Law enforcement and military personnel might finally have a way to track malicious drones and prevent millions of dollars in damage thanks to new artificial intelligence research. Academics at Israel’s Ben-Gurion University of the Negev have developed a way to locate the operator of a drone by looking at how the airborne vehicle moves.
Locating the pilots of malicious drones is a pressing issue. In December 2018, Gatwick Airport had to close its runways to avoid drones flying dangerously close. Officers believed that it was a deliberate attack on the airport. The same thing happened at Heathrow Airport just a few weeks later.
While drones are relatively easy to spot, it is a lot harder to pinpoint their pilots. Although technicians can try to locate them by monitoring radio signals, they must be relatively near the drone to do so, and operators can cloak their transmissions.
The Ben-Gurion research team worked on the premise that drones behave differently depending on where the pilots are. By tracking the drone’s path in the sky, they were able to analyze the reactions of the pilots to external stimuli such as sun dazzle and obstructions.
The team simulated drone flights using a software simulator, logging the path of drones across 81 simulated flights from three operator locations. It then ran this data through a machine learning algorithm and was able to guess the viewpoint of the operator with 73% accuracy, it said.
The researchers said that this drone-watching technique could be paired with traditional radio frequency scannung techniques. RF scanners have trouble identifying a signal related to a specific drone in a dense area where there might be other, similar signals, it said. “We can train our neural networks to identify command patterns of the signal transmitted from the operator when the drone is turning, rotating, accelerating, and decelerating and use it to connect to signal to a specific drone in the air,” it said.
By detecting the direction of an operator, defenders could also use techniques to obstruct their line of sight, it said.
The technique has applications beyond watching drones, the paper said. It could also be used to identify drivers by looking at behavior in different traffic situations, including how they use the pedals and the steering wheel, and how much distance they keep from the other cars.
Other researchers at Ben-Gurion Univeristy have also been working on anti-drone technology. In March, Prof. Amiel Ishaaya at the University’s School of Electrical and Computer Engineering revealed a laser-based defense system called Light Blade that will be able to down the next generation of attack drones.