Mavic 2 Pro Crash out of Control

[AtortPhotography]

It picked up momemtum to fly backward as seen near the end of the footage because full backward elevator control was applied and held for as long as 2 seconds. Such movement has nothing to do with ATTI mode.

So what do you think or are saying happened then?

 

[BudWalker]

So basically the remote control connection strength correct? I have a smart controller and then use the Fly App for the other drones. So you really think this is more crucial than the amount of satellites connected?

I don't understand how you thought my comments were about the remote control connection strength. But, no, it's the Flight Controllers confidence in it's location solution that matters. And that is displayed on both the RC and the Fly App.

 

[AtortPhotography]

I don't understand how you thought my comments were about the remote control connection strength. But, no, it's the Flight Controllers confidence in it's location solution that matters. And that is displayed on both the RC and the Fly App.

I have to find some articles on this so I can research more and know. I Googled and really that doesn't seem to pull up much with "flight controller confidence" at least. If you have anything good to read appreciated. Thanks.

 

[boblui]

So what do you think or are saying happened then?

If you pull the elevator stick fully backward and the craft responded accordingly then the craft was largely under contol. Sorry, I have to agree that the crash was due to pilot error.

 

[AtortPhotography]

If you pull the elevator stick fully backward and the craft responded accordingly then the craft was largely under contol. Sorry, I have to agree that the crash was due to pilot error.

No sorry needed. What I think exacerbated this was the fact I was in a narrow space with two tall buildings which winds always ramp up and an issue let alone in atti mode unexpected not prepared. Pretty much a recipe for disaster. I think would have been challenging honestly for anyone in this little corridor in atti mode even with experience.

 

[boblui]

No sorry needed. What I think exacerbated this was the fact I was in a narrow space with two tall buildings which winds always ramp up and an issue let alone in atti mode unexpected not prepared. Pretty much a recipe for disaster. I think would have been challenging honestly for anyone in this little corridor in atti mode even with experience.

The problem would have been very much reduced if it was at day time because vision positioning would work. I believe turning on the light at the belly would have help also.

 

[Meta4]

I Googled and really that doesn't seem to pull up much with "flight controller confidence" at least.

The flight controller (which is inside the drone) looks at the number of sats and how well their spread is to assess the reliability of the location data.
If you are out in the open, GPS Health will always be 4 or 5.
In a situation where the number of sats is low and/or they are not spread well across the sky (which happens when only a small area of sky is visible), the GPS health drops and if it's 0 or 1, the flight controller has no confidence in the data and won't use it as it is likely to include significant errors.

 

[AtortPhotography]

The flight controller (which is inside the drone) looks at the number of sats and how well their spread is to assess the reliability of the location data.
If you are out in the open, GPS Health will always be 4 or 5.
In a situation where the number of sats is low and/or they are not spread well across the sky (which happens when only a small area of sky is visible), the GPS health drops and if it's 0 or 1, the flight controller has no confidence in the data and won't use it as it is likely to include significant errors.

So on this image...the furthest from the left of course is satellite strength....then you have the bars and the wifi signal. The wifi signal is the connection strength from your drone to controller and the bars in the middle are what is most important in a way at least with being concerned with being thrown into atti mode?

 

[Meta4]

So on this image...the furthest from the left of course is satellite strength

No .. see post #100 and 105 again
GPS satellites are a constant distance from earth and the GPS signal strength doesn't vary that much.
The flight controller monitors the number of satellites and their spread.
The bars represent the flight controller's assessment of its confidence that the location data is good to use.

the bars in the middle are what is most important in a way at least with being concerned with being thrown into atti mode?

If you fly in the open where the drone gets a clear view of most of the sky, you should never have fear of "being thrown into atti mode".
If you fly where >half of the sky (and the sats in those parts of the sky) is blocked, you risk having unreliable GPS and flying in atti mode.

 

[AtortPhotography]

The flight controller (which is inside the drone) looks at the number of sats and how well their spread is to assess the reliability of the location data.
If you are out in the open, GPS Health will always be 4 or 5.
In a situation where the number of sats is low and/or they are not spread well across the sky (which happens when only a small area of sky is visible), the GPS health drops and if it's 0 or 1, the flight controller has no confidence in the data and won't use it as it is likely to include significant errors.

That's super interesting. Thanks for the clarification.

 

[Cymru]

Available satellite geometry is far more important than number of and to an extent signal strength.

The DJI drone is using geometry and maybe DOP to put on a 1-5 scale of " quality".

 

[BudWalker]

Available satellite geometry is far more important than number of and to an extent signal strength.

The DJI drone is using geometry and maybe DOP to put on a 1-5 scale of " quality".

The quality (aka navHealth or gpsLevel) is more than the DOP. The drone's flight controller is continually using several inputs to compute the drone's location. That's why the time/space resolution is better than what can be achieved from GPS alone. If, e.g., the flight controller decides that the heading data may not be valid then the location solution may also be invalid. I.e., the confidence in the location solution is lowered, usually to 0/5.

 

[Cymru]

Yeah its more than DOP but it does use it (or a form of it).
Ive seen people have some near misses because someone people seem to think "number of satellites" is important.
No point seeing 10 satellites if they're all in a tiny quadrant of sky...

 

[MavRik]

Well ... if the flight wasn't conducted near Astorga, Medellin, Colombia (as suggested by the DJI tech support & the GPS position) they have mixed up the cases.

Attach your TXT & DAT log in a new post here (both will be located in the mobile device you flew with). Go here & read up on how to retrieve them --> Mavic Flight Log Retrieval and Analysis Guide (read under section 3.)

I also plugged those reported coordinates and agrree, Medellin, Colombia is where Google Maps sent me. It appears the Technician has mixed up the case here. So....if they got that wrong....

 

[AtortPhotography]

I also plugged those reported coordinates and agrree, Medellin, Colombia is where Google Maps sent me. It appears the Technician has mixed up the case here. So....if they got that wrong....

You think DJI could have fouled up?

 

[DanMan32]

So on this image...the furthest from the left of course is satellite strength....then you have the bars and the wifi signal. The wifi signal is the connection strength from your drone to controller and the bars in the middle are what is most important in a way at least with being concerned with being thrown into atti mode?

It's probably clearer on the RC LCD display as both GPS confidence and RC/AC strength are represented as separate bars.

One of the things I especially appreciate about the M2 remote.

 

[DanMan32]

The problem would have been very much reduced if it was at day time because vision positioning would work. I believe turning on the light at the belly would have help also.

Landing lights don't reach very far. I figured that out on my very first test flight of my M2. By the time I finished work, set up the M2 and got out to the park, it was quite dark. I could hardly see anything off of FPV with landing lights on and gimbal pointed down.

 

[Thomas B]

It's probably clearer on the RC LCD display as both GPS confidence and RC/AC strength are represented as separate bars.

One of the things I especially appreciate about the M2 remote.

And on the CS too. Agree, easy to find and read

 

[MavRik]

You think DJI could have fouled up?

It certainly appears that way!!

 

[slup]

You think DJI could have fouled up?

Based on the log you shared... that will not make any difference I'm afraid.