Drones are growing up

On the first day of class in Malawi, the students at the African Drone and Data Academy went around the room and explained why they were there. Kevin Kochersberger, an associate professor of mechanical engineering, developed the UNICEF-funded program to help students learn to use drones for humanitarian, development, and commercial purposes. He remembers being struck by how personal their motivations were.

“The answers were all like, ‘Cholera is a big problem in my community and I want to learn how to manage data that would help me predict where the next outbreak is going to be,’” he said. “Or ‘Malaria is a problem, and we want to know how to control malaria by looking at breeding grounds.’”

Another student hoped that drones could improve vaccine access for his remote community. One, who grew up on a farm, wanted to use drones to collect data that could help mitigate the impacts of pests and drought.  

Many of the 165 students who have graduated from the academy so far have already gone on to careers with drone companies. One is working with the government of Sierra Leone to set up another drone academy there. But the range of aspirations invoked that first day encapsulated a truism about drones: they can do a lot. 

Unmanned aircraft can cover more ground more quickly than a terrestrial vehicle, are indifferent to challenging terrain, and can carry advanced sensors that disgorge astonishingly detailed data. This makes them really good at doing certain kinds of jobs faster, and sometimes more safely, than people can, and enabling better decision-making by supplying better information. 

Faster work and better information are universally appealing. And over the past decade, as drones have gotten better at providing both, they’ve evolved from a niche technology mostly interesting to hobbyists, the military, and specialized researchers to a tool popping up in almost every STEM department on college campuses. At Virginia Tech, applications are in development for everything from helping search and rescue teams find a missing person to feeding more accurate weather predictions.  

But the heady promise of virtually endless potential applications belies the staggering complexity of realizing them. In 2016, when the FAA released rules that cautiously nosed open the door to a set of commercial applications with tight limits, they predicted that 600,000 commercial drones would be flying within a year. Five years later, the U.S. still hasn’t reached that number.  

One reason is that the exponential growth in drones’ individual capabilities has far outpaced the maturity of technology that will allow them to safely share the airspace with each other and with manned aircraft — especially in the U.S., whose skies are the most congested in the world. 

Fundamental questions don’t have answers yet: How can we manage traffic when multiple drones are vying for the same airspace a few hundred feet above our heads? How can a drone, with no pilot in the cockpit to scan the sky, reliably detect manned aircraft nearby and safely stay out of their way? How do we know when autonomous systems are trustworthy? When new services bring drones routinely into neighborhoods, how will the public respond? Until we have answers to these questions, and other equally complex ones, only a tiny fraction of drones’ potential will make it past the research phase. The rest are penned up behind a dam of challenges the industry doesn’t have solutions for yet. 

Several years past a watershed of rapid development, progress is inching forward in a slow, halting ballet — some new technology enables more permissive regulations; other new technology enables new applications that require more permissive regulations; new regulations demand new industry standards; technology to meet those standards needs to be developed and tested. Technology, rules, and standards are all interdependent, and all evolving at the same time. But if we can get the balance right, solve the problem of successfully working drones into the fabric of aviation, they can help us solve a lot of other problems.

Virginia Tech has become a hub for all these intersecting threads of activity. Researchers are constantly devising new applications for drones and building and testing the technological framework that can allow them to play nicely in the airspace. Specialists at Agricultural Research and Extension Centers are studying how aerial imagery can boost productivity on Virginia’s farms. The research coming out of the university’s FAA-designated drone test site is informing new regulations and industry standards. The first commercial drone-delivery program in the country, launched in collaboration with the test site, is humming along in the next town over. An 85-foot-tall drone park is an outdoor classroom for first-year engineering students and a proving ground for advanced cybersecurity research. Teachers are taking advantage of all these resources to use drones as a springboard for reimagining what undergraduate education should look like. For students, drones are a tactile platform for learning, a piece of technology they can hold in their hands, experiment with, and use as a blank slate for their ideas. Through all of this, drones are expanding our ideas about what it means for humans to collaborate with robots. 

And despite the technological hurdles and regulatory challenges throttling real-world commercial applications, new ideas are blossoming everywhere, and small-scale trials hint at much grander impacts. 

It’s hard to see what’s over the horizon. Thomas Pingel, an associate professor of geography in the College of Natural Resources and Environment, uses drone imagery to knit together stunningly precise maps — maps that can help reveal the strength and magnitude of a tornado that ripped through a forest, or build an immersive virtual-reality double of a city, or measure, within millimeters, the shifting sands of a beach. 

“It’s like getting to look through a microscope for the first time,” he said. “We don’t quite know what we’re going to see yet.”