This article was first published in the January 2016 issue of WIRED magazine. Be the first to read WIRED's articles in print before they're posted online, and get your hands on loads of additional content by subscribing online.
Robin Murphy provides a special kind of emergency help: robots. in the wake of 9/11, the director of the Texas-based Center for Robot-Assisted Search and Rescue sent squirrel-sized rescue bots into the rubble of the World Trade Center.
In 2011, she used sonar-equipped remotely operated underwater vehicles to evaluate ports and piers after the Japan tsunami. "They can do things that people can't do," says Murphy. "And they can act in situations where humans would be extremely limited."
Because they don't have to worry about safety, robots can achieve a great deal in a short amount of time. When the Oso, Washington, mudslides occurred in 2014, Murphy and her flock of UAVs took seven hours to gather necessary information about the hydrology and geology of the landscape.
Without robots, it would have taken two to four days to compile the same intelligence, at a much lower resolution. "UAVs were able to get accurate data from angles that you couldn't get from a satellite," she explains.
The only downside is the sheer amount of information that can be gathered.
"It's called a data avalanche," says Murphy. "You get more data than you can actually go through."
The next step is using AI to do the work for them instead. During the mudslides, deep-learning programs were used to train computers to go through photographs and identify details that a human would miss, such as pieces of debris big enough for a person to be trapped under. "Having computers able to detect something we didn't and to specify it within 20 minutes, not two years -- that's really exciting," says Murphy.
This article was originally published by WIRED UK
