Unmanned “Unplugged”

Robin Murphy, Texas A&M

Robin R. Murphy is the Raytheon Professor of Computer Science and Engineering at Texas A&M University and directs the Center for Robot-Assisted Search and Rescue (CRASAR). She has used robotics at disaster sites around the country, including the 9/11 World Trade Center attacks, Hurricanes Katrina and Charley, and the Crandall Canyon Utah mine collapse. She is also a founder of Roboticists Without Borders.

You’re currently involved in diverse array of robotics-related work at Texas A&M. Can you tell us a little about what you do?

I have the best job in the world. I am a professor, so I teach, conduct research, and I am the director of two centers: the Center for Robot-Assisted Search and Rescue (CRASAR) and the Center for Emergency Informatics (CEI). I am involved with the CEI because the purpose of the robots is information. I get to work with more than 50 faculty involved in computing, engineering, geosciences, social sciences, and policy to help get the right information to the right person in time to save lives and accelerate economic recovery. 

You’ve been at this a while, too. In 2001, you founded the Center for Robotic-Assisted Search and Rescue (CRASAR) to prove that UAS could be a key tool of search-and-rescue operations in 2001 and became one of the first pioneers in the field. What led you to pursue such a path?

I started researching rescue robotics in 1995 after the Oklahoma City bombings, because it seemed like the hardware and intelligence for small planetary rovers being designed for Mars would have been useful — big heavy bomb squad robots were at the bombing but were too heavy to use without risking causing the structure to collapse further. John Blitch founded the CRASAR in 2001 just before 9/11 to transition the small ground robots from his DARPA Tactical Mobile Robots program to search and rescue, and I took over the organization in 2002. During Hurricane Charley it was clear that advances in small UAS from the DARPA Micro Air Vehicle program and similar DOD investments would be useful. Plus the damage to bridges, marinas and sea walls at Hurricane Charley showed that unmanned marine vehicles were needed.

At CRASAR, you focus on creating a “community of practice.” What does that mean in practical terms?

This means bringing together practitioners from the Texas A&M Engineering Extension Service, which trains more than 240,000 emergency professionals each year, with industry and academia. Industry and academia need to understand what the responders need. Responders need to see what’s possible and shape the development so that technology can be quickly adopted.

You focus on a combination of remote control and artificial intelligence and even teach Intro to AI Robotics, a course on the subject that is hosted on iTunes. In what way can UAS govern themselves?

All of the small UAS I’ve worked with in the past few years have been self-governing in terms of flight control. Many have autonomous waypoint navigation using GPS, return to home and takeoff and landing. Some are able to visually avoid obstacles and pick out safe landing zones. I’d like to see small UAS have more onboard health diagnostics, computer vision and decision aids.

You also have a new book, “Disaster Robotics” (Intelligent Robotics and Autonomous Agents series), where you study the many roles that unmanned systems can play in places that are difficult for human beings to reach. What are some of the unique ways unmanned systems are being used?

There’s too many to list. But one of the biggest surprises to audiences when I give talks is the use of marine robots for disasters. Unmanned marine vehicles have been at almost as many disasters as UAS, which makes sense when you think about how much of the world’s population lives near water. That means critical transportation and energy infrastructure is underwater. The underwater portion of bridges have to be inspected, ports and shipping channels reopened, pipelines repaired, and so on after an earthquake, tsunami or hurricane. Sending manual divers down is slow, expensive and risky — robots can add great value. 

As someone who has been in this field a while, could you tell us in your opinion how the use of UAS in search and rescue has changed?

One change is how rapidly UAS are adopted for missions. The Gulf Coast states now routinely fly UAS in the two days before a hurricane to get “before” imagery. Florida had their National Guard fly a UAS when Hurricane Rita was aiming for the keys in 2005. I think all the states did a collective “Wow, that’s a good idea!” 

As UAS technology continues to develop, how do you foresee it being used by researchers in the future? 

I think small UAS technology will continue to evolve for near-Earth applications where they fly below the tree line and near structures and even in interiors. GPS is unreliable near structures and there is wind shear, so this is particularly challenging for guidance, navigation and control. 

What does that future hold for you and UAS? Do you have any other exciting research projects planned? What would you like to do next?

There are so many directions. My group is primarily focusing on the human-robot interaction. Two of my students, Zack Henkel and Jesus Suarez, are finishing up our Skywriter — a more naturalistic interface that allows decision-makers to draw on the video feed in real time where they would like the UAS to go, take pictures, etc. Another student, Brittany Duncan, is programming small UAS for evacuating crowds. I continue to focus on how to make the small UAS smart enough that it is easy for many responders located through the disaster or off site to simultaneously get the information they need in real time.