Drone Collision Update
Two weeks on from the publication of the now infamous BALPA/DfT/MAA drone mid-air collision study and the controversy hasn’t receded.
My previous post critiquing the study drew a much larger response than I’d intended and as a result I was able to find at least a few answers to some of my many questions.
This post therefore seeks to cover the main issues that arose from some of the discussions I’ve had in the past two weeks, those that took place in public and some that took place in private, to shed some more light onto the findings of this study and the Government’s plans for drone registration. Both of which are inseparable.
Safety is the main reason that’s being used in the UK to justify the need for drone registration. It’s claimed that better education will solve many of the safety issues. The need for registration is so that education can be targeted at drone owners, hence the UK Government states it should be mandatory.
The often quoted rationale is that a recreational drone user has no real understanding of the law, so they fly in ways that cause dangers to manned aircraft.
This view is widely held, but it’s one that’s often held without any evidence to support it. It’s based on an assumption, one that I’ve read, heard experts talk about, and even been told by senior police officers and law enforcement professionals. Aviation experts tend to mostly agree with this view and yet there is no evidence to prove it’s correct.
The number of drone related prosecutions is very small in the UK and as such doesn’t constitute a valid body of evidence to then make sweeping statements or policy decisions related to tens of thousands of people. In addition a significant proportion of those prosecutions relate to intentionally flying drones into prisons to deliver contraband which hardly fits the model of someone who’s unaware what they’re doing is illegal.
On the basis of this perceived ignorance of the law the Government promotes the view that registration is essential for education and safety. This is despite the fact that recreational drone users by and large understand the law and are a very self-policing group. It seems that self-regulation is acceptable for the media, but not for the drone community.
What actually became apparent in my discussions with aviation experts however is that it’s they who don’t seem to know the law related to drones.
One expert, Scott Bateman MBE, describes himself as an, “experienced aviation consultant, and voluntary police officer”, and, “Head of Special Constabulary, & UAV Accountable Manager at Wiltshire Police”. As well as being a commercial pilot he’s also, “served as a special advisor and subject matter expert to the UK CAA [Civil Aviation Authority] on their Fixed Wing Advisory Group”.
One would then expect Scott to have an intimate knowledge and understanding of the legislation governing drone operations in the UK, certainly far more knowledge than any recreational drone pilot would be expected to have.
In my interactions with him I asserted that the legislation is unclear and applies unequally to commercial drone pilots and recreational operators, especially with regard to distances that drones can be flown away from the pilot. Scott dismissed this, claiming instead that the legislation was very clear and unequivocal.
The Air Navigation Order (ANO) 2016 section 94(4)(a) only stipulates a maximum height of 400ft above ground level (AGL) for drones over 7kg in weight. Its exact wording is:
“The person in charge of a small unmanned aircraft which has a mass of more than 7kg excluding its fuel but including any articles or equipment installed in or attached to the aircraft at the commencement of its flight, must not fly the aircraft—
“(c)at a height of more than 400 feet above the surface unless it is flying in airspace described in sub-paragraph (a) or (b) and in accordance with the requirements for that airspace.” You can read it here.
The ANO makes no provision for maximum heights or distances for which any smaller drone can be flown recreationally, just the proviso in 94(3) that,“The person in charge of a small unmanned aircraft must maintain direct, unaided visual contact with the aircraft sufficient to monitor its flight path in relation to other aircraft, persons, vehicles, vessels and structures for the purpose of avoiding collisions.”
This isn’t just my own interpretation of the legislation, it’s also the interpretation of the legislation by the UK Government which stated in its drone consultation response report that it was minded to, “Amend the Air Navigation Order 2016 to ban all drones of 7kg or less in weight flying above 400ft or 122m (drones of above 7kg are already banned from flying above this height).”
The Drone Code, a guide issued with most drones to educate recreational users, states: “Stay below 400ft to comply with the Drone Code”. However the Drone Code is a guide, it’s advice, and as such doesn’t constitute UK law and has no power to override UK legislation such as the Air Navigation Order.
Whilst the CAA can, and does, issue guidance in the form of publications (Civil Air Publications – CAP) and is a part of the Government, under the Department for Transport (DfT), it cannot create or amend laws without Parliamentary authority. In the UK this requires an Act of Parliament or some form of delegated legislation such as a Statutory Instrument (SI). CAPs do not constitute SIs, they are therefore simply advisory, not law.
It’s worth noting that no-one has ever been prosecuted in the UK for flying a sub-7kg drone above 400ft. Even the recent conviction of UK drone pilot Richard Smith on 5th July 2017 didn’t include any charges related to flying a drone above 400ft. The charges of flying through fog and above the cloud base all related to the offence of flying beyond direct unaided visual line of sight.
If there was legislation restricting drones from flying above 400ft he would have been charged with that, but he couldn’t because it isn’t illegal for sub-7kg drones. People can only be prosecuted if the drone is beyond their direct line of sight, or if it is too close to another aircraft, vehicle, vessel, person, or structure, or some other condition specifically stated in the ANO.
If such an expert of both commercial aviation and drone aviation can get it wrong then is it any surprise that recreational drone users can also get it wrong from time to time?
I must point out that although I’ve used my interactions with Scott to demonstrate this point he was in no way the only person who held this view. All of the commercial pilots I discussed this with held the view that it was illegal for a drone to fly above 400ft.
Just as they hold the assumption that recreational drone users don’t know the law, so they hold the assumption that drones are banned from flying above 400ft. Both assumptions however are wrong.
A lot of drone users understand the law and their obligations far better than commercial airline pilots, so where is there any evidence that safety will be improved through mandatory registration?
What was also surprising was that I was accused by most commercial pilots of promoting dangerous or illegal flying by pointing this fact out. It’s believed by them that as a commercial operator I should keep quiet about it in case it encourages people to fly their drones to high altitudes or near planes.
This was also the same justification commercial pilots used for attacking me for critiquing the mid-air collision study. They think the best way to prevent possible collisions between drones and airliners is to scare drone users with unreliable information.
I don’t agree with that approach at all, in fact I think it’s incredibly damaging and counter-productive. It leads people to distrust any future safety data in a field that’s already filled with plenty of distrust and conflict.
This was a hugely important element for the study to prove, if it failed to show there were critical issues between drones and airliners then the whole study would be undermined.
You couldn’t have the British Air Line Pilots Association (BALPA) co-authoring a study that showed drones posed no critical danger to airliners. That would be an unacceptable outcome for them.
The difficulty was that in all scenarios where airliners and drones were likely to meet the study failed to show any critical damage to the airliner windscreen. They already knew that drones wouldn’t cause an airliner to drop out of the sky if one got sucked into an engine since airliners are designed to be able to carry on flying and safely land with only one engine.
Someone who was present at the mid-air collision group meeting stated that Robert Curry, a Senior Inspector with the UK Airprox Board at the Civil Aviation Authority, said they’d already been in touch with several engine manufacturers. All of the manufacturers they had contact with were of the belief that a drone strike wouldn’t cause an uncontained failure of an airliner engine.
An uncontained failure of an engine is when parts of the engine are damaged and fly off causing damage to other parts of the aircraft, such as the fuel tanks, wings, cabin and so forth.
This supports the findings of Virginia Tech’s CRASH team and also directly contradicts Steve Landell’s (BALPA Flight Safety Specialist) claims that a drone strike could cause an uncontained engine failure.
I was also told that live testing of the commercial airliner windscreen stopped at 340kts and at that point there had been no critical failure caused to it by any of the quadcopter drones.
The study states, “critical damage did occur to the airliner windscreens at high, but realistic, impact speeds, with the 4 kg class drone components used in this study.”, but doesn’t say what speeds or if it was during live testing.
If the information I’ve received is correct it means that the study only showed critical damage to the airliner windscreen occurred in computer modelling, not in the live tests. That means that the 4kg class drone model never penetrated the windscreen in the live tests.
This raises questions about whether BALPA lied in their press release. In it BALPA claims, “The picture below, taken during the testing, shows a larger hobbyist-class drone penetrating an aircraft windscreen.”
The image appears to show the model they used to represent the 4kg class of drone, or as they refer to it, the larger, “typical hobbyist drone”, shown below:
The study states, “during one high speed live test with the Airliner-B windscreen, the 3.5 kilogram class fixed-wing drone components penetrated the windscreen.”, yet the image clearly shows the 4kg class quadcopter model, not the fixed wing model.
In one exchange with me BALPA stated that, “there was significant damage [to the airliner windscreen] at all speeds”, however they refused to explain what they meant by significant damage.
Despite claiming that they want drones to be safer BALPA has refused multiple requests to release further data from the study, instead going on the attack by accusing drone manufacturers of being in denial.
The above tweet was in response to a BBC report that the Drone Manufacturers Alliance Europe (DMAE) had called for the publication of the full study data in a press release entitled, “DMAE calls on UK Department for Transport to release full testing results of Mid-Air Collision Study”.
The DMAE questioned how they could design their products to be safer since it was unclear, “what specific drones were used, or exactly how they performed in their impact tests.”
It seems very strange that BALPA on the one hand call for safer drone use yet refuse to release data requested by manufacturers specifically to help achieve this aim.
The Need for Speed
The final nail in the coffin of BALPA’s claim that drones are a proven critical danger to airliner windscreens relates to the speeds they needed to run tests at. In my previous critique I’d estimated this could be anywhere between 300 – 575mph as the study gave no specific details. However they did state that, “they do not represent typical cruise speeds at altitude”, so it was more likely to be around the 340-400kts region as typical cruise speeds at altitude are around 500kts.
They also said: “These impact speeds would usually be encountered when the aircraft is at higher altitudes, 10,000 feet or above, but aircraft do sometimes operate at these speeds at lower altitudes.”
I was therefore keen to try and discover what sort of speeds this was referring to and found that airline pilots were by and large unwilling to say. One Captain with British Airways confirmed to me that 340kts would be flown at 10,000ft dropping to 180kts by around 3,000ft.
This would mean that to be in the vicinity of an airliner travelling at 300kts or above a drone would need to be at an altitude of 8,000ft or above.
The Guild for Air Traffic Control Officers (GATCO) in the UK pointed out that often airliners fly at 300kts with air traffic control approval between 5,000-6,000ft.
They also said that theoretically speed control could have been lifted in which case the aircraft would have reached 340kts at 5,000 – 6,000ft.
Even with these increased speeds below 10,000ft the scenario BALPA are referring to in terms of critical damage would require the drone to be operating at over 5,000ft, which although theoretically possible is highly unlikely.
If the claims I’ve received are true, namely that the mid-air collision group were told that no critical damage had been caused to the airliner windscreens up to 340kts in the live tests with quadcopters, then it becomes even more improbable. We’d then be looking at the 4kg class quadcopter causing critical damage to airliner windscreens at speeds well over 350kts which are only ever seen at altitudes far higher than those a drone would fly in.
I am now more convinced than ever that this study was never designed to test drone safety issues. If it was then the full data would have been made available to drone manufacturers and the Government would have announced immediate plans to form a working group to establish design standards.
BALPA released the most inflammatory image they could find along with an inaccurate description of it. At best this was mis-stated as “penetrating the aircraft windscreen”, at worst it was a blatant lie to intentionally create media hysteria.
According to information I’ve received this image actually shows the 4kg model of a quadcopter which has been fired towards the aircraft windscreen upside down to try and cause maximum damage. The black object seen on the upper side of the drone is the camera, which would normally be underneath the drone. This test did not penetrate the windscreen at all.
It’s alleged that in previous tests with the 4kg class drone model fired in the correct orientation (camera beneath the main body) the camera impacted the windscreen first and disintegrated, then the rest of the drone travelled up the windscreen parallel to it.
If correct it would seem that out of frustration with the results, in an effort to increase the amount of damage, the model was then fired upside down so that the battery would impact with the windscreen first. This still failed to penetrate the windscreen however, but it did give a better picture for BALPA and the DfT to use to make it look like it did.
Did BALPA intentionally lie to the media?
I can’t say for definite if they did or didn’t intentionally lie to the media. As it stands BALPA haven’t edited their press release, issued any statements of correction, or retracted their claims about this image.
Will the full data ever be released?
It’s highly doubtful. The full data would seem to confirm that there’s no evidence to prove that drones can realistically cause critical damage to airliner windscreens. As such releasing the full data would undermine the Government’s registration plans and give BALPA headlines they don’t want. As one of the sponsors of the study they will only want the outcomes that they’ve paid for.
In summing up I should also make it clear that I never advocate anyone flying a drone above 400ft or anywhere near major airports. I always recommend that drone pilots communicate with local Air Traffic Control to let them know if they’re nearby – there are lots of airports throughout the country and open communication always yields better results for everyone.
No-one wants there to be any mid-air collisions between drones and other aircraft, and no-one I’ve spoken to underestimates the damage and danger that would be caused if one did occur.