Word from Responders: “Small UAVs are Available, Now Help Us Use The Data They Generate!” REU Students Provide Apps

Virtual reality reconstruction of Blanco River flood by Dr. Russ Taylor from CRASAR flight

TEES just concluded a three day Summer Institute on Floods (July 26-28, 2015), which was hosted by our “mother” center, the Center for Emergency Informatics. The Summer Institute focuses on the data-to-decision problems for a particular type of disaster. This year’s Summer Institute was the second in two-part series on floods and brought together representatives from 12 State agencies, 15 universities, and 5 companies for two days of “this is what we did during the Texas floods” and one day of “this is what we could do or do better” experimentation with small unmanned aerial vehicles, crowd sourcing, computer vision, map-based visualization packages, and mobile phone apps for common operating pictures and data collection.

A portion of the Summer Institute focused strictly on UAVs. The session was organized by the Lone Star UAS Center and the TEES Center for Robot-Assisted Search and Rescue (CRASAR), both of whom flew during the Texas floods.

The UAV field exercises  spanned four scenarios witnessed during the floods:

  • Search of cars and other vehicles swept away by a storm surge for trapped victims;
  • Search for missing persons who may have fled campsites or been swept down river;
  • Assessment of damage to power lines and transformers and presence of standing water which prevents restoration of power infrastructure; and
  • Assessment of damage to houses and estimates of household debris, which is critical to insurance companies estimating damage and agencies such as the American Red Cross in projecting social impacts.

The exercises were staged at the 1,900 acre TAMU Riverside Campus with one fixed wing and three types of rotorcraft  flown by CRASAR and David Kovak, a member of CRASAR’s Roboticists Without Borders program.

A major finding out of the institute was the realization that while State agencies are adopting UAVs, the agencies can’t process all the imagery coming in. For example, a single 20-minute UAV flight by CRASAR at the floods produced over 800 images totaling 1.7GB. There were over a dozen platforms flying daily for two weeks as well as Civil Air Patrol and satellite imagery. Each image has to be viewed manually for signs of survivors, clothing, or debris that indicate a house or car (and thus a person too) was swept to this location.

Given the huge demand for automating image processing, it was no surprise that a panel of four judges from the agencies awarded $900 in prizes to three students from the National Science Foundation Research Experiences for Undergraduates (REU) program. The prizes were for software that classified imagery and displayed where it was taken.

First place went to Julia Proft for an application that used computer vision and machine learning to find images from the floods which contained a spectral anomaly such as color. Below is UAV image flagged with an amaloy and anomalies indicated on a parallel image to make it easier to find where in the image the anomalies occurred without marking up the image and blocking the view.

UAV image flagged with an amaloy and anomalies indicated on a parallel image

urban debrisSecond place went to Matt Hegarty for an application that used computer vision and machine learning to find images from the floods which contained urban debris having straight lines or corners. In this image, the program found  instances of trash, pvc poles, and other indications of where houses (and possibly victims) had been swept down river.

 

Third place went to Abygail McMillian for a program that visually displayed where and when UAV data or any geotagged data was taken. The spatial apsect is important for responders to see where assets have sensed- at Nepal, UAV teams often surveyed the same areas but missed important regions. The temporal aspect (a timeline scale on the map) is important because with the changes in the river flood stage, hydrological and missing person images become “stale” and flights should be reflown to get new data.

Data display with timeline

The 10 students (nine from under-represented groups) were from Colorado, Georgia, Puerto Rico, South Carolina, Texas, Vermont, and Wyoming and spent  the summer conducting research through the Computing for Disasters NSF REU site grant under the direction of Dr. Robin Murphy. Multiple State agencies requested those student-created apps be hardened and released. In previous years, the panels have awarded prizes primarily to hardware- UAVs, battery systems, communications nodes, etc. This year, the focus was on software and programming to help analyze the data the hardware generates.

 

 

Drone Helps Rescue 2 Boys from Raging River

Two boys needed to be rescued from the raging Little Androscoggin River in Maine after their tube overturned Tuesday. Only one of them was wearing a life jacket. Frank Roma, chief of the Auburn fire department, wanted to get that boy a life jacket before attempting the rescue. The water was rough, and rescuers had a hard time getting to the boys, so they used Roma’s personal drone to deliver a life jacket and a safety line to the boys.

“We wanted to make sure we got a life jacket on that second child so that if they did fall in the water we could catch them downstream,” said Roma. “We used the drone to fly a tag line out to the young man that was on the rock, we instructed him to untie and to pull life jacket over to him.”

Check out more information here

Stop Laughing at Those Clumsy Humanoid Robots

The Humanoid Robot, built like a linebacker with an oversized head, tiptoes on two feet through the dirt. It’s free of any wires. It’s unleashed—but it’s now wavering. They were all part of a competition in Pomona, California put on by Darpa, the far-out research wing of the Pentagon. After the Fukushima disaster in 2011, Darpa set out to encourage the development of robots that can assist in similar catastrophes: machines capable of working where humans dare not go. And so the yearly Darpa Robotics Challenge was born.

To explore something like a contaminated nuclear reactor, a robot would have to conquer not only piles of rubble in the facility, but also be able to open doors and climb stairs and ladders. In a human-designed space, the thinking goes, a humanoid robot would be best equipped to handle the job. And indeed, for all their clumsiness, the semi-autonomous robots (human operators still do much of the controlling in the challenge) passed some impressive tests, including driving an ATV.

The thing is, relief workers have had operational “tracked” robots for 15 years that roll along on tank-like treads. They even helped out in the aftermath of Fukushima, and still run tests there to this day. Bipedal humanoids, on the other hand, have never gotten near an actual disaster. They’re expensive, and you don’t have to be a physicist to notice these robots are top heavy.

So why even bother developing bipeds? Well, when Darpa’s handing out $2 million for first place in its challenge, building a walking robot that drives ATVs must seem like a sweet deal.

Check out more information at wired.com

Update: Flooding disasters continue to mount casualties and challenge response efforts

We are working with our partner the Lone Star Unmanned Aerial System Center, one of the 6 FAA test centers, to help support the flooding response and recovery efforts for the Texas floods. The floods are a sad example of how flooding costs the U.S. more than 80 lives and $8 billion in damages each year. And that excludes storm surge from hurricanes.

Technology based solutions have advanced to where employing robotic assisted solutions can aide in crisis assessments by federal, state and local officials and emergency workers. Yet more data adds another layer of complexity and extra coordination of robot assisted efforts.

Accurate ground and aerial surveys can help decision makers choose where to deploy limited resources in the best way. By quickly identifying survivors, any follow on threats from the natural disaster, collecting measurements of debris fields, volume of water flows or similar data can be provided without endangering the emergency responders.

Water gone wild and what the Center for Robot-Assisted Search and Rescue (CRASAR) is working to do:

Texas A&M’s Center for Emergency Informatics (CEI) continues its two-year study of information technologies, including unmanned systems, in response to floods. The CEI brings together practitioners, academics, and industry experts to converge and test their knowledge hardware and software to respond to an array of water borne disasters.

CEI as the host for the Summer Institute on Flooding, slated for June 16-18, will set the stage for testing ideas, marry technologies and coordinate decision makers in real-time flood based scenarios

From the 2014 summer institute 42 representatives from 12 states tapping 14 agencies, 14 universities, and eight businesses were able to judge what information technology was mature enough, or which needs more development to assist with flooding response.

This year’s summer institute will include three exercises, each representing the key problem missions identified by the participating agencies:

  • Swift water rescue – a scenario has been developed where suddenly rising waters cut off helping responders to rescue people. Responders from Texas Task Force 1 using the characteristics of a campground flooding will be offered realistic challenges. The TEEX designed exercises focus on real missions rather than around technologies.
  • Life-saving response and immediate mitigation: identify best ways to assess where’s the flood, who is at risk?
  • Restoration and recovery– exercise decision making for resource allocation to restore roads, electricity, sewage, etc. And engaging insurance companies to conduct property damage assessment and projections of debris generation estimation to manage post flooding problems with decay, vermin and disease.

The growth of small UAVs and flooding response:

Small UAVs have been tasked for at least eight disasters with flooding or had flooding associated with it:

  • Hurricane Katrina 2005
  • Typhoon Morakot, Taiwan 2009
  • Thailand Floods 2011
  • Typhoon Haiyan, Philippines 2013
  • Boulder Colorado floods 2013
  • Oso Washington Mudslides 2014
  • Balkans flooding Serbia 2014
  • Cyclone Pamela Vanuatu 2014

From the several UAV platforms that went airborne each provided rapid and often timely information to officials and for responders in search and rescue missions as well as the difficulties with recovery.

General reconnaissance: addressing where’s the flooding, where are the people cut off by the flooding, and what roads are still passable.

Hydrological situation awareness: both real-time and post-processed. The flooding caused by the Oso mudslide was a real problem and rotorcraft could hover and stare at the river, letting the hydrologists estimate the flow rate in different areas. The biggest need remains surveying the amount of flooding. And with the addition of photogrammetric image software – UAVs provided details of the terrain and potential for additional flooding or the best place to put a dyke, channel or other mitigation. Our partners at FIT took it further and printed out a 3D model of the terrain to help everyone visualize the terrain.

The power of Unmanned Autonomous Vehicles:

  • Over watch for swift water rescue teams: our friends at South Carolina Task Force 1 have been pushing us to help create the protocols for using small UAVs. Seeing something such as a logjam that might be coming down river and pose a threat to life and limb as they work to rescue people. They’ll get their wish as this is one of the scenarios to be unfolded in our 2015 Summer Institute on Flooding in June hosted by CEI.
  • Debris estimation: both the debris directly from the flood and the indirect debris that follows on from people having to rip out sheet rock and toss water logged carpets. The advances in photogrammetrics make it possible to estimate the volume of debris. That is based on having the “before” survey of the area. To test it, we flew with PrecisionHawk at the Bennett Landfill superfund site in February 2015 to estimate the volume of toxic trash, which was on fire and needed to be safely removed. Part of the data analysis included estimating the content type, because vegetation and construction materials have to get handled and processed differently.
  • Delivery of supplies to isolated regions. We learned during last year’s summer institute that if the locals can hold their own for 72 hours, usually that was sufficient. In Texas, where breeding stock can represent hundreds of thousands of dollars of investment, someone to stay behind may make sense. Determining that they are not a victim is vital. Disasters can take people by surprise with a bridge being washed out or a vital need of medicines and other perishables. A group of Texas A&M aerospace students won 2nd place for their small, hefty fixed-wing UAV that could be used to drop off heavier/bigger bundles of supplies from further distances. Groups like Matternet, who like CRASAR are members of FIT, are looking at delivering medicine with rotorcraft.
  • Delivering lifelines, life jackets, and small things to people trapped on roofs: A note about delivering things with a rotorcraft– Using rotorcraft to carry a line or bottle to someone is complicated by the weight and distribution of them. Those factors usually make the UAV very sensitive to wind and control errors. As such, if an open rotor system is used, more distance than normal from a person is needed to account for such errors. One senior design project from last year’s Summer Institute created a two-way audio system for rotorcraft. Since rotors can generate about 85 dB of noise hanging a microphone and speaker on the UAV is ineffective. The Computer Engineering team used noise reduction algorithms from National Instruments LabVIEW to prototype a lightweight 2-way audio system impervious to noise.

Look for more to come from the 2015 Summer Institute for Flooding, June 16-18.

 

Flooding and small UAVs

At this time, we aren’t involved in the horrible Texas and Oklahoma flooding. But we’ve been studying flooding since our deployment with Hurricane Charley in 2004. This blog will give a short background on what we do here at A&M and the upcoming 2015 Summer Institute on Flooding, then the history of small UAVs for flooding, and the potential uses generated by experts at our 2014 Summer Institute on Flooding.

About CRASAR and flooding

The Center for Emergency Informatics (CEI) here at Texas A&M is in the middle of a two-year exploration of information technologies, including unmanned systems, for floods. The CEI is a “center of center” that brings together practitioners, academics, and industry to fuse and apply their knowledge to disasters.  CRASAR is the “response” arm for CEI (actually we do participatory research versus response- similar to what anthropologists like Margaret Mead do. Sometimes the only way to learn is by doing by embedding with the responders, we don’t do disaster tourism).

The CEI is hosting a Summer Institute on Flooding June 16-18, the second on flooding. The choice of flooding was motivated by the fact that flooding is costing the U.S. over 80 lives and $8 billion in damages each year, excluding storm surge from hurricanes. Last year’s summer institute brought together 42 representatives and 12 states from 14 agencies, 14 universities, and 8 companies to consider what information technology is mature enough, or needs a bit of encouragement, to assist with flooding response.

This year’s Summer Institute consists of three exercises, each representing the key problem missions identified by the agencies:

  1. Swift water rescue (June 16)- helping responders rescue people suddenly cut off by rising waters. The plan for the swift water rescue reads like the campground flooding- showing the advantage of having the responders from Texas Task Force 1 and TEEX design exercises around real missions rather than around technologies.
  2. Life-saving response and immediate mitigation (June 17)- where’s the flood, who is at risk?
  3. Restoration and recovery (June 18)- restoring roads, electricity, sewage, etc. as well as insurance companies conducting property damage assessment  and cities generating debris estimation so that they can keep the rats out.

About small UAVs and flooding

Small UAVs have been used at least 8 disasters from flooding or had flooding associated with it: Hurricane Katrina 2005, Typhoon Morakot, Taiwan 2009, Thailand Floods 2011, Typhoon Haiyan Philippines 2013, Boulder Colorado floods 2013, Oso Washington Mudslides 2014, Balkans flooding Serbia 2014, and Cyclone Pamela Vanuatu 2014.

Historically they have been used for:

  • General reconnaissance: where’s the flooding, where are the people cut off by the flooding, and what roads are still passable.
  • Hydrological situation awareness, both real-time and post-processed. The flooding caused by the Oso mudslide was a real problem and rotorcraft were considered because they could hover and stare at the river, letting the hydrologists estimate the flow rate in different areas. The biggest push was for surveying the amount of flooding and- with the addition of photogrammetric image software- the terrain and potential for additional flooding (or the best place to put a dyke, channel or other mitigation). Our partners at FIT took it further and printed out a 3D model of the terrain to help everyone visualize the terrain.

Potential uses for UAVs

  • Overwatch for swift water rescue teams: our friends at South Carolina Task Force 1 have been pushing us to help create the protocols for using small UAVs to help them see that a logjam is coming down river and going to wipe them and the people they are trying to rescue out. This is one of the scenarios we will be playing in our Summer Institute on Flooding in June
  • Debris estimation: both the debris directly from the flood and the indirect debris a few days or weeks later from people having to rip out sheet rock and carpets. The advances in photogrammetrics make it possible to estimate the volume of debris— if you have the “before” survey of the area; we flew with PrecisionHawk at the Bennett Landfill superfund site in February in order to estimate the volume of toxic trash (which was on fire) that needed to be safely removed. The next step is to estimate the content, because vegetation and construction materials have to get handled and processed differently.
  • Deliver supplies to cut off regions. We learned during last year’s summer institute that if the locals can hold out for 72 hours, usually that’s sufficient. Indeed, in Texas, where breeding stock can represent hundreds of thousands of dollars of investment, it may make sense for someone to stay behind. But sometimes people get taken by surprise or a bridge washes out unexpectedly and need diabetes medicine and other perishables. Groups like Matternet, who like CRASAR are members of FIT,  are looking at delivering medicine with rotorcraft. A group of Texas A&M aerospace students won 2nd place for their small but hefty fixed-wing UAV that could be used to drop off heavier/bigger bundles of supplies from further distances
  • Delivering lifelines, life jackets, and small things to people trapped on roofs. A note about delivering things with a rotorcraft- You can use rotorcraft to carry a line or bottle to someone but the weight and distribution usually makes the UAV very sensitive to wind and control errors— if you are using a open rotor system just be aware and maintain more distance than normal from a person if at all possible. One senior design project that resulted from last year’s Summer Institute was to create a 2-way audio system for rotorcraft. The rotors generate about 85 dB of noise so if you’re a responders and want to try to talk with a person by hanging a microphone and speaker on the UAV, it probably won’t work. The Computer Engineering design team used noise reduction algorithms from National Instruments LabVIEW to prototype a lightweight 2-way audio system impervious to noise.

 

 

 

 

Nepal: recommendations for small UAVs

As the tragedy in Nepal unfolds, the immediate rescue response has ended and now efforts are shifting to agencies working on the mitigation of the event and dealing with continuing cascade of consequences and hopefully to recovery as well as humanitarian relief.  We have not been asked to participate and cannot self deploy but to those planning to fly small UAVs, I recommend that you look over the range of uses of small UAVs in the past 8 earthquakes in the past blog (and in more detail in Disaster Robotics). Plus:

  • Be aware that the altitude may change the performance of your platform
  • Working in complex terrains such as mountains will impact any preplanned paths. We have found that imagery reconstructions from fixed-wings will do better with a series of flights “stair stepping” along a hill or mountain than trying to cover the entire area at one altitude. Also a flight at one altitude may violate any flight AGL restrictions because to be high enough to fly at the top of the mountain, you’ve almost certainly exceeded the AGL limits for the lowest part of the terrain. We have found that rotorcraft flight plans work better as a set of vertical planes.
  • If you are planning to conduct structural inspection missions, you will most likely need to fly with 3-10m of the structure. Be aware that this creates wind effects and can interfere with GPS and wireless connectivity. Also, our research with civil engineers indicates that no matter how much video or photos we try to take, having a specialist who knows exactly what to look for is critical.
  • Expect engineers and structural specialists to use the raw images. Our studies at Disaster City indicate that orthomosaics do not accurately show straight edges on buildings and have a slight bit of ghosting, regardless if from fixed or rotorcraft.
  • Be aware that the country may have a temporary flight restriction in order to protect manned helicopters working at low altitudes in the area- that applies to anything that flies, there are no hobbyist exemptions. The normal procedure is for ANY aircraft, manned or unmanned, to coordinate with the air traffic control so that the manned helicopters can continue to operate. Regardless, manned systems cannot see small UAVs and thus cannot avoid. Should they see a small UAV operating and are not briefed, they typically have to return to base because of the possibility of a collision. Sending someone to the Air Branch of the incident command can go a long way to making sure ad hoc flights don’t accidentally interrupt other activities.

 

Nepal: disaster robots and earthquakes- history and uses

Our thoughts and prayers to the victims, families, and responders in Nepal where CNN is reporting over 777 people killed. Here is some information about how disaster robots have and can be used.

Uses: the primary use of disaster robots in 8 previous earthquakes have been to give authorities and experts rapid understanding of the damage and general situation, the state of the infrastructure – especially underwater portions of bridges and ports which is key for transportation of responders and supplies, and the state of building collapses- especially where there is the indication of survivors or where the building must be inspected by experts but it appears to be too unsafe to inspect.

History: Small unmanned systems have been reported for use in response and mitigation of 8 earthquakes. The first reported use in 2004- an experimental ground robot from IRS exploring a house in the Niigati Chuestsu (Japan) earthquake with the Japanese equivalent of FEMA. In 2009 a small UAV was deployed in Italy by La Sapienza with the Italian Fire Department for the L’aquila Earthquake. At the Haiti Earthquake in 2010, the Navy MSDU used underwater ROVs to clear the port so that ships could bring in responders and supplies without running aground or collapsing the piers. The Haitian airspace was under a temporary flight restriction but there was a  small UAV that self-deployed and performed reconnaissance. A small UAV was tried for indoor inspection of the cathedral at the Christchurch earthquake (2011) but the structural specialists shifted to ground robots.  Underwater robots were used extensively by municipalities with some use of the ground robots and small UAVs for structural inspection at the Tohoku earthquake/tsunami in 2011.  Small ground and UAVs were used by the NifTI team with the Italian Fire Department at the Finale Emilia earthquake for structural inspection in 2012. The Chinese military used small UAVs at the 2013 Lushan China earthquake and the 2014 Yunnan China earthquake for rapid reconnaissance of hard to reach areas.

 

 

Researchers make a robot with other bots

Engineers have created a modular spider robot that has snake bots as feet. It could be reconfigured to meet almost any user requirement and could also be useful in search and rescue operations. Carnegie Mellon University’s latest robot is called Snake Monster, however, with six legs, it looks more like an insect than a snake. But it really doesn’t matter what you call it, says its inventor, Howie Choset – the whole point of the project is to make modular robots that can easily be reconfigured to meet a user’s needs.

The Defense Advanced Research Projects Agency sponsored this work through its Maximum Mobility and Manipulation (M3) program, which focuses on ways to design and build robots more rapidly and enhance their ability to manipulate objects and move in natural environments. Snake Monster, as well as some of Choset’s other robots, will be demonstrated at the finals of the DARPA Robotics Challenge in June.

Check out more information at punemirror.in