Can We Detect Small Drones Like the One That Crashed at White House? Yes, We Can

null
Photo: U.S. Secret Service
The DJI Phantom drone that crashed on a White House lawn last week.

Last week, a small drone belonging to a federal worker crashed on the White House grounds. Just what caused the little flyer to head off to one of the most security-sensitive sites in the world is not certain, but in any event the results were pretty harmless. Still, the incident sparked much interest in the White House’s aerial defenses should someone want to use a small drone of this kind to do real mischief.

The New York Times reported that the ill-fated drone, a DJI Phantom, was “too small and flying too low to be detected by radar,” according to government officials. So how might the U.S. Secret Service—or others worried about drone incursions to the properties they oversee—detect them? I contacted T. Adam Kelly, the CTO of DeTect, a specialty radar company based in Panama City, Fla., to discuss this issue.

Kelly told me immediately that the notion that a DJI Phantom is too small to detect with radar is, well, not exactly correct. Sure, you can’t detect one with ordinary radar, but Kelly’s company, among others, has been making systems capable of doing just that. The tricky part, he says, is not so much in sensing the subtle radar returns but in distinguishing a small drone from the many birds that your radar will also pick up.

“You’d just be floored by how many birds there are around,” says Kelly. “And you’re looking for this one incident in this mass of activity.”

Despite the difficulty, Kelly says it’s indeed possible to distinguish small drones from birds, by measuring the motion of the targets and other subtle aspects of each radar return and then applying machine intelligence to the problem. That is, you can automate the process of picking out any targets that don’t match what you’d expect for a bird or other source of radar clutter. Kelly’s company has done this, for example, for a client in Spain who needed to monitor what was in the air at a UAV-testing site.

Using radar is not the only strategy for keeping tabs on small drones, though. DroneShield, a company based in Washington, D.C., offers a system that detects nearby drones by the distinctive sounds they make. The problem with that approach, Kelly points out, is that wind noise can make it very difficult for the microphones to do their job. And, I would surmise, such a system wouldn’t be able to detect an incoming fixed-wing drone programmed to dive silently toward its target while unpowered—in the style of an early V1, say.

There’s also the possibility of detecting small drones by virtue of the radio signals that at least some of them (or their controllers) produce. That won’t help if the drone has been preprogrammed to fly to its destination without radio uplinks or downlinks. But it could, presumably, help thwart some worrisome scenarios. Defenders might even be able to stop a malicious drone by jamming its controllers’ signals. “That’s technically illegal,” says Kelly. “But the rules might be different if you’re defending the Commander-in-Chief.”

Advertisement

Automaton

IEEE Spectrum's award-winning robotics blog, featuring news, articles, and videos on robots, humanoids, drones, automation, artificial intelligence, and more.
Contact us:  e.guizzo@ieee.org

Editor
Erico Guizzo
 
Senior Writer
Evan Ackerman
 
 
Contributor
Jason Falconer
Contributor
Angelica Lim
 

Newsletter Sign Up

Sign up for the Automaton newsletter and get biweekly updates about robotics, automation, and AI, all delivered directly to your inbox.

Advertisement