In 25 years’ time, when a disaster strikes on the news, we might gauge how well it’s being handled not by the aid crews en route–but by the number of robots on the ground. And if so, we’ll have Robin Murphy to thank: Along with a handful of collaborators, she’s a pioneer in developing robots that deliver aid to humans. A professor at Texas A&M, her creations have ranged from tiny helicopters and boats, to “caterpillars” that wriggle through rubble to find survivors.
Murphy’s founding flash of insight came in the wake of the Oklahoma City Bombing and the Kobe Earthquake, both of which happened when she was a young assistant professor; she realized that the advanced work in AI and robotics being done for a Mars lander might be better applied here on earth. Her boss, the department head, promptly told her she was committing career suicide. “He wanted me to focus on basic research,” Murphy says. “He told me not to worry about applications.”
But she did, and entire disciplines have now caught on: Where the focus used to be on pure academic silos, researchers now trumpet their “interdisciplinary” approaches. “If you’re trying to make a robot, you get a beautiful Gordian knot of problems,” she says. “You can’t work on one without understanding the other. If you’re doing emergency response, you need sensors. If you have sensors, you need compression algorithms. And then you have to understand how people respond to stress.”
Which is exactly what she’s working on now: “Survivor Buddy,” a robot that delivers two-way audio and video, linking survivors with a rescue crew. With Cliff Nass, a professor at Stanford, she’s working on giving the robot true body language. “We realized pretty quickly that robots are creepy because they violate social conventions,” Murphy says. “But you can program them in ways that are calming, if you understanding things like how to make them seem like they’re listening, how far a distance is comfortable, and how to approach someone that’s unconscious.”