Shoebox-sized Robots Deployed in Rescue Effort at Ground Zero
Graduate students and the experimental robots they helped to develop were among the early responders who joined the search and rescue efforts shortly after the Sept. 11 collapse of the World Trade Center towers.
March 24, 2004
Shortly after the Sept. 11, 2001, collapse of the World Trade Center towers, robotics expert Robin Murphy, an associate professor of computer science at the University of South Florida (USF), received the call.
Within minutes of the attack, Murphy had also spoken with one of her former students, John Blitch, who directs the Center for Robot-Assisted Search and Rescue (CRASAR) at USF.
Murphy's research on experimental robots for urban rescue operations was originally funded by NSF. The CRASAR response team also included three graduate students -- Jenn Casper, Mark Micire and Brian Minten -- who helped Murphy develop the robots and their software-guided "marsupial" systems.
The CRASAR team prepared its robots near Stewart Air Force Base in Newburgh, N.Y., where Murphy, Casper, Micire and Minten were fitted with asbestos-rated respirators. They gained access to Ground Zero at 8 p.m. on Sept. 12. Murphy, a veteran of many smaller-building collapse training sessions, knew instinctively that they had little chance of finding survivors.
"The firefighters were grimly determined to find their colleagues," she said. "We were all determined to find whatever was possible. If we didn't find someone alive, finding remains and ensuring rescuer safety was just as important."
The CRASAR team had expected to meet in New York under the direction of National Institute for Urban Search and Rescue (NIUSR) board member Raymond Downey, chief of the fire department's special operations section. However, Downey was among the rescue workers killed when the towers collapsed. Murphy and Blitch are also NIUSR board members.
Murphy and her students had expected to deploy their intelligent anonymous "marsupial" robots, so called because the "mother" robot releases smaller robots to explore tight spaces unreachable by other means. The mother carries the little ones in her "pouch" as far as she can maneuver into the site, then provides power as the "babies" descend from her to perform their search, negotiating smaller crevices and hidden spaces. Equipped to maintain balance on rough terrain, the smaller team can reach, sense and report on spaces that may be too small or too dangerous for human rescue workers to approach or enter.
The severity of the WTC damage, however, prevented the use of the mother; instead, the shoebox-sized robots were taken by backpack into the rubble.
Over the next 11 days, the CRASAR teams made five insertions onto the massive rubble piles, often interrupted when safety concerns forced an evacuation. The hardiest of the robots -- which cost between $10,000 and $40,000 -- were sent into the rubble whenever requested by Federal Emergency Management Agency (FEMA) task force teams or sector chiefs. The mechanized prowlers had 100-foot tethers, far out-distancing the fire department's seven-foot camera wands.
While CRASAR robots helped find five victims and another set of remains, Murphy expressed regret that they hadn't been more successful. The marsupial concept still remains promising, especially to handle remote tether management, says Murphy. In the future, tether-free robots enabled by artificial intelligence would guide themselves through collapsed buildings or other difficult terrain, Murphy believes. She also hopes to develop algorithms that could distinguish colors and shapes to help robots locate victims, living or dead.
Before graduate school, Casper received an award under NSF’s Research Experiences for Undergraduates program. Now she and fellow students Micire and Minten have acquired the kind of field experience none of them would have hoped for, yet none would have missed.
"This was the worst thing I or anyone else has seen in the field," Murphy said. "It says a lot for my students that they rose to the challenge. Now we'll take these lessons home, with new inspiration to make our next generation of search and rescue robots as advanced as possible."
-- Peter West
University of South Florida
CISE Research Infrastructure
Robotics and Human Augmentation
#9320318 Reactive Sensing for Autonomous Mobile Robots
#0229809 SGER: Adaptive Shoring for Robot-Assisted Search and Rescue
#0224401 CISE Research Resources: R4: Rescue Robots for Research and Response
#9617309 CISE Research Instrumentation: Intelligent Assistance for Multiple Robots
#9732601 REU: Multiple Autonomous Mobile Robots for Search and Rescue Applications
#9996356 REU: Multiple Autonomous Mobile Robots for Search and Rescue Applications
#9531730 REU Site: Multiple Autonomous Mobile Robots for Search and Rescue Applications
Federal Emergency Management Agency (FEMA)
NSF News Release (with more images): http://www.nsf.gov/news/news_summ.jsp?cntn_id=104424