The news stirred a lot of debate about nuclear plant security and highlighted the bold fact that no one had been able to prevent or trace these repetitive intrusions.
The fact that the French Gendarmerie may be allowed in the future to shoot down such trespassing drones (using firearms) may not be a deterrent against a swarm of malevolent drones, and it could cause collateral damage. Securing large industrial sites against non-authorized drones is still an issue, but drone geo-fencing countermeasures are already in the making.
Just a few weeks after these events, during Electronica, I stopped by the IMST GmbH booth, where a radar-bearing drone was exhibited. Siegfried Schulze, responsible for business development at the research-driven industrial engineering and design house, told me that an increasing amount of research was taking place to come up with preventive measures against such drone intrusions in established no-fly zones.
Talking about no-fly zones, if you buy a small drone, a notice is usually included in the box warning that it's pretty much illegal to fly it anywhere in populated areas or anywhere you could put someone at risk. (The definition is as broad as it is vague, except for airport routes or specific flying zones mapped in aeronautical charts.)
But this in itself won't prevent any malicious flights, and judging by the number of camera drones that are offered to consumers and pro users alike to map and record their surroundings, it is inevitable that more drones will venture beyond the strictly private boundaries of a garden or a living room.
“For the next months to come, until the industry comes up with drone interception solutions, drones will represent a new potential threat to a lot of industrial sites,” warned Schulze, who leads a research group on special security systems and is active in the German UAV DACH working group. According to Schulze, there will definitely be a market for catcher drones.
So what could be done to intercept a drone before it causes any harm?
“First, you want to identify it, know exactly what drone it is, what radio link it uses, and possibly communicate with it. Then you want to stop it before it trespasses the site's boundaries. Then you want to catch it or take it down,” said Schulze.
“Here we are exhibiting a high-precision 3D radar technology light enough to be used on board drones for anti-collision and flight control, but it could also be used to exactly pinpoint and locate simultaneously other drones, learn about their speed and flight direction.
“We are also developing sensor and radar solutions in order to determine the unique acoustic, thermal, EMI, and optical signatures of any given drone, because you don't want to raise false alarms every time a bird flies over your site.”
Practically, the idea would be to record these multi-domain signatures for all known commercial drones and have their unique signatures ready in a database for quick profile matching and 100% physical identification.
Such detection and identification units (including the radar, a video camera, and multiple sensors) could be pole- or wall-mounted at a site's periphery. Then, upon detection of a nearby drone, the actual interception step could require a specialized drone capable of getting close enough to jam the intruder's GPS signal.
“As a failsafe feature, when a drone's central unit no longer receives any radio commands or GPS signal, it hovers on site until it recovers its communication link or until it drops dead short of battery juice,” says Schulze, “so by knowing what drone we are dealing with, we could launch the appropriate jamming scenario to freeze the drone in the air.”
No need to hack into the communication link or try to establish a dialogue with the pilot for that, though you could imagine a small flag popping up saying, “Your drone is under arrest” for the video record. By knowing exactly what commercial drone is being used (and having studied its communication link beforehand), one could even try to hack into the system to take over flight control.
Catching the drone or taking it down is more problematic, Schulze admits. Simply taking the drone down could create accidents if it were hovering over high-voltage installations or chemical plants.
“We would have to build an intelligent mapping system so the interceptor drone would know exactly where it is, inside or outside the industrial estate, and if there are any roads or populated areas nearby.”
In some cases, you could just wait for the trespassing drone to drop short of battery power (typically a few minutes to around half an hour for most hobbyist drones), but for better control, one envisaged solution could be to equip the catcher drone with a releasable net, tied to a small parachute equipped with flashing LEDs for people on the ground.
“A tilt-wing drone like the prototype being developed by RWTH Aachen University for the AVIGLE European research project could be a good start,” Schulze told us when we asked what type of drone would be suited to the task.
The aim of the project is to come up with an autonomous flying service platform for operations within a swarm (so with anti-collision and communication capabilities).
The tilt-wing allows for vertical takeoff, with the hovering qualities of a rotorcraft. It also allows a progressive transition to a more energy-efficient horizontal flight mode like that of a fixed-wing airplane. Specs of the AVIGLE project include speed capabilities ranging from 0 to 40 m/s (144 km/hour), a maximum flight time of one hour, and a payload capability of 1.5 kg.
To read the rest of this article, visit EBN sister site EE Times.