My experience with drone detection goes back to summer 2020, when I was part of tests run by DFS, Fraport and Munich Airport. We invited six drone detection system companies to deploy their technologies at either Frankfurt or Munich, and we had drones flying at and around the airports to test their performance. Some systems were able to detect standard, off-the-shelf drones, but when we used custom-built drones, using non-standard frequencies, the only usable detection tool was radar. At that time, the range of such radar was quite limited, the number of false detections was high and they also missed many drones.
One of the conclusions was that a human operator would have to monitor the drone detection systems’ screen to decide if a detection is posing a real threat or not. You can imagine the level of cost that such a solution would require, both from a technical and manpower point of view.
I was discussing this with counter-uncrewed aerial systems (C-UAS) and airspace safety specialist Philip Butterworth-Hayes recently, who explained that things have changed a lot recently, particularly because of the war in Ukraine.
Question: How has the picture changed since 2020 and is detection better nowadays for off-the-shelf drones?
It is certainly true that for many years the C-UAS industry was notorious for a “fake-it-till-you-make-it” culture, with many companies claiming all sorts of performance capabilities for their systems, but only rarely managing to achieve them. All that changed on 24 February, 2022. Russia’s invasion of Ukraine sparked the outbreak of a drone war where technology evolved at an unprecedented rate and the weaknesses of C-UAS system capabilities – in terms of technology, resilience and affordability – were very quickly exposed. Now we have a clear understanding of what works and what doesn’t. In general terms, the gold standard is a layered defence system combining acoustic, optic and radar sensors, networked within a command-and-control system which identifies an optimal mitigation system from a range of possibilities, including intercept drones, missiles, guns, directed energy and spoofing/jamming/cyber takeover measures. This is expensive and not very mobile but it works. In the non military arena, over 99% of all drone threats come from commercially available drones or drones with commercial command-and-control communications components. We now have a range of highly effective RF-based detection and mitigation solutions available, as well as C-UAS performance measurement systems, such as Interpol’s Project Courageous, which gives an objective view of real system capabilities.
Question: What about more advanced, custom-made drones, or drones that fly without using any radio-frequency communication?
Outside of the battlefields of Ukraine and the Middle East these are extremely rare. It is true that fibre optic cabled FPV Russian drones which do not emit RF signatures have caused huge problems this year for Ukrainian defenders and about 20-30% of the drones used by Russian forces are these types of drones. Ukrainian soldiers have had to develop a range of tactical mitigation measures, such as mobile radar systems, fields of barbed wire (targeting the cables that float down to the ground), specially-designed shotguns and anti-drone rifles, and nets, to deal with the threat. Over the horizon is the looming prospect of AI-enabled drones without any RD signature or cable. But these require huge amounts of government investment and will not be available to non-military personnel for many years to come.
Question: Are there solutions today that can operate fully automatically, or is it still necessary to have a human in the loop to decide if a detection represents a threat? What about the rate of false positives?
We are quite close, perhaps six months away – although some companies say they are already there – from the first fully AI-enabled C-UAS layered network which can distinguish between drones, loitering munitions and missiles all fired together in a swarm. In this context you need a C-UAS system with a three/four second engagement rate per target (requiring a 100kW laser for example) before moving to the next one. This is where AI comes in. We are also starting to see AI applied to classifying drone threats in the non-military domain. What type of drone is it? Does it have a payload? Is it flying towards a target? This makes the intervene/non-intervene decision a lot easier for the human defender.
Through AI and improved software – plus more powerful databases which log commercial drone and component identification information – we have dramatically reduced the amount of false positives in the last three years.
Question: Do you see an impact of the war in Ukraine on the speed of technological development?
In the drone war above Ukraine there is an evolutionary cycle with a predictable end-state: on one side, AI enabled drone swarms exploiting AI and quantum technologies for navigation, targeting and communications while on the other, AI-enabled C-UAS direct energy weapon systems which can affordably shoot drone swarms out of the sky. Affordability is key. With 10,000 drone engagements a month any effective mitigation must be cheap and fast. The latest trend is the emergence of a new generation of fast, cheap and expendable intercept drones from Ukrainian companies – and the export of this technology in terms of industrial partnerships, to Western Europe and North America.
We have just published the Unmanned Airspace 2025 Global Counter-UAS Systems Directory which itemises over 1,000 C-UAS systems, including the latest Ukrainian solutions, with performance details, sales and the partnership arrangements being developed to fill in any C-UAS defensive gaps. It is updated every month and broken down into niche sub-sectors (net-capture, missiles, intercept drones, detectors, etc) to give C-UAS procurement personnel and industry marketing executives a clear view of what capabilities are now available.
If you are interested in the Counter-UAS Systems Directory, mention Global Airspace Radar when you order it to get a 10% discount on it.
Question: You are also covering countermeasures, which is not really an air traffic management issue, but can you share some insights about the trends, especially how a drone intruding the airspace above an airport can be neutralised without the neutralisation process being a threat to people and assets?
I would argue that C-UAS is very much an air traffic management issue, especially in the lower airspace. The key technical challenge is that normal RF spoofing and jamming mitigation measures cannot be used around a communications-rich environment like an airport. But there are proven technologies such as cyber takeover, net capture and (soon) laser weapons which, when allied to effective detector systems, can identify, track and deal with rogue drone incursions without disturbing the surrounding traffic. This is an area of considerable research but as the technology improves it will require the legal frameworks to evolve in parallel so, in the non-military world, the organisation which is empowered to detect the rogue drone is the same organisation which can take it safely out of the sky.
