When rescuers need rescuing

Get Adobe Flash player
By Hiawatha Bray
Globe Staff / August 30, 2010

E-mail this article

Invalid E-mail address
Invalid E-mail address

Sending your article

Your article has been sent.

Text size +

WORCESTER — The blinded, disoriented firefighters who crawled up the stairs of a century-old brick building here recently had to find a missing colleague, and find him fast.

Their only hope was a battery-powered homing beacon worn by the lost firefighter. Fortunately, the gadget worked, leading the rescuers directly to him.

Immediately, all three of them sat up, stripped off their oxygen masks — and their blindfolds — and took deep breaths of fresh, smoke-free air.

Then they ran the exercise all over again.

The Worcester Fire Department conducted this search-and-rescue test three times this month. They worked inside a mathematics building on the campus of Worcester Polytechnic Institute, which hosted a major national conference on technologies to tackle a remarkably tough task: finding firefighters stranded inside burning buildings.

WPI scientists have wrestled with the challenge for more than a decade, along with scientists and entrepreneurs around the world.

The tests suggest their work is at last bearing fruit. On average, it takes about 24 minutes to rescue a missing firefighter. But during the conference, systems developed by a University of Michigan researcher and two private companies were able to locate a firefighter in as little as seven minutes.

Now there are plans to begin using the technology in the real world.

“We’re within 24 months, 36, or 48 months of having a commercially available product that will do exactly what we’re talking about,’’ said Jim Duckworth, an associate professor of electrical and computer engineering at WPI.

The Department of Homeland Security wants to start field tests next year, and a Utah company that participated in the conference hopes to begin selling a system as early as this year.

Today, “rapid intervention teams’’ sent to rescue firefighters can do little more than grope and guess. Technologies that can spot body heat through walls have long been a staple of high-tech television shows like “24,’’ but they don’t exist in real life. And a seemingly obvious solution — the Global Positioning System technology found in millions of cellphones — turns out to be useless.

“GPS works very well outdoors, but it doesn’t work at all indoors,’’ Duckworth said. So he and other scientists have been working on alternatives.

Duckworth and his WPI colleagues began their quest in 1999, after six Worcester firefighters trapped in a burning warehouse died. John Orr headed the electrical and computer engineering department at the time. As he watched the funeral procession, he said, “it literally occurred to me, technology can and should solve this problem.’’

Orr met with other members of the faculty and persuaded them to take on the challenge. WPI was especially well qualified to tackle the job. Apart from its strong electrical engineering faculty, the school hosts one of the nation’s top academic programs in fire safety.

The idea was to create a device firefighters could wear. The incident commander managing the fire scene would be equipped with a computer to intercept the radio signals and display each firefighter’s location on a screen. If someone got in trouble, the commander could radio his location to an intervention team.

Such a device would have to be lightweight, tough, and inexpensive enough for the nation’s 30,000 fire departments to afford. And since it would be used in smoky rooms with near-zero visibility, the gadget had to be highly accurate.

“The only way they could find the victim is actually when they got close enough to touch him,’’ Duckworth said. That meant getting within about three feet of a victim. The device would also have to calculate altitude, so rescuers would search the correct floor.

But building such a system has proven brutally tough. “It is an amazingly difficult problem,’’ Orr said.

Researchers tried using a radio beacon, then homing in on its location by picking up the signal on multiple receivers — a process called triangulation. But walls often reflect radio waves, so the signal seems to come from several directions at once. That “makes it appear like you’re in a house of mirrors,’’ Orr said.

Scientists have found some promising alternatives, however.

Two of the systems tested at WPI used inertial trackers like those used to guide commercial airplanes. The firefighter switches it on while standing at a known point outside the building. Chips inside the device detect every movement of his body. If he moves say, 20 feet to the east, and then 30 feet to the north, the device can calculate where he’s standing.

During the Worcester conference, Johann Borenstein, head of the mobile robotics lab at the University of Michigan, demonstrated a system mounted in a “lost’’ firefighter’s boot. Along with accurate motion tracking, the device contained a barometer that measured air pressure and thereby the wearer’s altitude, revealing the victim was on the third floor. The device used a radio to transmit the location to the incident commander. He, in turn, sent voice commands by radio to guide rescuers.

Due to a technical glitch, it took 12 minutes to complete the search. That’s half as long as a typical rescue, but not good enough for Borenstein. “It didn’t work quite as well as I had hoped,’’ he said.

Next, a rival inertial product, from Seer Technology in Salt Lake City, was strapped to the missing firefighter’s back. He was found in nine minutes.

The third device, from Q-Track Corp. of Huntsville, Ala., sends a short-range AM radio signal. But instead of using traditional triangulation, the Q-Track receiver compares the incoming radio signal and its accompanying magnetic field to figure out the source of the transmission. The device located the firefighter in seven minutes.

“The technology’s actually better than we were expecting,’’ Duckworth said.

Firefighters won’t be the only ones who will welcome an indoor location system. Police officers could use it to keep tabs on each other while searching a building, and the military needs a better way for soldiers to track each other in house-to-house combat.

So the WPI conference featured presentations from researchers at the Army Soldier Systems Center in Natick and the Defense Advanced Research Projects Agency, the Pentagon think tank that sponsored the development of the Internet.

More than 10 years after the effort began, indoor location devices could soon become available to rescue workers.

Jalal Mapar, a program manager for the science and technology directorate of the Department of Homeland Security, said it’s working on prototype systems that could fit inside radios used by firefighters and police. Field tests are to begin early next year.

That’s too long for Dick Gibson, vice president at Seer Technology. His company hopes to ship a system as early as next month, with an average price per user of $5,000. Hardly cheap, but it will seem like a bargain if it saves a life.

Hiawatha Bray can be reached at