by David Pescovitz (Lab Notes, October 2001)

"Researchers launch Smart Parking project at Rockridge BART" by Sarah Yang (Media Relations, 07 December 2004)

Center for Information Technology Research in the Interest of Society (CITRIS)

Sensys Networks Inc.

ParkingCarma

California Partners for Advanced Transit and Highways

 


Lab Notes is published online by the Marketing and Communications Office of the UC Berkeley College of Engineering. The Lab Notes mission is to illuminate groundbreaking research underway today at the College of Engineering that will dramatically change our lives tomorrow.

Media contact: Teresa Moore, Lab Notes editor, Director of Marketing and Communications
Writer, Researcher: David Pescovitz
Web Manager: Michele Foley

Subscribe or send comments to the Engineering Marketing and Communications Office: lab-notes@coe.berkeley.edu.

© 2005 UC Regents. Updated 5/1/05.