DJI Mavic, Air and Mini Drones
Friendly, Helpful & Knowledgeable Community
Join Us Now

Lost drone in the UK (article about GPS interference)

berisen

Member
Joined
Sep 9, 2019
Messages
15
Reactions
0
Age
35
Location
Sweden
I stumbled over this article about a lost drone in the UK.


It’s states that the crash was due to gps interference and that’s why the drone was drifting but isn’t this a perfect example of plausible magnetic interferences and not actual gps problems?

Never heard about a drifting drone due to GPS failure, is this normal?
 
It’s states that the crash was due to gps interference and that’s why the drone was drifting but isn’t this a perfect example of plausible magnetic interferences and not actual gps problems?

Never heard about a drifting drone due to GPS failure, is this normal?
A drone drifting in atti mode is not unusual.

The author and probably the drone pilot aren't experts so it's hard to interpret the account and work out exactly what happened.
It sounds like the drone suffered a compass error (the article mentions GPS compass error, but that's not a thing) that put the drone into atti mode.
That's not unheard of.
And in atti mode they failed to bring the drone back, either due to inexperience or possibly loss of signal but it's hard to see how from the illustration of the flight path.
None of that requires mysterious GPS interference to explain .. but it sounds good to an uninformed audience.
 
I stumbled over this article about a lost drone in the UK.


It’s states that the crash was due to gps interference and that’s why the drone was drifting but isn’t this a perfect example of plausible magnetic interferences and not actual gps problems?

Never heard about a drifting drone due to GPS failure, is this normal?

I don’t know why the author of the report referred to GPS Compass, they are separate and independent devices. GPS only gives a location, the compass gives a heading in conjunction with the IMU, which tracks rotational acceleration along 3-axis.

Drifting is normal and can be expected with loss of GPS. In this circumstance the AC enters ATTI mode and will maintain its attitude and altitude but can drift given sufficient wind. The remote controller will warn you if this occurs and just requires a little extra skill to control positioning. It’s worth practicing in ATTI mode so you know what to do should it occur.

Magnetic interference would be unlikely to cause this - the AC continued in a straight line by the looks of the flight plot. Ultimately it’s down to pilot error if you follow the report about them being caught out.

All that said, I have yet to experience loss of GPS in all the time I’ve been flying.
 
It sounds like the forbes writer didn't read the AAIB report https://assets.publishing.service.g...f596648297e/DJI_M600_Pro_UAS_reg_na_07-20.pdf


Here's an extract
The recorded log from the accident flight was analysed by the aircraft manufacturer, who stated that the reversion to ATTI mode had been caused due to a mismatch between the aircraft’s GPS derived heading and its magnetic compass heading data. This was attributed by the manufacturer to have been caused by signal interference that had affected the magnetic compass. The error had continued throughout the flight. The manufacturer advised that if the error had subsequently cleared, the automated flight modes would have been re-established.

DJI didn't suggest "GPS Jammers", that came from Forbes. Fly-away due to compass interference is a well known phenomenon. They launched a second drone from the same place and got a similar warning so landed at once (not 'nearly suffering the same fate' as Forbes put it).

Sounds like the pilot didn't realise that RTH wasn't available and froze. They never lost the link to the drone and should have been able to fly it back, but it reads like they were pressing RTH for the 75 seconds it look for the drone's path to go through someone's roof. It also appears they were trained, certified and working for a reputable outfit and did everything else write.
 
  • Like
Reactions: berisen
Good spotting.
The "pilots" were out of their depth.
The incident report says that: No manual flight control inputs were made using the controller.
It would have been a very simple exercise to bring the drone back.

Despite the author's bio notes, it seems that he was going well beyond his actual level of expertise too.

David Hambling Author of 'Swarm Troopers: How small drones will conquer the world,' following cutting-edge military technology in general and unmanned systems in particular.
 
It sounds like the forbes writer didn't read the AAIB report https://assets.publishing.service.g...f596648297e/DJI_M600_Pro_UAS_reg_na_07-20.pdf


Here's an extract


DJI didn't suggest "GPS Jammers", that came from Forbes. Fly-away due to compass interference is a well known phenomenon. They launched a second drone from the same place and got a similar warning so landed at once (not 'nearly suffering the same fate' as Forbes put it).

Sounds like the pilot didn't realise that RTH wasn't available and froze. They never lost the link to the drone and should have been able to fly it back, but it reads like they were pressing RTH for the 75 seconds it look for the drone's path to go through someone's roof. It also appears they were trained, certified and working for a reputable outfit and did everything else write.

An excellent report from the AAIB and a good call for us all to practice using ATTI mode, RTH (when useable) and above all Manual control.

Like everything with automation we rely on it way too much and lose our ability to quickly respond in an emergency situation.

Curious if this really was ‘magnetic interference’ or poor GPS. I guess it’s possible some heavy machinery in the industrial units could be creating interference but not totally convinced. Would love to see the logs analysed by the guys on this forum.
 
Good spotting.
The "pilots" were out of their depth.
The incident report says that: No manual flight control inputs were made using the controller.
It would have been a very simple exercise to bring the drone back.

Despite the author's bio notes, it seems that he was going well beyond his actual level of expertise too.

David Hambling Author of 'Swarm Troopers: How small drones will conquer the world,' following cutting-edge military technology in general and unmanned systems in particular.

That's my take too. It seems there was some kind of EM/RF anomoly (it was a construction site they were surveying, which may have entailed a large amount of steel being around - that can absolutely interfere with the onboard compass). That caused the mismatch with GPS, so the aircraft switched to ATTI mode and started to drift. So far nothing unusual, and all within accepted behaviour patterns in that scenario.

Then it goes wrong. The pilot and spotter apparently did not even attempt to manually control the drone until it drifted out of VLOS. The report indicates that there was *no* disruption to comms between controller and aircraft, so it should have been perfectly possible for the pilots to bring the aircraft in for a manual landing. They also did not try and climb vertically once they lost VLOS, a fair obvious safety measure at that point given they had nothing else to lose.

Then it goes seriously wrong. Having suspected that the drone had crashed, they failed to consider possible causes of the first failure and *sent up a second drone* to look for it, which promptly experienced the same problems. This time they managed to bring it down safely in ATTI mode, which *really* makes me wonder why they didn't try that with the first one. They've both now been sent on a refresher training course, apparently.

Still, even if the interference was the result of deliberate (and illegal) jamming, I'd put this 100% down to pilot inexperience; issues like this can and do happen, and pilots, especially those with PfCOs (commercial licenses) should be aware of how to deal with them. I also know for a fact that at least some of the UK's pilot training schools include flying in ATTI mode as part of their tuition, so hopefully that is part of their new course.

IMO, if you are within VLOS then the *only* thing you need to bring an aircraft home safely is for it to respond appropriately to the sticks. Beyond VLOS, you can add the current bearing and (ideally) speed/distance to that - you should already *know* the minimum safe altitude you need to fly at. If you need RTH, to see what the drone's camera is looking at, or an arrow on a map, then that's just not good enough - especially if you are flying something like an Inpsire or a Matrice that could do some serious damage if it came down hard. And if you have a drone that lets you manually switch to ATTI mode, then you should regularly find a large open space on a breezy day and practice, practice, practice.
 
It may not be pilots out of their depth. People are known to get fixated on using problem and one solution, its not uncommon to read an accident report and find someone kept trying to do X and you think "why didn't they do Y".

If you ask this pilot why he kept pressing RTH and didn't fly back "old school" he might well say "I should have done that. Could have done it. Don't know why I didn't". Good emergency training makes that less likely.
 
An excellent report from the AAIB and a good call for us all to practice using ATTI mode, RTH (when useable) and above all Manual control.

Like everything with automation we rely on it way too much and lose our ability to quickly respond in an emergency situation.

Curious if this really was ‘magnetic interference’ or poor GPS. I guess it’s possible some heavy machinery in the industrial units could be creating interference but not totally convinced. Would love to see the logs analysed by the guys on this forum.
TIG welding give off high frequency's. Chances are ther is a welding shop in the Wallsend factory's.
 
  • Like
Reactions: zocalo
TIG welding give off high frequency's. Chances are ther is a welding shop in the Wallsend factory's.
To cause a compass error that would force the drone into atti mode, the drone would have had to be very close.
The presence of a welding workshop in the neighbourhood or even flying over one won't do that.
 
Lycus Tech Mavic Air 3 Case

DJI Drone Deals

New Threads

Members online

Forum statistics

Threads
130,598
Messages
1,554,236
Members
159,603
Latest member
refrigasketscanada