Share on twitter Twitter Share on facebook Facebook Share on linkedin LinkedIn Share on reddit Reddit Share on print Print

The pace of change in the civilian unmanned airborne systems marketplace is staggering, and much like the mobile telephone, it will present both several threats and opportunities for defence and security forces.

While unmanned systems’ defence market is well established, the real growth and innovation are in civilian markets. Like many of the other Think Defence ‘long reads’, this is an accumulation of posts and ideas from across the years. The first post on the subject was in 2011, covering the use of a civilian ‘drone’ by Libyan rebels to observe Gadhafi’s forces and direct their own artillery strikes.

I asked;

With this kind of unmanned system, relatively unsophisticated in comparison to a Reaper, of course, defence economics come into play. Systems like these, costing less than £10,000 can be easily obtained on the open market in significant quantities, operated without extensive training or worrying about airspace management, carry day and night sensors and generally provide enemy forces with a big asymmetric advantage.

It doesn’t take an overactive imagination to see how they could be easily weaponised either.

If all we can counter them with is hundreds of thousand pound missiles do we have an operational and economic problem?

Since then, operational use by both conventional and non-conventional forces in Ukraine, Syria and Iraq has demonstrated just how they are being exploited in a defence context. Increasing concerns about their use near airports and nuclear power stations have also elevated the importance of understanding.

In response, several countermeasures have emerged, both conventional and unconventional.

There might be a perception that non-state actors use of commercial unmanned systems means they are unsophisticated. It might even be tempting to think of small commercial UAS as not far removed from toys. Whilst there are many toy-like devices, there are as many that bear no relationship to a toy whatsoever, professional users in the inspection, survey, filmmaking, agriculture, archaeology, emergency response and journalism sectors are driving innovation and improvements in all sorts of interesting directions.

More than Toys

Most drones are relatively fragile, open rotors and lightweight construction make them prone to damage from minor collisions and rough landings. Aftermarket modification kits can provide rotor guards and specialist crash proof designs have also emerged for use in crevasse exploration and search, stop sniggering at the back.

The Flyability Gimbal has also been proposed as a suitable design for search and rescue in urban environments. The Vantage Robotics Snap uses four rotors with integral guards, safety and robustness feature heavily in their marketing materials.

It is easy to envisage lightweight blade guards being mandatory in some markets to prevent injury.

[the_ad id=”53415″]

Most devices are specified to carry a camera payload but there have been many instances of organisations wanting to use them for high-value parcel delivery (Amazon, DHL etc.). Drug delivery in remote locations in Africa has also been seen as a possible use for small UAS. This has resulted in an increase in interest in improving payloads.

These are readily available, there are hundreds of specialist providers developing more cargo and logistics oriented designs.

The real Achilles heel of consumer/prosumer types is their endurance, typically, only 20-30 minutes. Although battery technology is always improving, one solution currently being explored by a number of manufacturers is hydrogen fuel cells. Intelligent Energy, from Loughborough, have developed a hydrogen fuel cell that can extend flight duration to several hours. EnergyOR from Canada, likewise, in fact, EnergyOR have announced years ago the H2Quad that has a 1kg payload and 2-hour endurance.

If communications or autonomy allows, this puts the device into the 80-100km range. If the size is not an issue, larger UAS can carry larger payloads and batteries, 10kg and 1-hour endurance, for example, is not uncommon for the larger devices.

Many systems are available with mission planning software that allows the operator to pre-programme flight paths and waypoints to reduce workload and reliance on skilled operators. Many of these are open-source, available on software repositories and ‘app stores. Third-party mission planning and flight control software is also increasingly available from specialist vendors.

Flight control software is available to circle around an object, move on a virtual wire, object following, full systems monitoring, and flight planning with waypoints, station keeping, and automatic return to launch point, camera control, automatic data export and autonomous search.

For use in areas where availability or visibility of a GPS signal may be in doubt, underneath a bridge or in the shadow of a large building, for example, precise positioning would normally be a problem. Inertial positioning is one option but for other locations, GPS can be augmented with Differential GPS that makes use of publicly available beacons (Trinity House in the UK for example) or using a system like Local Positioning System (LPS) from Novadem. Attacking GPS is often seen as a critical weakness, but only if the system actually uses it.

High definition day/night sensors are a given but thermal imaging, LIDAR and Multi-Spectral Sensors are also available to commercial users. Gimballed EO/IR sensors weighing less than 700g that can stream video imagery at a pixel resolution of 720×576 and IR at 640×480 are now commonplace. LIDAR scanners are used for collision detection and 3D mapping, and multispectral sensors are commercially available. Broadcast and film quality cameras are available from Phase One, RED, Black Magic, Nikon and others. Tamron has recently introduced a camera module for small UAS that includes onboard optical stabilisation, together with full HD output and FLIR have a number of packaged IR systems.

On-board image processing devices from vendors such as Headwall allow multiple sensors and data inputs to be fused.

Common control stations and those that can be used to control multiple platforms are available, and how about a smartwatch control system, now available. Control systems typically require line of sight and this naturally reduces utility, waypoint navigation can be used and payload systems are exploring alternatives to GPS and Wi-Fi. Although regulatory conditions normally specify a line of sight as being the distance limit there are some professional applications where this is being exceeded. Mesh network technologies have also started to appear in unmanned systems for robust and non-line of sight communications.

This is one of the main focus areas of current research for obvious reasons, either for low speed or high-speed operations. Laser, ultrasonic and even synthetic vision using a Microsoft Kinect sensor have all been implemented with the objective of allowing them to fly in close formation, inside buildings and urban areas. The videos below show systems using onboard automatic obstacle avoidance, human operators are not making the decisions.

Swarming technology is also in receipt of a great deal of interest in both civilian and military markets.

Manufacturers have also started to specialise in niche markets, mapping, survey and mining for example. senseFly can produce hyper-accurate geo-referenced digital imagery using autonomous unmanned aircraft. The post-flight analysis is carried out to add reference points, ground information and stitch the imagery together into an ortho-mosaic image. It can then be exported in a number of formats suitable for inclusion in mapping and GIS systems such as MapBox.

As can be seen from above, extremely sophisticated systems are now readily commercially available. There is no need for permission from governments, just a credit card, they are much more than toys.


The most common threat seems to be mostly non-malicious, careless use of commercial UAS near airports or groups of people for example. Terrorist threats might seem far-fetched but the media attention given to ‘drones’ plays into their potential use against civilians. Most systems cannot lift heavy payloads but in many regards, they don’t need to.

A small explosive charge surrounded by nuts and bolts can be devastatingly effective in the right circumstances. From a terror and societal disruption perspective, imagine what would happen if that explosive were replaced with a bag of flour and a warning about anthrax, and the drone was flown into Wembley stadium. It is not beyond the realm of the impossible that these systems could be obtained by terrorists and weaponised with alternative payloads.

In a battlefield setting, we have seen hundreds of examples of the use of commercial UAS. Ten years ago when I wrote about their use in Libya, it was limited to simple observation and fall of shot adjustment.

Since then, from the Phillipines to Syria, civilian UAS have been utilised in a variety of ways, from direct attack using freefall munitions to coordinating complex ground assaults using suicide vehicles and dismounted infantry.

Although Russian forces reported successfully repelling it, in 2018, they were subject to a non-line of sight swarm attack.

What makes this such a threat has almost nothing to do with their capabilities or lack thereof, it is their potential for ubiquity driven by mass-market cost reduction. Sophisticated military unmanned systems are expensive and it is this that puts them out of reach of many of our potential enemies but when those potential enemies can buy one from Amazon for hundreds of dollars then the specification difference between proper military systems and remote control toys becomes of decreasing relevance, their very lack of sophistication and low cost becomes the problem because it will drive us to counter with increasingly expensive measures.

3D printing designs shared online and open-source control software means that the technology cannot be contained and given this, we can also make a case that innovation cycles will turn over faster in the commercial market than in the military. Current commercial offerings are designed around the carriage of a video or still camera payload and the means to transmit to a control station, roughly the same weight as a hand grenade or two.

With a bit of imagination, one could see an enterprising and imaginative enemy using 50 of these, each carrying an explosive and bag of nails to simply fly to a pre-programmed pattern above a military airfield and destroy a handful of very expensive but very fragile Typhoons or F35’s.

Impossible you say, except the software is freely available online that enables swarming behaviour and formation flying, not remotely piloted. The enemy would not need 50 pilots as each would formate against a pre-arranged pattern to maximise fragmentation coverage and all it would take is a large commercial van to carry them.

A poor man’s cluster bomb.

The Taliban attacked Camp Bastion and destroyed six per cent of the USMC’s Harrier inventory, they were housed in fabric shade shelters.

Threats are increased by swarming systems. The relative economic exchange rate is significant, buy a hundred, arm them with a grenade, pre-programmed to fly over a military airfield and drop their grenades in a pattern to maximise fragmentation coverage and you have a very effective means of putting very expensive combat aircraft and helicopters out of action, maybe not permanently but certainly good enough to do a denial job.

A command-detonated Claymore mine weighs less than 2Kg but has an effective range of approximately 50m. Although it is an anti-personnel system, the effect on light vehicles, helicopters or aircraft would be equally devastating. All far-fetched, perhaps, but consider one of the fastest-growing areas for civilian drone systems is for package delivery. Package delivery might be for something from Amazon or an emergency flotation device for a swimmer in difficulty, both have been demonstrated. Substitute the parcel for an anti-personnel mine or unintended ground sensor, all of a sudden, it seems less far-fetched.

There is no need to improvise package release mechanisms, they are being sold as accessories.

More conventional defence manufacturers are also exploiting commercial UAS.

Defence manufacturers have been exploiting civilian UAS technology for many years but this is outside the scope of this document to some extent, interesting to note that military forces are exploiting civilian technology now, not the other way around.

All the innovation is being driven from the civilian market.

The battlefield threat is twofold, their use in intelligence gathering today and the potential weaponisation, both are a reality now.



Recognising there is a threat is the first step, after a slow start, clearly, this has happened.

The first counter, obviously from a non-battlefield view, is regulation

Many of the safety security domain challenges are likely to be met by regulatory approaches, not all security threats come from those with ill intent. The regulatory agencies have probably been somewhat behind the curve, as might be expected in an industry where increasing innovation and decreasing cost lead to widespread adoption. There has been somewhat of a ‘Wild West’ feeling in some parts of the market as regulators, insurers and user groups struggle to coalesce around a single sensible and supportable approach.

Regulation is also being increasingly used with technology solutions.

Most of the larger manufacturers have software that includes ‘geofencing capabilities that ensure the system cannot be piloted into a ‘no fly’ zone. No-fly zones are maintained in real-time, in the USA for example, by a company called AirMap. The software also ensures altitude limits are no exceeded. For responsible users, this provides a great deal of assurance, they cannot inadvertently fly their aircraft in restricted airspace. With real-time updates available, even temporary restrictions such as those over a forest fire can be incorporated.

Some of the most typical causes of accidents that might cause a security issue such as a low battery or lost communications are also being increasingly addressed with software, automatic landing upon a defined battery threshold or ‘return home’ on lost communications, for example. More advanced collision avoidance systems are also increasingly being deployed, especially as large technology organisations enter the market. 

Through a combination of registration, policing guidelines, regulation and technology, together with sensible and lawful operator communities and trade bodies, the majority of non-malicious safety and security threats should be gradually addressed, allowing malicious threats to be the focus, or in a critical location (e.g. airport) safety context, catching the idiots.

Whilst small commercial UAS can be difficult to detect, they are not impossible to detect. The four main methods are optical/thermal, acoustic, passive RF and radar.

If the system is being actively controlled or transmitting video imagery, it will be radiating RF energy that can be detected and triangulated. For fixed locations, this can be relatively simple to install, but more difficult for moving areas or temporary locations. Acoustic detection uses signature recognition, each drone creates a unique noise. This may not work well in cluttered environments like urban areas, and at extended distances. DroneShield and Alsok are providers of this type of technology.

Thermal and visual detection can use the UAS movement and contrast against a static background. With advanced scene change algorithms, this can provide highly effective detection but again, potentially limited in cluttered environments. Small UAS tend to be battery-powered and use electric motors, neither well characterised by high thermal signatures. Radar is the most widely used, but small UAS are still difficult targets to detect, and there is the ever-present problem of bird rejection, Blighter and Kelvin Hughes, both from the UK, have specialist small object detection radars.

The Blighter A400 radar has a published detection range for compact UAV of 6km. The Elvira radar from Dutch company Robin Radar is another good example.

In Ukraine for example, one tactic used was to avoid attacking the airborne system but instead, track it back to the recovery point and attack the area with indirect fire. The operators, in this case, were far more valuable targets than the systems. Tracking the device back to its recovery point can also be used in a law enforcement context.

Means of defeat can be physical or electronic in nature.

Blowing UAS out of the sky with anti-aircraft missiles or automatic cannons might be OK on the battlefield, but it is expensive and not very well suited to civilian areas were behind target effects might be more serious than the attack. Beyond highly kinetic means, and accepting geofencing and NOTAMS might only have limited utility against those with determination and malicious intent the two principal means of defeat are electronic and physical.

Electronic attacks might include exploiting a cyber-vulnerability to remotely hijack the device, usually over WiFi or Bluetooth. An example of this is the Maldrone malware. However, with an increasing regulatory concern for flight safety, consumer devices will become increasingly difficult to exploit. Randomised frequency hopping and encryption for control links are being implemented on newer devices. Attacks against the GPS subsystems have proven to be effective but again, secondary positioning with GLONASS and inertial systems may blunt even this.

Active RF jamming seems to be the most common active denial approach in the newer systems.

As can be imagined, it is an increasingly active market with all the defence primes and many SME’s now offering product solutions.

The Anti-UAV Defence System (AUDS) integrates the Blighter A400 Series Ku-band electronic scanning air security radar, Chess Dynamics’ stabilised electro-optic director, infrared and daylight cameras and target tracking software, and a directional radio frequency (RF) inhibitor/jammer system from Enterprise Control Systems to detect, track, classify, disrupt and neutralise UAVs at ranges of up to 8 km. The AUDS system is even effective against so-called Group 1 micro UAVs at ranges of up to 2 km and Group 1 mini UAVs at ranges of several km.

Mark Radford, CEO, Blighter Surveillance Systems, said: “We formed the all-British AUDS team in 2014 as we were each acutely aware of the urgent operational requirement from our customers for an effective and affordable anti-UAV system. Working in partnership, we have developed some clever technology (patents are pending) that integrates the different sensors, effector and electro-mechanical positioning systems to disrupt and bring down any malicious drone in a phased and controlled manner.”

According to the AUDS team, the technology has been extensively tested in South Korea along the 250km demilitarised zone (DMZ), where Blighter radars have been deployed for some years. In March 2015, the AUDS team took part in multi-supplier French Government trials in Captieux, France, where its counter UAV system proved highly successful in detecting and neutralising a variety of fixed and rotary wing micro, compact and standard UAVs. And last week, the system also performed well in UK Government sponsored counter UAV trials (known as Bristow 15) in West Freugh, Scotland.

The AUDS system is designed for counter UAV operations in remote border sites or urban areas. It can be operated from fixed locations and from mobile platforms. Key features and benefits include:

Fully electronic scanning radar technology with Doppler processing allowing all-weather, 24-hour detection of both fast and slow moving micro and mini-UAV targets with unsurpassed ground clutter suppression for near horizon operation;

Highly accurate stabilised pan and tilt director combined with the very latest electro-optic infrared day and night cameras and state-of-the-art digital video tracking technology to automatically track the UAV and classify the target;

Smart radio frequency inhibitor to selectively disrupt various command and control communication links employed by the UAV. Disruptive effect can be carried out in an intelligent, proportional and non-kinetic manner to mitigate collateral impact, and the inhibitor/jammer system is software controlled giving the capacity to counter new and emerging threats;

The Anti-UAV Defence System (AUDS) from Blighter Surveillance Systems, Chess Dynamics and Enterprise Control Systems. Chess Dynamics bring their Hawkeye EO/IR and tracking system that allows the operator to manually confirm and identify the target that has been detected by the Blighter A400 Ku surveillance radar, the Enterprise RF jammer making sure the target is no longer a threat.

There are hundreds of alternatives available now.

One of the eternal challenges with this type of active electronic attack is spectrum management, avoiding disrupting one’s own systems or providing something by which counter-counter systems can use. It could be argued that this is context, countering drones in an airport security setting must have a very high degree of safety, but not being concerned with the counter-counter attack.

As ever, solutions selected will be subject to many decision points.

Physical disruption can mean many things, simply shooting them out of the sky with a shotgun or automatic weapon is the obvious choice but this method has complications in built-up areas and against a rapidly manoeuvring vehicle will be very difficult. One of the Dutch project participants, Delft Dynamics, has developed a concept called Project Drone Catcher that uses another quadcopter to fire a grappling net at the target machine. The net can be tethered to the launch device which allows both to descend under control. Using one drone to destroy another with an intentional collision has also been demonstrated, even eagles have been demonstrated as being able to be trained in the counter UAS role.

An interesting counter small UAS system has been developed by the UK company, OpenWorks Engineering, called SkyWall. SkyWall uses a physical capture method to ensure drones are removed from sensitive areas.

Although the MoD’s response to the small UAS threat has not been widely publicised, it is not ignoring it. DSTL started a number of projects running under Project BRISTOW several years ago.

The five-day trial comprised flight profiles by four UAS providers: Callen-Lenz Associates Ltd, Blue Bear Systems Research Ltd, Resource UAS, and DSTL from various locations around MOD West Freugh. Each flight was monitored by seven sensor systems, consisting of commercial and military sensor equipment provided by Plextek, Selex, Thales, Saab, QinetiQ, UK Army and UK RAF teams attempting to detect, track and identify each UAS within the sanitised airspace provided. A substantial amount of data has been recorded and is now being analysed to produce a final report for Dstl

Bristow 15 was completed in May 2015 and included the British Army’s Giraffe Agile Multi-Beam radar and the Thales Squire radar, shown below.

Since then, the MoD has purchased a number of systems from different providers, Leonardo Orcus for example, read more here

Rafael Drone Dome systems have also been purchased.

A number of research and development programmes are also active. This presentation from a couple of years ago provides a good summary

Wg Cdre Rob Wilson And Peter Clarke

And in a wider UK context, click here to read more


Table of Contents

[adinserter block=”2″]

[adinserter block=”1″]

Change Status

Change Date Change Record
 01/03/2017 Initial issue
 05/05/2018 Update 
 16/07/2021  Update
Notify of
Newest Most Voted
Inline Feedbacks
View all comments