We recently faced with a huge problem that makes indoor flights almost impossible: vibration. The flight computer is undamped and vibration reaches 0.8G in X/Y axis.
Does anyone has an experience of damping this particular frame? Did anyone try to replace motors to those one that vibrates less? Default ones are not isolated at all and frame shakes like crazy.
According to the documentation (https://github.com/intel-aero/meta-intel-aero/wiki/03-First-flight 03 First flight · intel-aero/meta-intel-aero Wiki · GitHub), for indoor flights you should use Manual or Altitude Hold modes.
INDOOR FLIGHTS: use Manual (mode 0) or Altitude Hold (mode 1). Manual flight mode requires reasonable piloting skills. Until you are experienced, use caution while operating in this mode.
That being said, we tested the Aero RTF indoor in Manual mode and did not see any vibration.
We recommend you to flash the latest firmware (1.5.1) and to update the BIOS, FPGA and Flight Controller as well:
Please also do a frame reset and go thru the calibration process as it's described here: https://github.com/intel-aero/meta-intel-aero/wiki/02-Initial-Setup# calibration 02 Initial Setup · intel-aero/meta-intel-aero Wiki · GitHub
Have a nice day.
Drone doesn't vibrate visually. It vibrates on very high frequency(according to FMU measurements) due to motors and propellers. In offboard mode(I don't use remote controller) with LPE this is crucial for velocity estimation.
BTW, is there any way to save ALL the messages that are generated by FMU? Currently with QGC I'm able to get ~3Hz for HighresIMU messages and ~0.5Hz for local position messages due to maximum bandwidth limitations.
Could you please provide additional information on how you are analyzing the non-visible vibrations? Also, we'd like to suggest you to try the EKF2 estimator as it handles vibrations much better than LPE.
Regarding to your question about the log files and messages generated by the FMU, we're still looking into it. As soon as we have information about it, we'll post it in the thread.
Have you had any chance to confirm if the same behavior happens when using the EKF2 estimator?
We recommend this tool since it handles vibrations much better than LPE.