DJI Mavic 2 Pro - Positive Feedback Loop Crash

[sar104]

Thanks sar, also found these in the event stream log after I posted.


The entry at T-18.518 seems to indicate a switch from IMU(0) to (1) and FUSION errors are reported later on IMU(1). But note that there are also GPS errors as early as T+47s. Could the GPS errors be the root cause here?

No - you can see the direct cause in the graphs of the IMU rate gyro data. The GPS data are fine, and the flight was good until attitude sensing failed at 374 s.

 

[sar104]

Isn't all FC control, even recalibration mid flight (yaw correction) "trial and error"?
I realize my proposals are oversimplified, but I don't think it is as impossible as you make it out to be either.

No - sensor fusion flight control is not based on trial and error at all - if that's your misconception here then you need to read about how it works. Your proposal is not oversimplified, it's incorrect.

 

[slup]

Thank you all for the great input.

Additional question: I bought this unit from someone online used. Everything worked great for the month or so I owned it until this. Would it be worth reaching out to DJI Support in hopes that they agree with all the above supporting information and determine it is a hardware/software issue. Is it likely that DJI would provide a new unit, discount on a new unit, none of the above? Suggestions on moving forward?

Also, moving forward, how can I and other pilots avoid this issue or something similar to this?

Thanks again for all the help and knowledge!

In this particular case it's been pretty well established that it was HW not pilot error ... all man made items will eventually fail at some point, others sooner then later. With that said, the only thing we can do is to plan for that it will happen, meaning minimize damage if... My impression though is that pure, not beforehand detectable HW failures is pretty rare compered to the vast majority of pilot errors and neglect when it comes to having proper knowledge achieved before take off.

What will DJI do ..? Well it depends. Is it still under warranty ... then you will have the greatest possibility. If not, then it's up to their goodwill. Why not contact them.

 

[Prismatic]

If that were true, Sar wouldn't be arguing against my point.

You can't be serious. The flight control always knows the x-, y-, and z-axes relative to the aircraft. How could "forward" possibly be unknown?