The organizers of Fifth Precision Personnel Locator (PPL) Workshop (held on the campus of Worcester Polytechnic Institute, or WPI, August 2-3, 2010) collaborated with the Worcester Fire Department to conduct realistic firefighter rescue exercises. All three systems tested enabled rescues substantially faster than the 24 minute unaided rescue baseline. The quickest rescue of the day was performed with the assistance of Q-Track’s Near-Field Electromagnetic Ranging (NFER®) Firefighter Location and Rescue Equipment (FLARE) prototype system. For more details, please see this presentation or a conference paper preprint describing the system from the Q-Track website. Now, there is a video available of the simulated rescue.
The organizers of the WPI PPL Workshop were kind enough to make available a DVD with the full video of the exercise to my Q-Track colleagues. We added some subtitles to aid in understanding what was being said and added some annotations to aid viewers in understanding how the system worked and what it did. We had to speed up some parts of the video to fit in the fifteen minute limit for YouTube.
The WPI PPL Workshop helped open my eyes to the difficult situations under which firefighters have to work. In this exercise, the firefighters wore opaque face masks to simulate operation in a dense, smoke-filled environment. Their progress literally slows down to a crawl as they stay low – below potentially high temperature combustion products. In this exercise neither the original team (one member of which gets “lost”) nor the Rapid Intervention Team (RIT) have been in the building before. It’s almost painful watching how they have to grope around to feel their way through a building. Any assistance that can guide rescuers to the correct floor and help them home in on the correct location is invaluable.
Here’s a complete description of the exercise (excerpted from the conference paper preprint):
As part of this workshop, Ric Plummer, Engineer, Berlin, MA Fire Dept. worked in conjunction with John Sullivan, Deputy Fire Chief, Worcester Fire Department to create a realistic fire rescue scenario involving Stratton Hall on the WPI campus (see Figure). Stratton Hall has four levels and two entries: a basement level entry on the front or “A” side, and a first floor entry on the reverse or “C” side. This ambiguity in floor numbering complicated the rescue scenario.
Firefighters designate the working or front side of a building “A,” and label sides from their clockwise in alphabetical order through “D.” In their scenario, Stratton Hall had a fire at the top A/D corner of the building with possible victims trapped. The Primary Search Team (Engine 22), uses the first floor entry on the far or “C” side to enter the building. They proceed to the top (third) floor. As conditions in the building deteriorate, the Incident Commander (Chief Sullivan) calls for an evacuation of the building.
During the rapid evacuation, one firefighter on the Primary Search Team (Engine 22) becomes lost on the second floor in a classroom toward the center of the building. His partner calls MAYDAY upon exiting the building and realizing a firefighter is missing. The Incident Commander orders Ladder 33 to serve as a Rapid Intervention Team (RIT) and enter the building. RIT team members wore blacked out visors to simulate the no-visibility conditions of a smoke filled building. In a baseline physical search using standard procedures and unaided by any location technology, the Worcester Fire Department required twenty four minutes to locate the “lost” firefighter: sixteen minutes to clear the top (third) floor, and then eight additional minutes for a second RIT to locate the “lost” firefighter on the second floor. These results provided a baseline against which various location approaches were evaluated, including Q-Track’s NFER® FLARE prototype.
The location of the “lost” firefighter was unknown to the RIT or to the Q-Track team. Q-Track’s Steve Werner interpreted the FLARE GUI to aid the Incident Commander (Chief Sullivan) in guiding the RIT to the “lost” firefighter (as shown in Figure 4). By interpreting near-field RSSI levels, FLARE guidance directed the RIT to the center of the second floor. The RIT entered the correct classroom on the correct floor and began clearing the room clockwise with the “D” side wall. FLARE detected that they were off the path, and Chief Sullivan directed the team to back-up to the last known contact with the path. The RIT exited the room to try further down the hall. FLARE detected that they were off the path again, so the RIT re-entered the correct classroom and began clearing the room in a counter-clockwise fashion beginning with the “C” side wall. The RIT found the “lost” firefighter with an elapsed time of 7:02. Two primarily inertial systems were also evaluated, yielding rescue times of 11:56 and 9:42.
The prototype system demonstrated by Q-Track had some warts:
- The NFER® FLARE prototype required some set-up around the incident scene. Firefighters have lots to do at an incident scene and would prefer minimal to no set-up.
- At one point in the video, you’ll note we erroneously thought the victim might be on the top floor (“fourth,” from our perspective), an error that was quickly corrected as the RIT approached the second (“third” from our perspective) floor.
- The NFER® FLARE prototype is best at path following – so when the rescuers had to use a different entry, we weren’t able to provide more than general guidance until the RIT intersected the path of the “lost” firefighter.
- The NFER® FLARE prototype works by correlating the rescuers’ path to the victim’s path. The system guided the rescuers to the center of the second (again, third from our perspective) floor – directly to the same room where the “lost” firefighter lay under a table. But because the victim entered the room following the wall in a counter-clockwise fashion and the rescuers began clearing the room in a clockwise fashion, the rescuers left the victim’s path. Noting this, the NFER® FLARE system guided them to leave the room to reacquire the trail. When they did, they got on the path again, re-entered the room, and began clearing counter-clockwise: on the “lost” firefighter’s trail.
When the rescue team got within about ten feet of the lost firefighter, we were able to instruct the rescuers to “conduct a careful search of that area.” Before the Chief could relay that instruction, the rescue team found the “lost” firefighter. Without further ado, here’s the video of the simulated rescue:
Again, I gratefully acknowledge the assistance of the organizers of the WPI PPL Workshop as well as the members of the Worcester Fire Department who worked so hard to provide Q-Track and other RTLS vendors with a realistic test of our systems’ capabilities. My Q-Track colleagues and I will be taking the lessons we’ve learned and we will continue to work on improving the NFER® FLARE system as we can afford to spend time away from our funded projects.
Previously on ÆtherCzar:
- 5th WPI PPL Workshop Proceedings Online: an update with a link to the conference proceedings,
- Q-Track’s Tracking System Prototype Successful in Trial: coverage from the Huntsville Times,
- Q-Track’s Firefighter Location System Succeeds in Trial: the “day-after” report on the firefighter rescue trials including photos, and
- Precision Indoor Personnel Location and Tracking: background information on the WPI PPL Workshop.
Additional information and a report on the WPI PPL Workshop are available from the TRX Blog, here.