Pix32"Pixhawk" Holybro FC Plane 3.9.10 suddenly goes crazy during flight. Nav.Roll and Nav.Pitch freezes during mission. Intermittent Problem


I have been experiencing this problem for more than 2 weeks and I couldn’t figure out what is the root of the problem. The problem sometimes happens in the middle of the flight and stops, and sometimes it persists until the flying wing lands. The setup I have is:

1 meter wingspan, 1,3kg Flying wing,
T motor 2212 motor and T motor AIR 20A ESC
Holybro Pix32"Pixhawk" Flight Controller
Holybro Pix32 digital differential airspeed sensor (Measurement Specialties 4525DO sensor)
UBLOX M8N module, as well as IST8310 compass
2x BlueBird servos BMS-390DHM

I have built many airplanes with this same setup as well as used the same parameters configurations and all of them flies pretty well doing photos missions. It takes off, flies the predetermined path and lands all automatically.

The problem is that during the flight, suddenly the flying wing starts to roll and pitch pretty crazy. The plane rolls really fast to both sides, it goes from -30 to +30 degress really fast and keeps doing it as crazy and also starts to climb oscilating the pitch together. Sometimes it starts and finishes the oscillation during the flight, but sometimes it lasts longer and is very scary.

On the logs I have noticed that the nav.roll and nav.pitch they just stuck at some angle and the roll and pitch oscilations get really big. It all hapenns with the airplane on Auto mode flying a mission. The airplane seems just to get out of control. When this problem happens the plane can’t just maintain its mission path.

Attached here I have 2 logs. The first one was a flight that the problem didn’t happened, I was tunning the PID for roll and pitch to check if this was the cause of the oscilations, and the airplane flew well. On the second log, it was the next flight and on this one the problem suddenly happened and this time was really scary.

First Log (great flight): 00000012.log
file: https://drive.google.com/open?id=1MCEUx36G69YAQTK23gm8Iv0MeQcOmIc2

Secong Log (the one the problem happened again 2 times during the flight): 00000013.log
file: https://drive.google.com/open?id=1y47m8Gyfg6uldnAhUUb4Ij41xGRcrtrW

The problem happened with 2 different FC, both new, and I have already tried using new servos.

Could someone with more experience analyze the logs and help me find the root of the problem?

I would appreciate it too much. Thanks in advance

The phenomenon I encountered
It should be after 3.9.4. In vtol mode, after the GPS gains a steady signal, all fixed-wing rudder surfaces will appear as you described and all rudder surfaces will lock or sway. However, in fixed wing mode, the rudder surface is normal.

The problems in vtol mode have been bothering me for a long time. I hope to add a parameter to choose whether to lock the rudder surface in vtol mode or stabilize the rudder surface control body.