What is it with disasters? They’re coming fast and furious. Here’s the 411 on robots at the China quake.
The Qinghai quake is the latest of the series of tragedies. Prof. Bin Li at the Shenyang Institute of Automation and an active member of the IEEE Technical Committee on Safety Security Rescue Robots, contacted the Chinese national earthquake response service this morning. It doesn’t look like ground robots are appropriate– the structures are mostly small and constructed from brick and mud. That type of construction is problematic– the brick and mud turns to a liquidized dust, acting like water to fill all the voids and displaces air. Even if there are voids, the suspended dust causes respiratory distress. Eric Rasmussen InSTEDD has many tales to tell of the similar Turkey earthquake.
China, by the way, does have at least one rescue robot. Bin tells me it was deployed to the mine collapse but could not be used because it wasn’t waterproof. (A gentle aside to manufacturers: d’uh!)
Aerial vehicles might be helpful for tactical operations and I can’t help thinking that an unmanned marine vehicle with an acoustic camera capable of penetrating turbid waters could provide more information about that crack in the big dam…
Bin was a participant in the NSF-JST-NIST workshop at Disaster City at the first of the month and we look forward to working with him and his group. In the meantime.. I’m speaking tomorrow at AUVSI day at the Capitol– I hope that in the future we can do more than offer our prayers.
Just as I was sitting down to blog (with relief) about the rescue of the Chinese miners, the explosion in West Virginia hit the news. Terrible, terrible. Our prayers go out to the miners and families.
Rescue robots have been used in mine disasters in the past and perhaps they will be of service in Raleigh, WV.
Some background on mine rescue robots. From 2007-2008, I led a study for the Mine Safety and Health Administration (MSHA) on underground mine rescue robots, getting to work with Dr. Jeff Kravitz and his team, attending a rescue competition at the MSHA academy in West Virginia, plus participated in two different mine disasters: Midas Gold Mine and Crandall Canyon Utah. A summary of the study was published in IEEE Robotics and Automation Magazine.
Jeff had been exploring rescue robots from mine disasters since before 9/11(!) and has pioneered their use- MSHA is second only to CRASAR in the number of deployments to actual incidents.
Coal mine disasters are tricky for robots- in part because methane is often present. As in “hmm… which would be worse- the explosion or the huge amounts of coal catching fire?” I know of only 1 robot rated to work in those conditions and not trigger an explosion- MSHA’s modified Remotec Wolverine, called V2. Making it intrinsically safe made it much bigger and much heavier, which are not necessarily pluses for agility. MSHA has an iRobot Packbot and is upgrading it, while NIOSH has a set of Gemini robots developed by Sandia Labs. The smaller robots we used at Midas and Crandall Canyon didn’t have to be intrinsically safe because one was a gold mine and the air quality testing at the coal mine showed there was no significant presence of methane- mine environments in the Rockies are different from mine environments in the East.
Mine disasters are also tricky because there are at least three different scenarios for using a robot- and each scenario favors a different type of robot. One size does not fit all. I’ll try to upload some slides to go through the unique challenges posed by each scenario.
The news this morning of the Chilean earthquake continues to highlight the possibilities for robots to help in the response and recovery.
Certainly the big surveillance UAVs such as the Predator and Global Hawk variants can give a “big picture” overview, but don’t forget that search and rescue is largely done by small teams working independently who need information “on demand”. They can’t tell you what they want until they get there. And if they don’t get within a few minutes, they move on. Small helicopter like UAVs can help them assess a collapsed building, see people trapped on the other side of a pile of rubble, and get the views they want. Small, shoe-box sized robots can drive into rubble deeper than a search camera or boroscope can go, finding voids that would be otherwise missed. And unmanned marine vehicles can determine if the substructure of bridges and sea walls is still intact and whether rubble and debris is being washed into the footings and will cause damage.
We’re standing by, hoping to help and hoping to learn. And hoping that one day responders all over the world will have these robots to use immediately.
I’m getting a lot of questions about why wasn’t CRASAR, or any robotic technology, in Haiti. Interestingly, some of the questions indicate anger at agencies and while I thought CRASAR could have been of use in saving lives and learned important lessons for science, I’d like to defend the decision not to invite us.
It is always a difficult call for an incident commander to bring in new technology that they have not trained with or has been shown definitely to work. A disaster is just that—it’s usually a surprise and by definition exceeds resources. So the command structure is busy just trying to do the regular things of getting traditional resources to the right place. And people under stress fall back to the things they are most comfortable with. Consider too, the individual responders are too tired and stressed to take on out new equipment (more things to lug around) even if they are familiar with it. That’s why for the first 4 years following 9/11, I gave nearly a hundred talks to response agencies and participated in many exercises so that we could show the responders what the robots could do and get their comfort level up. We created a 2-hour awareness course and a 10 hour introductory course that responders could get continuing ed credits for. The role of giving responders hands-on time with robots has largely been taken over by NIST and their rescue robot standards program.
And remember most new equipment has terrible interfaces and ergonomics, so it is a true pain to use. This means bringing trained operators to use the equipment on behalf of the responders adds to the logistics footprint- here’s a couple more people that aren’t on the official roster and have to be accounted for. And the tech operators may have no experience or response training- and there are basic procedures and terminology that you need to know. The liability and logistics is just hard. It is way easier for the incident commander to just to say “no.” CRASAR is all about the technicians getting response training so we won’t be a burden or a liability.
Also keep in mind that disasters always bring out people who are well-intentioned but have no clue whatsoever. I have some horror stories from the Crandall Canyon Utah mine disaster so I can definitely sympathize with the incident commanders. The fire service typically just says “no” based on past experiences because they don’t have time to get distracted with such things—so if they didn’t know you and felt comfortable with you before a disaster, you aren’t likely to get your foot in the door. Trying to pressure them just makes it worse for the rest of us. The robots used at 9/11 were invited by the NY State and City emergency departments through the connections of Lois Clark McCoy at NIUSR, but the responders viewed them warily and did not take us to the field. John Blitch led a small group that used the robots on the first day but the second day when you couldn’t get to the site without being part of a tasked assignment, we just sat there. A guy from a major government lab in a suit was wandering around the Javits Center where all the response teams were housed talking about how great the lab’s sensors were. (Back up- A guy in a suit. At a disaster site. That certainly undermined any credibility that these guys had ever stepped outside of their lab, much less did rigorous field testing. ) Then suddenly the FEMA teams started asking for us to come with them- primarily because Chief Ron Rodgers (bless him!) at Florida Task Force 3 posted to a responder chat room that my group had worked with him in the field and the students and I had completed basic response training. We became known quantities.
There’s also a matter of scale. The incident command team is responsible for doing the most good for the most people. Will a couple (or even a hundred) of experimental technologies really make a difference and be worth the disruption to the already stressful way of doing things and additional personnel and logistics burden? Or is a more rational decision to focus on doing the basics? That’s the incident command teams call and I respect that.
The point is not whether CRASAR participates in a disaster but rather whether we are getting closer to the day when the responders routinely take the robots and other technologies that they own and operate to the incident- that’s our mission.
Tomorrow is the anniversary of the 1995 Kobe Earthquake. The irony that I am in Kobe accepting the Motohiro Kisoi Award for Academic Contributions to rescue engineering instead of in Haiti does not escape me. There is always a gap between possibility and reality, but gaps about high definition TVs seem trivial compared to gaps in life saving and recovery.
Yesterday Ms. Ikuko Tanimura from the International Rescue Systems institute took me to the Hyogo Perfectural Emergency Management and Training Center and the full-scale earthquake testing facility at the National Research Institute for Earth Science and Disaster Prevention. Suffice it to say that the Japanese have the technology to shake entire 6 story buildings and bridges in three dimensions and understand collapses. Recently, they shook to pieces a wooden house and let the IRS researchers apply their technologies (I am so envious!) Dr. Akiko Yoshimura, an architect, designed a clever facility where teams can practice victim management in wet, confined spaces designed to tax the ergonomic constraints of responders. As I travel the world, I see so much good science, good ideas, good inventions!
The Japanese researchers from IRS are sanguine about progress and the time it takes to go from research to the field. IRS director Prof. Satoshi Tadokoro started what became IRS in 1995 in response to the Kobe earthquake and the loss of Motohiro Kisoi, a promising graduate student in his department. I also started in 1995, motivated by the Oklahoma City Bombing. The research directions Satoshi and I initiated back in those days are a little embarrassing in retrospect- we didn’t understand disasters and there was little data or experience base. Now as we’ve profited from being engaged in exercises and actual responses, being able to apply cognitive work analysis methods, and collect performance data on machines and people, the community is beginning to isolate and address more meaningful issues that will lead to truly useful technology that will be easy to use and maintain.
But as we discussed last night at the reception, good science isn’t sufficient to help a disaster like Haiti. We need industry to (cheaply) manufacture the devices, agencies and NGOs to accelerate adoption.
But what we really need are early adopters and caches all over the world, so that even it doesn’t take 3 days for response teams to bring in the sensors and robots (and comms and power), that the local responders can make the most of the critical 72 hours.
The sad news of the mudslide in Angra dos Reis, Brazil, brings up memories of our deployment to the 2005 La Conchita, California, mudslides. Rory Rehbeck, then a captain with LA County Fire Department, invited CRASAR out to assist Ventura County Fire Department. There really aren’t survivors of a mudslide- the mud is a liquid, penetrates like water, and covers everything. The best you can hope for is survivors from the collateral damage. The houses on the slope of La Conchita were either buried, squished as if inside a giant trash compacter, or untouched. We attempted to use the new Extreme robots we had purchased through a NSF grant to search some of the damaged houses as a family of 6 was still missing (they were on vacation) and the canines were giving some ambiguous hits.
Our journal article “Rescue robots for mudslides: A descriptive study of the 2005 La Conchita mudslide response” Journal of Field Robotics, vol 25 no 1-2 (Jan 2008) p 3-16 gives the details of what Sam Stover and I experienced: the robots did not do well in the mud and vegetation when we tried to go under a house to get in it nor work in deep shag carpeting when we entered another house through the garret window. See the Media Gallery for photos. But being there did identify the need for remote sensor networks dropped off by UAVs to continously monitor for further slides (geologists checking manually every 6-8 hours isn’t good enough)- sensor networks for advanced placement already exist, they just don’t get used. We’re looking forward to combining the UAV work here with Prof. Dez Song’s work in sensor networks.
The families in Brazil are in our prayers and hearts.
I joined the team led by Prof. Daniele Nardi (Universita’ di Roma La Sapienza) for a one day visit to L’Aquila, Italy, the cultural epicenter of the April 6 earthquake that killed 300 and displaced thousands of others.
It was a long day- I flew into Rome on an overnight flight, arriving Wed at 7:45AM to a quick change into my response gear, a warm welcome from Daniele and his colleagues, and a one and a half hour drive to the mountains. Daniele had gotten the local fire rescue department to give us a heartbreaking tour and permission to fly his group’s small Ascending Technologies Quadrotor on-site. I was along as an observer on this trip, so I didn’t bring any robots, just cameras and my field notebook.
As we drove east from Rome, the mountains were a carpet of spring green dotted with the purple blooms of the Judas trees and the outskirts of town a maze of gardens of tulips and wisteria in full bloom, ins sharp contrast to the maze of collapsed buildings just a few kilometers ahead.
L’Aquila is a historical town in the mountains, along a ridge in a valley, famous for saffron. Think of a 15th century New Orleans French Quarter, the Canal Business District, and the Garden District all along a ridge, providing a mixture of very old, very new, commercial, apartment buildings, and villas. The town has a large university with the flavor of Tulane.
Our first stop was the ruins of an apartment building where a 24 year old student was pulled out alive.
Our second stop was an apartment building where 24 people died.
And everywhere, toys, stuffed animals, legos, and baby blankets peeking out from the rubble causing my heart to clench in fear, “what about the children?”
The team flew four flights among the ancient villas and then a flight downtown along buildings tumbling into the narrow streets, notice the Italian flag in the background. The flight conditions were perfect, no wind and the rain clouds eliminated glare. While the team had no mission, the rubble served as natural targets for trying out camera configurations and flying strategies. The environment offered many examples the close quarters of the types of clutter that make flying in urban areas so challenging: trees, flags, telephone lines, etc.
The displaced will remain so for a long time. The aftershocks are expected to persist through August and it is difficult to assess the structural condition of the buildings and even more difficult to remove or repair structures. We passed two large tent cities on the outskirts of town.
As with hurricanes in the US, the recovery from earthquakes may be harder than the response. To me, it cries out for embedded sensors and smart structures to measure the impact- either before or inserted after a quake in anticipation of aftershocks. It also suggests that we need to explore ways to develop new sensors and use humans and unmanned systems better to more quickly inspect buildings, perhaps using the internet to send data to experts all through out the country rather than having 1 or 2 inspectors or claims adjusters act as a bottleneck. And I wonder how much information is known to the population but unharnessed because we still lag in exploiting connectivity and the “wisdom of the crowd” showing up in twitter, flickr, and other social software?
We’re in the airport now (painfully early in the morning at the Cologne airport, making our way home. The firemen found one of the victims early Sunday morning, the old fashioned way by slow, tedious extrication. As of this morning (Monday), it doesn’t look like the second victim has been found. We were not called out on Sunday but continue to learn more about robots that would be suitable for these conditions. It was a bit embarrassing, the Mayor, Deputy Mayor, and the Chief all shook our hands and were very enthusiastic, despite us doing nothing tangible. Indeed, we are very grateful because we learned so much.
The one photo that “got away” was that of a dozen women with cardboard boxes working in the light rain, picking up documents from the rubble a few meters from the rescuers carefully excavating for victims in the dangerous, unstablized rubble. The women looked like they about to weep from the lost and damage to the Archives, while the fire rescue workers were grim. A sad moment
Quick notes before getting some sleep… we have been on standby since going to the site on Friday around noon. City of Cologne and fire department (still fuzzy on naming translations) have been terrific and the Fraunhofer Institute IAIS incredibly supportive.
The Archives buildings was a new, modern multistory commercial building surrounding by multistory residential buildings. When it collapsed into the hole, some of the surrounding buildings were damaged. The one where the two victims are missing is an older (like more than a 100 years old looking) brick building. When brick buildings collapse, the brick crumbles into sand and small pebbles, filling every possible void. Even the ASC couldn’t get in.
The robots were requested this afternoon for the mixed rubble from the houses and the Archives, but it wasn’t a good fit. There were small voids but we couldn’t stand at the face of the rubble due to safety reasons- and the ASC requires us to be right there. The voids big enough for the larger Extreme, which we could operate from the safe location ~10m away, were shallow and thus didn’t require a robot. More notes on how to build better robots.
We’re back on site tomorrow- a huge crane is being brought in to do more excavations and more voids may open. It looks like the old fashioned tedious manual rubble removal is the best technology for this job for now.
Lots of pictures but haven’t gotten permission to release them yet, here’s link to pics from the media and gratuitous coverage 😉
