If you want to use a drone capture net as a counter measure you have one big hurdle that no manufacturer is willing to publicise on their website. How do you establish a legal authority to intercept a drone?
In the US you can’t legally deploy a drone net at present due to the fact that the Federal Aviation Authority state it is illegal to interfere with the flight of an aircraft. As they classify drones as aircraft you can’t try to capture them. In the UK it’s a bit more of a grey area, but the problem persists. You cannot interfere with a drone if it’s being operated legally.
How do you know if the pilot in charge is a commercial operator with permission to be in the air? You may own the land under which the drone is flying, but you most certainly don’t own the airspace above it. Unless the airspace is controlled or restricted and you can establish that the drone doesn’t have authorisation to be there you are breaking the law if you fire a net at it. You can also be sued for loss of earnings by the pilot if you do anything to interfere with their lawful business.
Even if the airspace is controlled and you can clearly establish the drone is operating illegally you then have to establish your right to try and intercept it. The Civil Aviation Authority state that, “If you have any concerns about drones being used in your area, either from a safety or privacy perspective, contact your local police on 101”. If you did try to interfere with the drone and damaged it you can also be arrested for criminal damage. Under certain circumstances the drone pilot would be allowed to arrest you until the police arrive.
The legality of using a drone capture net is a huge problem even before we’ve looked at the modes of deploying them and their effectiveness. That should give you some idea why the manufacturers of this type of counter measure avoid the subject altogether.
Let’s assume you’ve established a legal authority to take the drone down. You have two options (1) a ground based net launcher, or, (2) an airborne net launcher.
Ground Based Net Launchers
Perhaps the most well known of these systems is SkyWall, manufactured by Openworks Engineering Ltd in Northumberland. Skywall 100 resembles a black plastic replica of a Stinger rocket launcher. You can bet that the designers had that in mind when they were producing the prototypes.
Their brochure reads, “SkyWall allows an operator to capture and bring down a drone at a time and place of their choosing, using a combination of a compressed gas powered launcher and an intelligent projectile. Damage to both it and the surrounding area is minimised by controlling the descent of the captured drone.”
Essentially the system launches a finned pod towards the drone and when in range it bursts open releasing a net that snares itself in the drones propellors. At this point drones have all the aerodynamic properties of a brick, so to slow their descent the net has a small parachute attached.
The parachute element is quite important to understand here as it’s put forward as a safety feature to prevent damage to the payload if the drone has been weaponised. However, now the drone is in an uncontrolled, albeit reduced speed fall. It has no counter-action to the wind so will descend in a drift. This means that before launching the net the operator needs to take into account the possible landing site of the drone. Is it safe? Will the drone drift closer towards the principal? Will it drift towards a road? Could it cause an accident or potential loss of life as a car swerves to avoid it?
The advantage of the SkyWall system is that it is very portable and can be reloaded. It also has a maximum effective range of 100m, which is fairly impressive.That said, it has some significant gaps.
For instance the capture net parachute slows the fall of a DJI Phantom 3 drone to 3.3 m/s, or 7mph. Whilst the Phantom 3 has an operating mass of 1.2kg, the DJI Inspire 2 has a maximum take off mass of 4kg. This means that if you try to bring down an Inspire 2 with the net it’s going to be falling much, much faster. Certainly well in excess of 20mph. Therefore the operator of SkyWall has to correctly identify the type of drone and account for that in their decision before engaging it.
The other gap is SkyWall’s effectiveness at intercepting moving targets. Every promotional video I’ve seen of drone capture nets invariably show an almost stationary drone being captured. The reason is because it’s incredibly hard to hit a moving target, even with SkyWall’s built in laser ranging system. SkyWall’s own promotional video shows it engaging only with stationary drones. This raises an immediate concern – if it’s as effective as claimed against moving targets why wouldn’t they demonstrate that as a key feature?
SkyWall’s brochure states that the maximum target speed if it’s approaching you is 34mph, if it’s crossing your path that drops down to 28mph. The Phantom 3 has a maximum speed of 36mph which sounds like a good match. It’s unlikely the drone will be at full speed when SkyWall is taking aim at it, so you’d think it would be effective in targeting this type of drone.
Problems occur when you start to look at the maximum speeds of other drones though. The Phantom 4 has a maximum speed of 45mph, whereas the Inspire 2 has a maximum (limited) speed of 58mph. When the Inspire 2 was first launched it had a maximum speed of 70mph and the new limit is controlled through the on board software. It’s therefore possible to restore it to the previous setting of 70mph.
Put simply if the drone is flying faster than 34mph SkyWall will be unable to aim at the target correctly, so it becomes useless.
The best scenario for SkyWall to be effective is in targeting drones that are closing in on their targets. They will most likely be slowing to a stop to line up with a prison window, or be hovering to try and get good footage of the principal, or aim a payload. In these situations the operator will have just a few seconds to get into position, assess the drone weight, likely outcome of where it will fall and fire the net pod at it.
If all those factors line up then this is an effective counter measure, and in certain circumstances it will be an effective tool. It’s just that the circumstances where this will occur are not very often and I feel there are other security solutions that are more effective.
For example, the outlying security detail should be able to hear a drone flying towards them and be able to communicate to the close team that it’s incoming. This gives them the time to move the principal to a safe position. Likewise in a prison scenario the officers would be able to identify the cell the drone is hovering outside of and proceed accordingly.
Airborne Net Launchers
While the Dutch police, being accused of getting high on their own supply, were training an eagle to attack drones mid-air the Japanese were trialling a drone fitted with a net.
The Japanese drone flew into the path of ‘rogue’ drones, capturing them and landing them safely on the ground. Since then people have realised that it would be better to have a net fired from a drone in case the total weight of both drones was too much. Unsurprisingly no one has really embraced the Dutch police’s eagle idea.
Both Delft Dynamics and Search Systems offer what they refer to as Counter UAV’s or C-UAV’s. These are drones with the ability to launch capture nets in the air. Delft Dynamics C-UAV is presently undergoing the final stages of its development which they claim is, “very promising”. However, the video Delft Dynamics have released showing the system in action is once again targeting a stationary drone.
Sparrowhawk is Search Systems C-UAV and it boasts an on-board ranging system that notifies the pilot when the drone is in capture range of its target. The operator then deploys the net and, as in the SkyWall system, the drone is gently lowered to the ground by a parachute.
The distinct advantage an aerial system has over a ground based system is its manoeuvrability. The pilot can effectively chase the drone and hunt it down in a mini ‘Top Gun-esque’ dog fight. This sounds great in theory but does it work in practice?
Sparrowhawk is an hexacopter (a drone with six propellors) that looks remarkably similar to DJI’s own hexacopter, the S800 (pictured right). Even if Sparrowhawk has some hardware differences it is highly likely that the flight controller, the brains of the drone, is manufactured by DJI.
Search Systems don’t publish any detailed specification data on their website about Sparrowhawk. Given its external similarity to the S800 I think it’s fair to say that its flight characteristics will be similar too.
The S800 has a takeoff mass between 5kg and 7kg which makes it quite a beast compared to the much lighter Phantom 3 (1.2kg). Weight and size are significant factors with drones that shouldn’t be underestimated. The bigger the drone the more susceptible it is to wind, which is a definite issue when manually flying. It may have a good wind resistance, but that only applies when the drone is in a fixed position or knows where it’s going ahead of time. When you’re manually flying, the computer has no time to counteract the wind, so you get pushed off course much more.
The other factor is inertia. Because there’s practically no friction between the air and the drone, inertia plays a bigger role than it would if you were driving a car on the ground. In that scenario you have the traction of the rubber tyres to counter and control the effect. Granted, a 6kg drone has much less inertia than a 1250kg car, but when flying through the air you’d be surprised how much of an effect inertia can have.
Even before we look at top speeds we have to take into account the different flight characteristics between the Phantom 3 and the Sparrowhawk. The Phantom 3 will be much more nimble and less affected by the wind making it harder to chase. It will be able to out-turn, out-brake and outmanoeuvre the Sparrowhawk which is an issue. It means that in this context the pilot’s skills are a key factor in the ability of this system to be effective. Whereas SkyWall reduces the skills required by the operator, Sparrowhawk is entirely dependent upon them.
Now we get to the other issue – the maximum flight speed of Sparrowhawk. DJI don’t give the maximum speed of the S800 due to the fact that it can be configured in a variety of ways with different gimbals and cameras, which has an obvious impact on flight performance. However, users report that its top speed averages around 40-45mph.
If we take the upper limit of those figures it would place the Sparrowhawk in the range of the Phantom 4, having a higher top speed than the Phantom 3. With all models of the Phantom we have to consider manoeuvrability and how possible it would be to sustain a chase through the air.
Whilst Sparrowhawk has a higher straight line speed than the Phantom 3, I strongly doubt whether it would be able to keep up with its turns. That said, if in a panic, the other pilot would be likely to fly in a straight line to get away from Sparrowhawk rather than trying to outmanoeuvre it.
The other downside of Sparrowhawk’s size and 6 rotors is that it won’t be very stealthy. It won’t be sneaking up on the target drone because the other pilot will see it and hear it coming from a long way off. This then places us into the problem area of the legal flight distance for drones.
In the UK you can only fly a drone up to 500m away from the pilot unless they have forward observers and a licence to operate an extended range. In real world scenarios that won’t happen. Your observer is as likely to be out of position 500m behind you because you don’t know which direction the threat will come from.
Travelling at 45mph you will be out of the 500m range within 25 seconds. If the target drone is 100m away from you Sparrowhawk will have to close that gap and launch the net in a very short space of time. If the target drone can outpace Sparrowhawk that gap will widen and you’ll have no ability to capture it. Recalling the figures for the DJI Inspire 2, that can cover the 500m in less than 17 seconds and it can fly 25mph faster than Sparrowhawk.
Once again there will be certain scenarios where a Sparrowhawk type system will be effective, but just as the SkyWall system, those scenarios will be few and far between and can be countered with more effective conventional methods.
Having worked in the security industry for many years I can empathise with the desire to find a solution to the potential threat a drone can cause. However, so far I haven’t seen a single example of a net based system that would justify its expense.
I’d welcome some more openness and honest debate from manufacturers of this technology because I do feel that they are being somewhat disingenuous in hiding the blatant flaws of this approach.
For me this an example of old world thinking to try and solve a new world problem. However you engineer it, it’s never going to be truly effective.
In my opinion the great strength of airborne drone capture systems is simply that they are airborne. You don’t need to capture the drone, all you need to do is repel it. If a drone infiltrates an area the best strategy you can use to counter it is by launching another drone.
If the pilot of the encroaching drone is simply unaware they’re in a sensitive area the presence of another drone is likely to scare them off. If they are intent on causing harm their job is now much harder because they have to contend with an airborne obstacle.
I should also make clear that rather than this being a theory it is a practice I have successfully used before. A pilot’s instinctive reaction to another drone in the airspace is to move away from it. Capitalising on this you can shepherd drones away from your secure area rather than trying to capture them.
In the very worst case scenarios it is far simpler and easier to fly your drone into the rogue drone, causing it to crash. Whilst that wouldn’t enable a graceful parachute descent it would at least stop the drone and give your organisation ‘plausible deniability’. You can claim the crash was accidental and caused in part by the other pilot. (Note: I haven’t had to use this method)
In most situations the best defence of airspace against drones is the presence of another drone. It also helps to establish that the other pilot is operating illegally since they cannot fly within 50m of your drone. Your own drone can be performing vital perimeter security as well as providing airspace security, you just don’t need to overcomplicate it by adding a net device.
In this scenario you pro-actively establish an aerial presence which acts as a deterrent, rather than the net deployment scenarios that are principally geared towards reactive approaches.