What motivated you to get into rescue robotics?
The in 1995. At the time, robots were being made very small and agile, but rescue robots were golf-cart-sized devices or things developed for the nuclear industry that weighed tonnes. They could only go on top of the rubble, not into it. But there might be people trapped deep inside that you could save if you could just get to them within 48 hours. Small robots, the size of a suitcase or lunch box, would clearly have been of benefit.
Two decades on, what’s your approach?
Advertisement
At the , our programme finds robot-makers willing to donate their robots and expertise to us. We essentially audition the robots, large or small, for a spot on our team. We develop potential uses for the machines we select, making them smarter and adaptable for a variety of disaster scenarios. Right now we have about 20 types of ground, air and sea robots, many at my lab in storage cases, ready for deployment when something bad happens.
Have you deployed a rescue robot recently?
Yes. When we heard about refugees drowning off the coast of Greece we called Tony Mulligan, CEO of a company called Hydronalix, which makes remote-controlled . He said, “I will take two experts and two EMILYs to Greece.” I met him out there in January. He , one to the Hellenic Coast Guard and one to the Hellenic Red Cross.
Why is it dangerous for refugees to land boats on Greek shores?
They have bad motors and bad boats that are overloaded, and the wind conditions in that channel can suddenly change. The shores are also very rocky. If you find a good beach, great, but if you hit a bad stretch of shore, there’s no way to get off the boat or for anyone to come and rescue you from the rocks, so you are trapped. Roughly 600 people died last year.
How does a marine robot help?
One way is to get the refugee boats to follow an EMILY to a better shore. These robots have cameras and two-way radios to allow rescue boats to communicate with the people in peril. The robot can also deploy a flotation device if the boat goes down. Finally, if a boat gets stuck on a rocky piece of coast, EMILY can run a line out so they can clip it in tight around their boat and let one of the bigger boats pull them back out to safety – we saw that a lot. In short, we are preventing drownings.
Are you working on any developments to these lifesaving robots?
Yes, more automation. EMILYs are remote-controlled, but what we would like to be able to do is have a lifeguard look at the video feed from EMILY’s onboard camera and say, “Over there, that’s the cluster of people I want you to assist”, just by circling on the screen.
What other rescue situations have you attended with robots?
We helped after mudslides engulfed a rural community in Oso, Washington, killing more than 40 people in 2014. In one day we went out and flew a reconnaissance drone over the mudslides to get high-resolution images from angles that crewed helicopters and satellites could not. We flew above the area for 48 minutes and processed the data on a laptop while we were driving back to the incident command centre. We were able to give them a high-resolution 3D representation of the entire area fast. As a result, it became obvious to the hydrologist where the river had shifted to, where the lowest points were and where was the best place to create a flood-bypass channel to prevent further catastrophe.
What technological advancements with rescue robots are you most anticipating?
So far everything’s been focused on what robots see, but what they feel is important to us as well, so I’m looking forward to robots with a sense of touch. It’s really hard to clean the dirt off a trapped survivor in order to start using infrared techniques to see their blood vessels and check their pulse. And if I touch them I need haptic feedback so I don’t hurt or scare the bejeezus out of them.
And the newer types of mobility are exciting too, particularly snake-like robots that are able to burrow into rubble, for example. Big robots can only access big spaces, so smaller, burrowing robots are a big deal.
Have your robots saved a life?
No saved life has been directly attributed to one of our robots, but that would be like saying a fire truck or a search camera has saved someone’s life. They are important tools of the teams that saved lives, but at the end of the day it’s the emergency responders who do it. They are the heroes.
What’s holding back the rise of the rescue bots?
Right now the barriers tend to be lack of accurate knowledge about what they can do. That’s why we want them used as much as possible – seeing is believing if you want people to adopt this technology. The second thing is regulations on what government agencies can spend money on.
Aren’t these machines costly?
Most of our robots cost less than a fire truck; less than the fire chief’s SUV. They certainly don’t cost millions of dollars – even if you factor in training people to use them and other associated costs. The important thing is, if you reduce the physical recovery time of a disaster by just a few days you can take years off of the economic recovery time. That’s what robots allow you to do.
You have said it’s unethical not to use rescue robots – why?
We have now reached the point at which these technologies are proven. Not every robot is perfect and not every robot is ready, but in general these are hardened technologies that have been used in 50 disasters in 15 countries since 2001. It’s not like we suddenly came up with something out of nowhere. It was the same scenario with vaccines – at some point you just have to pony up the money and commit to it.
Robin Murphy introduces rescue robots that are ready to roll
Profile
Robin Murphy is director of the Centre for Robot-Assisted Search and Rescue at Texas A&M University in College Station, where she is Raytheon professor of computer science and engineering
This article appeared in print under the headline “Robot rescue? We’re ready to roll”
