Another Watchkeeper down

[airsound]

A fifth Watchkeeper has crashed - this time near Aberporth in Wales. In case you're not familiar, Watchkeeper is an ISTAR UAV asset of the British Army, built by a company set up by Elbit and Thales, and based on the Elbit Hermes. Report from the Cambrian News
Fire crews called after drone crash near West Wales Airport News Cambrian News (http://www.cambrian-news.co.uk/article.cfm?id=121277&headline=Fire%20crews%20called%20after%20drone%20crash%20nea r%20West%20Wales%20Airport&sectionIs=news&searchyear=2018)

Defense Aerospace comments:
The Watchkeeper programme was originally contracted for £800 million by the British Army, but delays and overruns have boosted its cost to over £1,200 million.
In exchange, the British Army obtained 54 Watchkeeper aircraft, several of which were rushed to Afghanistan so they could fly some token missions before the British Army pulled of the country.
They have since been mostly operating on training missions, with regular crashes diminishing their number: five having been lost to date.

airsound

[melmothtw]

I remember the army's programme chief telling me some years ago that attrition losses were factored into the 54-aircraft buy, but they are going through airframes at a rate of knots.

Genuine question - why do UAVs need to be flown and potentially lost in peacetime? Why can all the training not be done synthetically? For the pilot, he is not in the aircraft anyway so will lose nothing in terms of realism from sitting in a simulator . Any ground forces he is supposed to be interacting can do so synthetically also, no?

[Heathrow Harry]

I remember the army's programme chief telling me some years ago that attrition losses were factored into the 54-aircraft buy, but they are going through airframes at a rate of knots.

Genuine question - why do UAVs need to be flown and potentially lost in peacetime? Why can all the training not be done synthetically? For the pilot, he is not in the aircraft anyway so will lose nothing in terms of realism from sitting in a simulator . Any ground forces he is supposed to be interacting can do so synthetically also, no?

got to keep Aberporth busy I guess....................

[Mechta]

Synthetic training does not iron out the rarely occurring faults in the hardware, airborne software, RF link, weather related incidents, maintenance issues etc. Applying the same logic, none of our armed forces would ever leave their bases until war is declared.

Anyway, if the UAV is in a tree, it hasn't crashed, it just hasn't landed yet!

[melmothtw]

All of those failures can be included in the synthetic training, just as engine failures etc are currently practised on simulators.

There is value to real-world training where that training replicates the actual environment where the person will be operating, but I still don't see why a drone operator who in real life will be sat in a windowless shipping container with no physical connection to his or her aircraft cannot experience synthetically everything they will ever experience in the real world.

[Mechta]

What you are suggesting is that the time to discover faults with the air vehicle is when our troops on the ground are depending upon it? Without knowing what the contingency plans were, there is no reason to suppose that the drone operators did anything wrong at this stage.
A British operated Hermes 450 was lost at Camp Bastion, with the root cause being a blocked oil feed pipe due to the engine oil reacting with the plasticizer in the pipe. Hermes 450 ZK515 loss (https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/34176/zk515_hermes_part_1_4_1.pdf) Synthetic training won't bring things like this to light.

[airsound]

Mechta, you saidSynthetic training won't bring things like this to light.

Quite so, and, although the erudite discussion on the thread about synthetic training is interesting, I'd like to know why these things keep crashing. Any ideas, anyone?

airsound

[melmothtw]

What you are suggesting is that the time to discover faults with the air vehicle is when our troops on the ground are depending upon it? Without knowing what the contingency plans were, there is no reason to suppose that the drone operators did anything wrong at this stage.

At this rate it won't matter what the fault(s) might be, because by the time our troops on the ground are depending on it there won't be any left.

[Fareastdriver]

Only £22.22 million a throw; peanuts.

[Pontius Navigator]

Melmothw, simulation is used to practise actual flight conditions and procedures. Without actual flight experience you do not know if the simulation is accurate. Also, a general rule in simulator flying is to concentrate on likely scenarios as there is unusually insufficient simulator time to explore every possibility with every pilot.
​
​​​

[melmothtw]

Thanks Pontius, I know what simulators are for. The actual flight conditions and procedures for the drone pilot are surely no different to the simulated flight conditions and procedures, no? There is no physical connection between the operator and aircraft in either environment so surely anything and everything can be simulated.

The lack of simulator time you speak of has no real bearing on my point - that could be adjusted to meet the need.

I'm still no wiser as to why a drone operator/pilot can't learn and train for everything they might encounter in a synthetic enviroment, and so save expensive airframes from falling out of the sky until none are left.

But no matter.

[ORAC]

At least it’s not as bad as the “buggeroff” (Phoenix).

[nonsense]

Genuine question - why do UAVs need to be flown and potentially lost in peacetime? Why can all the training not be done synthetically? For the pilot, he is not in the aircraft anyway so will lose nothing in terms of realism from sitting in a simulator . Any ground forces he is supposed to be interacting can do so synthetically also, no?
Pilots aren't the only people involved in operating UAVs, nor the only people to benefit from training.

[airsound]

Sorry dears - I'm still not with you.

Surely, how people might be trained to deal with all the possible emergencies that could occur is a secondary concern? Wouldn't it be a good idea to decide why so many of these expensive airframes [are] falling out of the sky until none [is[ left. and then stop it happening?

airsound

[melmothtw]

Pilots aren't the only people involved in operating UAVs, nor the only people to benefit from training

Indeed, but I still don't see why ground handling crews, air traffic controllers, forward controllers, troops on the ground etc, need the Watchkeeper to be flying to get their training value.

Their only interaction with the drone in real-life would be via radio communication with the operator /pilot (who could be in a networked simulator for training) or viewing a video feed (which could be synthetic or fed from a surrogate manned platform that wouldn't fall out the sky).

[Mechta]

November 2014 - WK031 - Master Override selected to get aircraft down on first approach ahead of approaching thunderstorm. Aircraft thought it had landed when it hadn't, so gave down elevator command. Dived into ground as a result. WK031 accident description (https://www.theregister.co.uk/2016/08/12/watchkeeper_drone_wk031_crash_report_1m_uav_destroyed/)
November 2015 WK006 - Master Override selected to get aircraft down after two failed automated landing attempts. Aircraft dived into runway at 35° nose down. WK006 Accident description (https://www.theregister.co.uk/2016/12/15/watchkeeper_wk006_crash_report_poor_software_laser_altimeter s/)
3 February 2017 - WK042 - Lost whilst testing de-icing equipment.
24 March 2017 - WK043 - Training flight WK043 & WK043 Accident description (https://www.theregister.co.uk/2017/11/01/thales_watchkeeper_crashes/)

It appears that in an effort to achieve a system with operators rather than pilots, the ability to do a fully manual landing has been designed out. This leaves a system dependent on the automated system sensors always working properly; which they don't.

Lettting the manufacturer provide 'surveillance by the hour' is the way to go, rather than letting a succession of two-year tourists tell the experts how to design a UAV for a soldier to operate.

The Israelis will get it right if you just let them get on with it, in the knowledge that crashes will come out of their pocket and not yours.

[KPax]

Basrah late 2007/08 Hermes flown into the side of the main terminal in very poor vis, I also saw a picture of a Hermes trying to fly into the side door of aMmerlin that was also airborne. No criticism but am I correct in thinking that the Army UAV's are not flown by pilots.

[Mechta]

Basrah late 2007/08 Hermes flown into the side of the main terminal in very poor vis, I also saw a picture of a Hermes trying to fly into the side door of aMmerlin that was also airborne. No criticism but am I correct in thinking that the Army UAV's are not flown by pilots.

KPax, the Hermes 450 would have had an 'external pilot' doing the takeoff and landing, and an 'internal pilot' in the ground control station for the rest of the flight. The external pilots would have been experienced R/C pilots, as those are the skills required, and would have been civilian contractors. I don't know about the internal pilots. Watchkeeper is intended to be operated by the Army, although a lot of the development flying has also been by non-military personnel, as would be expected.

[EAP86]

Mechta, you said

Quite so, and, although the erudite discussion on the thread about synthetic training is interesting, I'd like to know why these things keep crashing. Any ideas, anyone?

airsound

The absence of a pilot means that the aircraft needn't be designed to be as safe as a normal military aircraft. Quite a lot of the cost savings come from this decision. Unfortunately they then tend to crash a lot more often but this is regarded as an economic issue rather than a safety issue. The safety of the overflown population can be helped substantially by clever choice of the aircraft's course/operating area.

EAP

[airsound]

Thanks, EAP for that - and also Mechta for the accident accounts. Are the Service Inquiry reports publicly available, do you know? I haven't been able to find out so far - sorry!

Also, while accepting EAP's point about UAVs' not having to be as safe as manned (personned?) aircraft, I don't recall hearing of anything like this rate of accidents to other advanced UAVs like Reaper.

airsound

[Nige321]

Thanks, EAP for that - and also Mechta for the accident accounts. Are the Service Inquiry reports publicly available, do you know? I haven't been able to find out so far - sorry!

Also, while accepting EAP's point about UAVs' not having to be as safe as manned (personned?) aircraft, I don't recall hearing of anything like this rate of accidents to other advanced UAVs like Reaper.

airsound

They are here... (https://www.gov.uk/government/collections/service-inquiry-si#service-inquiries-(si))

[Sun Who]

"Also, while accepting EAP's point about UAVs' not having to be as safe as manned (personned?) aircraft, I don't recall hearing of anything like this rate of accidents to other advanced UAVs like Reaper."

https://dronewars.net/drone-crash-database/

[EAP86]

SW, very interesting. I'm pretty sure that the list is not exhaustive. If safety isn't an issue, why would they report all mishaps?

EAP

[Mechta]

SW, very interesting. I'm pretty sure that the list is not exhaustive. If safety isn't an issue, why would they report all mishaps?

EAP

In their own words, " Drone Wars UK provides information and comment on the growing use of armed drones. As we are based in the UK we focus on the use of British drones but also include information about armed drones in general. " It is natural that they would want to highlight how frequently drones/UAVs fall out of the sky.
With regard to the Government reports, that's freedom of information for you, plus the desire to learn from incidents and accidents in an effort to prevent an identical reoccurence.

Safety is ALWAYS an issue whenever a manmade object takes to the air.


They've got it down from the tree now. Needless to say, it 'just missed' a school sports day. BBC report - Watchkeeper UAV recovery (https://www.bbc.co.uk/news/uk-wales-mid-wales-44502731)
:
https://cimg2.ibsrv.net/gimg/www.gmforum.com-vbulletin/736x414/_102075838_drone_78cc9d94a0d94c1d4837612f84b248bd92bfa147.jp g

Watchkeeper UAV WK005 Recovery 16 June 2018

[EAP86]

Mechta, sorry loose wording on my part. I wasn't referring to Drone Wars UK; the "they" was referring to the various operators of UAVs. Many industrial concerns and national Air Forces have experimented with UAVs investigating technical and operational viability. The industry operators will only report mishaps when there is a regulatory imperative to do so. I suspect many Air Forces will only report mishaps only when the knowledge is already in the public domain. I'm aware of the odd mishap which doesn't appear in their listing hence the first part of my statement. The strict definition of 'safety' relates to the risk to life, not the economic losses accruing from a mishap. If you can operate within a sanitised range, you can achieve adequate safety as there is little or no risk to life. I'd hope this explains the second part of my statement.

I note that their database only starts in 2007. In the early 90s I attended the UAV Systems Conference in Bristol. The first Bristol conference was held in 1979. A lot of mishaps can occur in 28 years.

EAP

[Mechta]

EAP86, The difference with Watchkeeper is that it is no longer just operating over military ranges and areas of conflict. With governments wanting to operate Watchkeeper and Predator or 'Protector' (as they like to call it now!) over the heads of their voting public, replacing manned aircraft, transparency is needed to keep their voters on side.

Approval to operate Predator in UK airspace was sought (https://www.theregister.co.uk/2018/02/12/predator_drone_approval_airway_flights_uk/)

Before 2007, UAV operation in the UK was, to the best of my knowledge, confined to military ranges and sparsely populated areas, so maybe of less interest to the Dronewars people. Prototype and development UAVs, not destined for production, which get bent on a military range, are of little concern to anyone other than UAV developers and potential operators, unless the future intention is to use them elsewhere.

[Windy Militant]

Unfortunately this is going to stir up a hornets nest as there has already been very vocal protests locally and this is only going to make them worse. As to the School it's about a mile and a quarter west of the runway and almost directly under the extended centre line, and it was there before the airfield!

[Davef68]

Don't these 'fly' autonomously and just respond to directional commands from the operators?

[Alber Ratman]

I was out at Basrah when the Hermes was flown into the terminal building. It was a very foggy morning. I did hear the operator was asked by Air tragic if he wished to be talked down onto the runway. He did not. I saw the wreckage later in the day. Very funny!

[SlopJockey]

Don't these 'fly' autonomously and just respond to directional commands from the operators?
So not autonomous then ;)

[VX275]

I was out at Basrah when the Hermes was flown into the terminal building. It was a very foggy morning. I did hear the operator was asked by Air tragic if he wished to be talked down onto the runway. He did not. I saw the wreckage later in the day. Very funny!

The crash at Boscombe was also a very foggy day and I heard (but never saw) the Watchkeeper that crashed from my office in Bldg 801. May be someone needs to tell the Army that relying on lasers for height finding should rule out ops when the cloud is on the deck.

[ORAC]

US Army point if view from recent copy of AW&ST:

.......”Another concern is airworthiness. This was not an issue when UAS were purchased and delivered into war zones, but it became a problem when the service wanted to bring the systems home and train with them in domestic airspace.

The Army has three levels of airworthiness for UAS. Level 1 is the same as for manned aircraft and calls for a loss rate not higher the 1/100,000 hr. Level 2 is 1/10,000 hr., and is the minimum required if the platform is weaponized. Level 3, at just 1/1,000 hr., is the minimum acceptable level of safety, says Dave Stephan, associate director for technology with the Army’s Aviation Engineering Directorate. The Army’s most capable and expensive UAS, General Atomics Aeronautical Systems (http://awin.aviationweek.com/OrganizationProfiles.aspx?orgId=45148)’ MQ-1C Gray Eagle (pictured), is a Level 2 system—and that is for the kill chain, from sensor to operator to weapon. The air vehicle itself is only Level 3, which means the Army has to treat the Gray Eagle as expendable, “which is hard when it costs $10 million,” Stephan told the forum.

The reason for this low level of airworthiness qualification is historical. The Gray Eagle is a program of record, but many of the system capabilities and upgrades have been developed by the manufacturer and acquired outside the normal procurement system, he says. “Our legacy UAS were developed to meet urgent needs, and they have flight restrictions,” Stephan says. “Software certification is not possible, or too expensive, and propulsion certification may be in progress but will not meet Level 1.” They were not fielded with the intent of flying in U.S. airspace, and when it trains with these systems, the Army “accepts risk at multiple levels,” he says.

Future UAS is an opportunity to reset the safety bar and could have broad implications for industry. “Future UAS is expected to be Level 1 and to meet the manned safety requirement,” he says.

Industry is already grappling with how to certify commercial UAS, but progress is being paced by the types of operation the FAA (http://awin.aviationweek.com/OrganizationProfiles.aspx?orgId=31159) will allow. There is limited demand for certified UAS, while the FAA places strict limits on UAS operations. And the limits are unlikely to be relaxed until there are certified UAS. It is a chicken-and-egg situation the Army’s emphasis on airworthiness could help hatch sooner rather than later by raising the bar for UAS safety.“........

[EAP86]

Its a few years ago but I got the chance to listen to someone from the US Army who had airworthiness responsibility for their aircraft and the scale of the issue was surprising. If I remember correctly, they had 12000 air vehicles to consider of which, 7000 were unmanned.

EAP

[Davef68]

So not autonomous then ;)

Autonomous in that the operator doesn't actually 'fly' the aircraft, just tells it where to go.

[Chugalug2]

Thank you ORAC for the AW&ST quote in your #32 above. It would seem that the UK is in much the same situation as described in the piece, ie that unmanned aircraft systems may not meet domestic airworthiness requirements, but are flown in domestic airspace for training, testing, or development purposes. The solution of ingenious routeing previously mentioned does not seem viable, if only because of the ever presence of sod's law. What is acceptable in an operational area (clinging to the outside of an Apache on a rescue mission into a firefight for example) is often a no-no back home. From what has been said about the issues with Watchkeeper I would suggest that it fails to meet the most basic of airworthiness definitions (EASA):-
"Airworthiness is the measure of an aircraft's suitability for safe flight."

[Fortissimo]

First, autonomous drones are those that require no external intervention by a pilot or operator beyond the initial programming of the flight; they work entirely on their own once you hit the go button. Remotely Piloted Air System is a better description (even though I don't like it!) of what goes on with the larger platforms such as Watchkeeper.

On the airworthiness front, 'suitability for safe flight' depends on the level of 'safe' you are trying to achieve. There is an ongoing debate across the industry about equivalence (ie an equal level of AW assurance with manned flight), as the current risks from drone ops are self-evidently 3rd-party only. That position will change when RPAS (or possibly autonomous drones?) start carrying passengers, as is now being explored in UAE, because the systems will need to meet the relevant weight-related CS standards applicable to CAT platforms. Until then, the use of intelligent routing as a means of mitigating the 3rd-party risks appears to be both ALARP ('practicable' not 'possible'...) and proportionate.

What really does need to be better handled is the professional aviation knowledge of those operating Watchkeeper and the like. The report on the Bastion accident pointed clearly to this. Whilst the root cause of the engine failure was the blocked oil pipe, an individual witth better knowledge and training might have understood the implications of rising temperatures more quickly and protected the engine thereafter. Easy to say with hindsight etc, and I am not trying to point fingers here. Similarly, the Boscombe fog accident may have been avoided had people (at all levels of the procurement and operating processes) thought about the potential pitfalls of flying a system designed for Israel in the prevailing weather conditions of N Europe. All it needed was a line in the RTS to say 'landing in visibilities less than X metres is not permitted'.

[Chugalug2]

Fortissimo:-
On the airworthiness front, 'suitability for safe flight' depends on the level of 'safe' you are trying to achieve.

Well let's drill down to a less ambiguous airworthiness definition:-
"The ability of an aircraft or other airborne equipment or system to operate without significant hazard to aircrew, groundcrew, passengers or to the general public over which the airborne systems are flown".

I would suggest that all embracing wording would include UAS's, and if you exclude for the sake of argument all except the general public you cannot exclude them. I believe that the original concept of airworthiness was for 'the protection of those wot was down below'. With the coming of the drones time perhaps to revisit that concept.

[Lloydybar]

Being a newbie I'm unable to post urls at the moment but there are a couple of reports which have been recently added on the Defence Safety Authority website regarding the WK042 and WK043 incidents in Feb/March 2017. Apologies if this info has been previously posted.

[Chris Kebab]

Interesting stuff, bit of a theme with these WK accidents; remarkable how a piece of kit which, on the face of it, looks so simple can be apparently so complex and poorly understood.

Does anyone fully understand how this thing works; do the Israelis themselves not really know how the final system integration all works or are they simply reluctant to pass on that knowledge to their customers (not just MoD but Thales/UTacS) simply leaving them to find out themselves?

Not too sure anyone would have managed these situations any better than the guys that were there did?

[Lima Juliet]

Links are easy these days - just cut and paste from the URLs

https://www.gov.uk/government/publications/service-inquiry-report-into-the-loss-of-watchkeeper-wk043-unmanned-air-vehicle-over-cardigan-bay-in-west-wales-on-24-march-2017

https://www.gov.uk/government/publications/service-inquiry-report-into-the-loss-of-watchkeeper-wk042-unmanned-air-vehicle-over-cardigan-bay-in-west-wales-on-3-february-2017

[BEagle]

Throughout the flight, there were examples of poor communication; the crew was unsure of who was doing what and why.


:\ VVVVVVVVVVVVVVVVVVVVVVVVVVVV

[Lima Juliet]

Yes BEagle and if you read the Service Inquiry then you will find this little chestnut on the WatchKeeper (WK) Unmanned Aircraft (UA):

WK is a fully automated system, which means that the crew have no way of taking control from the autopilot. On other platforms taking control from the auto pilot and manually flying with a level attitude and a safe power setting would usually be sufficient to protect against stall should CAS become erratic and erroneous. The Panel noted that ESL's Hermes 450 could be flown manually in 'sticks' mode, giving the operator or external pilot full throttle and stick control of the platform as required. The system architecture for WK is, however, quite different and there is currently no way of manually flying the UA.

So I read the fact that Watch Keeper is ‘point and click’ without flying controls it is very different to Reaper or Protector that has a throttle, stick and rudder. Indeed, the one Reaper that suffered an issue an engine failure over Afghanistan was successfully ‘dead sticked’ onto the desert floor by a Reaper pilot (who was a FJ QFI) and if it wasn’t in a contested area when the CH47 went to recover the sensitive parts with chainsaws then it probably could have been fixed and flown again. The final coup de grace was delivered by a GR7/9 with a Paveway to ensure that nothing was left to use by enemy forces.

The latest advanced Ground Control Station (GCS) for Protector makes you feel like you are looking out of a multi-engine ISTAR aircraft cockpit window. I’ve flown this GCS and it gives an unparalleled level of situational awareness to the crew with overlaid information on that 3D cockpit view - I wish I had that in the Tonka!


https://cimg7.ibsrv.net/gimg/pprune.org-vbulletin/1280x853/image_29a9a24b920611ee22dfdf60d85ede5268984383.jpeg

https://cimg9.ibsrv.net/gimg/pprune.org-vbulletin/1200x1041/image_6dc9b0063b2018ea84ad9d1ed68c91d0476eae26.jpeg

This is a significant convergence between manned aviation and unmanned which is why it is now a separate specialisation within the Pilot sub-branch within the Flying Branch rather than sitting totally separately. Whilst their selection will still be separate from the others but with the ability to ‘cross-over’ like FJ, RW and ME have in the past. The pilots of Reaper and Protector complete EFT and some live instrument flying before going to fly live Reaper (soon Protector) and much simulator time. A procedural IR is also becoming a must as these aircraft start to ‘mix it’ within various FIRs with manned traffic. So treating these unmanned pilots as de facto Aircrew is one thing the RAF (and other nations) have got right and possibly where the JHC/Royal Artillery/Army have not? But then again, with just ‘point and click’ with Watch Keeper, then you are going to struggle to keep traditional pilots interested.

[tubby linton]

I remember a similar looking craft exhibited in Larkhill camp next to a self propelled artillery vehicle a number of years ago. When I drove through there recently it had gone which I thought strange.

[Vendee]

The latest advanced Ground Control Station (GCS) for Protector makes you feel like you are looking out of a multi-engine ISTAR aircraft cockpit window. I’ve flown this GCS and it gives an unparalleled level of situational awareness to the crew with overlaid information on that 3D cockpit view - I wish I had that in the Tonka!



The yellow and black handle on the seat is a bit of a worry.

[Blossy]

The yellow and black handle on the seat is a bit of a worry.

I know the feeling and can only blame our military training!

[reader8]

Synthetic training does not iron out the rarely occurring faults in the hardware, airborne software, RF link, weather related incidents, maintenance issues etc. Applying the same logic, none of our armed forces would ever leave their bases until war is declared

What a unique approach to T&E. You're not a PTL by any chance are you?

[Lima Juliet]

The yellow and black handle on the seat is a bit of a worry.

It’s so you don’t inadvertently pull it and roll back over someone’s foot. There is always someone watching over your shoulder! Look at the ‘no step’ markings - they wouldn’t put them there if it hadn’t happened before... :eek:

[LowObservable]

Rudder pedals on a GCS = Steering wheel on a horse

[Blossy]

It’s so you don’t inadvertently pull it and roll back over someone’s foot. There is always someone watching over your shoulder! Look at the ‘no step’ markings - they wouldn’t put them there if it hadn’t happened before... :eek:

I thought that those of us who were in harms way aloft didn't give much thought to those in harms way in a simulator. Didn't have such fancy contraptions in my day.

[pr00ne]

Blossy,

It's not a simulator