Mavic Air 2 crashes into Lighthouse

[Order66]

That can happen. Every pilot of a Boeing that takes off knows that. He takes that responsibility. We should never blame the machine... i am a motorbiker: my biggest efforts go to correct errors other drivers make as well. Upto now i only had one accident and i was not enjured, apart from some blue spots on my hip and brest when i decided to jump off the motorbike when the car driver started to panic and freeze. If you crash against a bird, if your equipment is not maintained... i have nobody in my pilot environment who had an engine error, except in heavy rain where the battery got water (known problem) and fell out. He admitted as well he should not have taken off so... that he made an error as pilot. Weather conditions IS the responsibility of the pilots. Even tyre pressure on a motorbike and car is the ‘pilots’ responsibility, it is on the checklist and that is why i have PTMS on my vehicles as well (1% of motorbike/car incidents come from to low pressure on tyres followed by overheating or taking a sharp corner with too low pressure where the tyre looses all air suddenly or even explodes). Overpressure is hardly an issue. +je

I'm glad you said this, since motorcycling is a good example of 'it's all the operator's fault.' I know this because I'm a motorcycle safety instructor. While the probability of most m/c incidents can be mitigated to an extent by the rider, mechanical failures can and do happen, and can also be beyond the ability of the average rider to detect. You can't blame a rider, for example, for a sudden blowout caused by a sharp object that's all but invisible at highway speeds. But flying is different.

- weather - yes it's our responsibility to check, but to think that there are weather events that can happen suddenly without warning is to be purposely ignorant of the things that can happen.
- 'if you crash against a bird?' That's gotta be willfully ignorant on your part. My Inspire 2 went down when a small bird flew through my right rear prop while it was stationary, filming an event. How on earth does anyone avoid that? Birds fly at US - we don't fly into a bird.
- you don't know any pilots who have had an engine error? Lucky for you and them. I've had a motor fail for no reason at all (per DJI; it just died mid-flight). Just because it hasn't happened to you doesn't mean it can't happen.

We, as pilots, do have to take every precaution and check everything we can to avoid issues. Many incidents can definitely be pinned on the pilot; I know I've had my share. But to say that every crash is the pilot's fault is blame-y and inaccurate.

 

[BossBob]

RTH was never activated - it was in ActiveTrack continuously until impact. I'm not really familiar with how ActiveTrack works when combined with heavy stick input, but you were applying full forward elevator and full right aileron approaching the lighthouse, and the FC was yawing counterclockwise to keep it in view. But it looks like you basically just spiraled it into the structure.

View attachment 129408

View attachment 129409

From experience when in active track, if you give hard left or right aileron the drone will circle around the object being tracked, forward and backward moves closer or further away respectively. So the inputs indicated would lead to an ever decreasing spiral until the drone is unable to track the object.

 

[Icarus-50]

RTH was never activated - it was in ActiveTrack continuously until impact. I'm not really familiar with how ActiveTrack works when combined with heavy stick input, but you were applying full forward elevator and full right aileron approaching the lighthouse, and the FC was yawing counterclockwise to keep it in view. But it looks like you basically just spiraled it into the structure.

View attachment 129408

View attachment 129409

I have a file on my iPad that shows the drone on its route around the lighthouse and in the beginning there is a

From experience when in active track, if you give hard left or right aileron the drone will circle around the object being tracked, forward and backward moves closer or further away respectively. So the inputs indicated would lead to an ever decreasing spiral until the drone is unable to track the object.

my problems began when the drone went behind the lighthouse, lower than its height (so I could not see it) and I thought it took more than the expected time to come out on the other side, which I think prompted me to initiate RTH, although SAR says the records don't indicate initiating RTH. I am attaching a separate video that says RTH height was auto reduced to 30 meters, which is both weird and lower than the height of the lighthouse.

1. What are your RTH settings?

2. Does the lighthouse have any overhangs? If RTH was initiated and your RTH setting was for a return height higher than where you were flying, you could have climbed into an overhang since the MA2 does not have topside collision sensors.

3. If your settings allow sideways motion and if your RTH is set for "go around", any momentary object in front of you (like a bird) might have caused your drone to fly sideways into the lighthouse. You did say that you saw a series of strange objects on the monitor.

4. I try to avoid changing mode settings during a flight. I think the MA2 is supposed to handle that OK, but I've seen reports here that suggest that it isn't always the case. If the mode change from Normal to Tripod temporarily confused the drone, who knows what it might have tried to do.

5. How close were you trying to fly around the lighthouse on the second pass? I assume the frame of the lit portion of the lighthouse is metal, which could have disrupted your control signal and maybe even altered the path lengths (and therefore the timing, and therefore the position information) of the GPS signals if you were close enough.

I had not touched RTM since factory setting but it used it w/o issues; No overhang; I thought I was a good 8 feet from t Lighthouse when I saw it go around it.

 

[Meta4]

it took more than the expected time to come out on the other side, which I think prompted me to initiate RTH, although SAR says the records don't indicate initiating RTH.

You initiated Active Track at 379 seconds and didn't cancel it or initiate RTH before the collision happened at 416 seconds.
Active Track was cancelled (probably automatically) at 417.8 seconds as the drone was tumbling to the ground.

I am attaching a separate video that says RTH height was auto reduced to 30 meters, which is both weird and lower than the height of the lighthouse.

The RTH will only ever change if you change it.
The message about RTH height is just a notification that the RTH height was set at 30 metres.

The collision happened with the drone 142 feet above the launch point.
If RTH had been initiated, the drone would not have climbed above the lighthouse as it was already higher than the set RTH height (100 ft).

 

[Icarus-50]

RTH was never activated - it was in ActiveTrack continuously until impact. I'm not really familiar with how ActiveTrack works when combined with heavy stick input, but you were applying full forward elevator and full right aileron approaching the lighthouse, and the FC was yawing counterclockwise to keep it in view. But it looks like you basically just spiraled it into the structure.

View attachment 129408

View attachment 129409

From experience when in active track, if you give hard left or right aileron the drone will circle around the object being tracked, forward and backward moves closer or further away respectively. So the inputs indicated would lead to an ever decreasing spiral until the drone is unable to track the object.

My problems began when the drone went behind the lighthouse and I stopped see it. It was flying lower than the height of the lighthouse. I waited for it to reappear at the other end, but with the passage of a few seconds which I thought were normal, it did not reappear and this prompted me to initiate RTH, at least this is how I remember it while SAR here says the records don't indicate RTH initiation. I am attaching a video that says RTH height was auto reduced to 30 meters, which is weird as I did not make this change in the setting but more importantly, the RTH ceiling was set to a height lower than the height of the lighthouse.

You initiated Active Track at 379 seconds and didn't cancel it or initiate RTH before the collision happened at 416 seconds.
Active Track was cancelled (probably automatically) at 417.8 seconds as the drone was tumbling to the ground.

The RTH will only ever change if you change it.
The message about RTH height is just a notification that the RTH height was set at 30 metres.

The collision happened with the drone 142 feet above the launch point.
If RTH had been initiated, the drone would not have climbed above the lighthouse as it was already higher than the set RTH height (100 ft).

Thank you but I never touched the RTM height. But something unusual did take place the day before; when I took off near the beach on the sand dunes, the drone rose to about 20 feet or so & then stopped climbing & t controller/software announced “you have reached your maximum height”, again I had not touched this setting ever, since the factory setting. Since there were no consequences w this, I moved the max height slider to a more normal altitude, probably 200 or 300 feet & flew around w no issues. The height issue never came up again.

 

[sar104]

My problems began when the drone went behind the lighthouse and I stopped see it. It was flying lower than the height of the lighthouse. I waited for it to reappear at the other end, but with the passage of a few seconds which I thought were normal, it did not reappear and this prompted me to initiate RTH, at least this is how I remember it while SAR here says the records don't indicate RTH initiation. I am attaching a video that says RTH height was auto reduced to 30 meters, which is weird as I did not make this change in the setting but more importantly, the RTH ceiling was set to a height lower than the height of the lighthouse.

Thank you but I never touched the RTM height. But something unusual did take place the day before; when I took off near the beach on the sand dunes, the drone rose to about 20 feet or so & then stopped climbing & t controller/software announced “you have reached your maximum height”, again I had not touched this setting ever, since the factory setting. Since there were no consequences w this, I moved the max height slider to a more normal altitude, probably 200 or 300 feet & flew around w no issues. The height issue never came up again.

The RTH height was set to 30 m at the start of the flight, but it was completely irrelevant to the crash, which occurred because you actively flew the aircraft into the structure in ActiveTrack mode. There is no mystery here.

 

[Meta4]

Thank you but I never touched the RTM height.

It's RTH .. return to home.
There's been no suggestion that you did anything to change your RTH height.
The recorded flight data always shows what it is set to at the start of each flight, even if you haven't ever reset the height.
That's handy for people analysing the flight data.

But something unusual did take place the day before; when I took off near the beach on the sand dunes, the drone rose to about 20 feet or so & then stopped climbing & t controller/software announced “you have reached your maximum height”, again I had not touched this setting ever

This sounds like a completely separate issue and has nothing to do with RTH height.
If you had a Max Altitude reached message at approx 20 ft, that sounds like you launched without waiting for GPS.
After the drone acquired GPS that limit would have been removed..
It would be easy to confirm whether this was the case from a look at the recorded data.

 

[Icarus-50]

The RTH height was set to 30 m at the start of the flight, but it was completely irrelevant to the crash, which occurred because you actively flew the aircraft into the structure in ActiveTrack mode. There is no mystery here.

Thank you Sar, appreciate your time and knowledge. My mistake w t right/pitch stick.

 

[Icarus-50]

It's RTH .. return to home.
There's been no suggestion that you did anything to change your RTH height.
The recorded flight data always shows what it is set to at the start of each flight, even if you haven't ever reset the height.
That's handy for people analysing the flight data.


This sounds like a completely separate issue and has nothing to do with RTH height.
If you had a Max Altitude reached message at approx 20 ft, that sounds like you launched without waiting for GPS.
After the drone acquired GPS that limit would have been removed..
It would be easy to confirm whether this was the case from a look at the recorded data.

Thank you Meta. Makes sense.

 

[Icarus-50]

It's RTH .. return to home.
There's been no suggestion that you did anything to change your RTH height.
The recorded flight data always shows what it is set to at the start of each flight, even if you haven't ever reset the height.
That's handy for people analysing the flight data.


This sounds like a completely separate issue and has nothing to do with RTH height.
If you had a Max Altitude reached message at approx 20 ft, that sounds like you launched without waiting for GPS.
After the drone acquired GPS that limit would have been removed..
It would be easy to confirm whether this was the case from a look at the recorded data.

Good point.

 

[Icarus-50]

Am I the only one who will say that this flight was irresponsible on the part of the operator.
Luckily, no one got hurt by the drone or falling debris.

It was after the lighthouse had closed. There were no people or cars on the premises or in the vicinity. The parameter around the lighthouse is fenced.

 

[Van Belle Jean Marc]

I'm glad you said this, since motorcycling is a good example of 'it's all the operator's fault.' I know this because I'm a motorcycle safety instructor. While the probability of most m/c incidents can be mitigated to an extent by the rider, mechanical failures can and do happen, and can also be beyond the ability of the average rider to detect. You can't blame a rider, for example, for a sudden blowout caused by a sharp object that's all but invisible at highway speeds. But flying is different.

- weather - yes it's our responsibility to check, but to think that there are weather events that can happen suddenly without warning is to be purposely ignorant of the things that can happen.
- 'if you crash against a bird?' That's gotta be willfully ignorant on your part. My Inspire 2 went down when a small bird flew through my right rear prop while it was stationary, filming an event. How on earth does anyone avoid that? Birds fly at US - we don't fly into a bird.
- you don't know any pilots who have had an engine error? Lucky for you and them. I've had a motor fail for no reason at all (per DJI; it just died mid-flight). Just because it hasn't happened to you doesn't mean it can't happen.

We, as pilots, do have to take every precaution and check everything we can to avoid issues. Many incidents can definitely be pinned on the pilot; I know I've had my share. But to say that every crash is the pilot's fault is blame-y and inaccurate.

That is what i stated: There is ALWAYS risk and the driver/operator knows:
1. motorbiker or cyclist (did yesterday 120 km with >1200 m up and i have Garmin Radar for the dangerous vehicles after me):
https://www.komoot.nl/tour/371808297?ref=itd): legally most of the times we are considered as a weak user. As a teacher, i learn it is better to think of ourselves as the stronger: We dare it and are prepared this way to even die or crash. This attitude of high responsibility that everything is based on our fault, is the best survival mechanism we can have.
2. Weather can change: You admit you know, so knowing is accepting the responsibility. In your first topic you stated something on the road can show up unseen: That is because you drive to fast then. If you crash, in 95% of the insurance cases (i work in statistics as well for them), the accident could have been avoided if they drove slowlier or like me: they would have much less injuries. We know in insurances the risks models for men and women are different (men have more ‘perte total’, women have more sideview accidents and much lesser damage), but Europe forced us (Scandinavia) to use the same insurance prices so women motorbike drivers pay more for their insurances than they should pay now, none if my women pupils on motorbikes EVER had a crash with another vehicle, but some slips in the sand).
3. Birds are in nature. They are meant to fly. If they crash in your propellor: You are the only one who should feel guilty. This is one of the reasons they softened the propellor blades. Like the Tesla, we caused MORE accidents with low noise with other ‘trafficants’ (here birds, Tesla: pedestrians and cyclists) so every improvement creates new casualties. When i see a bird close to my drone, i put it in Sports modus and go up to max power so the bird knows my drone is dangerous ‘bird’.

i would like to see other critical comments from others here.

Last: Every pupil of me has its own driving or flying style. After the exam i support them following THEIR personal style, not to take over mine. Even in traffic in Sofia, if all drivers would be really defensive, on a crossroad with 4 cars, they would stay waiting for each other there for ever. Fast accelerating cars even come in head of the traffic lights and it does not FEEL right, but they make the traffic going more smoothly and it made me a better driver knowing there are always ‘dangerous cowboys’ around me. I just did one criminal case with a procureur and the man’s insurance will pay a life time long a woman who was hit while passing all other cars not seeing the woman that broker her leg while crossing near 5 Corners in Sofia... thanks to my dash cam. But he showed remorse and admitted he was the only one to blaim, so neither the victim, procureur or me are mad at him.

Starting drone pilots and children MUST make errors to learn.

Let us accept paying a price for the risks we take: WE take (always will). +je

 

[test2000Anafi]

The main error is not flying VLOS:

it is completely possible that a ninja was hiding behind the lighthouse, and it hit your drone with his ninja-stick when he decided the drone came to close to him and was a threat. In feudal Japan they had primitive drone but it would happen often to young inexperienced drone pilots not flying VLOS, a ninja would throw a shuriken at a drone once it was out of sight of the operator

 

[Drones.n.Pixels]

Not long ago I lost signal on my MA2 when I flew it below a ridge line and it triggered an automatic RTH. The camera footage did not show any "retrace 50 meters" ... the drone simply turned, climbed to the preset elevation, and started to return home. At that point of course I reaquired signal. I don't remember any such "retrace 50 meters" in any of my documentation, and in any case I wouldn't want it to do so.

I also did not see the „retract 50m” behaviour in my RTH instances

 

[Mirlo36]

the drone went behind the lighthouse, lower than its height (so I could not see it)

Sorry off-topic: free nation, free rules.

In Finland this would be illegal - loosing VLOS.
It is allowed only with special license, which is not for recreational flyer.

Of course you have to have Drone Operator License anyways, even for Mini 2. One year 30€ / 37$.

We have a lot of regulations, and more is coming all the time...

 

[DameEdnaBrown]

I was on a POI flight around a lighthouse in Outer Banks, NC. It flew around it and on the second round, I decided to fly it manually in Tripod mode to get some good video and fly by the tower, the top part where the actual rotating light is. As the drone went around it, I could no longer see it but I noticed a series of strange images on the monitor and then it went dead. I walked behind the lighthouse and the drone was on the grass, with its lights flashing. One propeller was broken, the others were all scratched, the camera was dislodged and dangling on a wire, there was a loose wire next to it, one camera bracket was on the grass. I was crushed. Lots of concrete on the top of the drone, which to me indicated that it had rubbed against it, either as it tumbled down or its top hit the side of the lighthouse - cant understand how or what happened. I cleaned the drone and started it just to check. It started but the camera would not turn while on the ground. I did not start the motors or do anything beyond just starting the drone (the buzzing sound). My assumption is that the drone was drawn to the lighthouse for some reason and my last minute controls did not respond. A couple of questions: 1-What in a lighthouse can interfere with the drone? 2-Are there magnets in the lighthouse, esp the top part or is it the concrete the culprit? 3-Who do I send the drone for repair and diagnostics (DJI or an authorized repair shop)? 4-Is it better to replace it? I am a serious hobby photographer so with the Mavic Air 2s in the market now, should I dump this and buy the 2s, although the difference, even with a 1" sensor, would not be significantly different for home videos. 5-Would any repairs on this damaged drone make it as good as new? Thanks for your thoughts. I have been flying drones for years, Yuneec, DJIs, and others so I am not a novice pilot. No pilot error, unless one considers proximity to the lighthouse to be one.

I think it was simply a case of the drone losing the signal when the lighthouse blocked it out then you have no idea what it's going to do. I would have thought it should have gone up to its set RTH height

 

[Meta4]

I think it was simply a case of the drone losing the signal when the lighthouse blocked it out then you have no idea what it's going to do. I would have thought it should have gone up to its set RTH height

This has already been looked into and it's been established from the recorded flight data that:
The drone never lost contact with the controller
It was already higher than the RTH height
The drone did not enter RTH
The drone was still in active track and the flyer fought against active track and flew the drone into the lighthouse

 

[Wash Pilot]

Day time; Hatteras. I filmed 2 lighthouses on this trip too; love the perspective. Yes, t obstacle sensors didn’t stop the crash.

Hatteras Light is on National Park Service land (Cape Hatteras National Seashore). Definitely a "no drone zone."

 

[JAW]

Hatteras Island, on which the lighthouse sits, is about 70% National Park land. Meaning No Drones. And yes, the park gendarmes actively enforce the rule. First time offenders usually get a verbal warning and a short dissertation on the rules pertaining to UAS opertaions in National Parks. While I have never gone past the "first offender" level, I am told second offenders get their equipment confiscated.

Still, it is possible to legally photograph the lighthouse without violating National Park rules. The actual shoreline up to the primary dune is NOT part the of National Park. This means a beach takeoff which at its closet point to the lighthouse is about 1500' away and requires a vehicle capable of handling the sandy oceanfront trails in order to get to the take-off point.

But what about VLOS requirements? If you can see your drone at a distance of 1500', you're good to go.

 

[Icarus-50]

Hatteras Island, on which the lighthouse sits, is about 70% National Park land. Meaning No Drones. And yes, the park gendarmes actively enforce the rule. First time offenders usually get a verbal warning and a short dissertation on the rules pertaining to UAS opertaions in National Parks. While I have never gone past the "first offender" level, I am told second offenders get their equipment confiscated.

Still, it is possible to legally photograph the lighthouse without violating National Park rules. The actual shoreline up to the primary dune is NOT part the of National Park. This means a beach takeoff which at its closet point to the lighthouse is about 1500' away and requires a vehicle capable of handling the sandy oceanfront trails in order to get to the take-off point.

But what about VLOS requirements? If you can see your drone at a distance of 1500', you're good to go.

View attachment 129575

It was Bodie, not Hatteras.