Hi Tony, thank you for your suggestion. I managed to access BlueOS, it turns out the SD card was corrupted, so after re-flashing it, everything worked.. Now I’m moving on to the next steps, which is vehicle calibration. On the Motor Config page, I’m trying to test the motors in QGroundControl, but I keep getting this error during the motor test:
“MAV_CMD_DO_MOTOR_TEST command failed.”
Any ideas what could be causing this issue?
Thank you, and we appreciate your guidance.
Hi @Nurul -
Glad you resolved the initial issue.
Before trying to do the motor test, have you verified in BlueOS that the ArduRover firmware has been loaded?
Has the vehicle connected with QGC ok? Can you arm and cause the motors to move with the connected joystick?
You should also make sure that the vehicle setup page is not open at the same time as QGC - I prefer to use BlueOS to do motor tests (with QGC closed) and we will be updating our guides eventually to drop QGC usage all together…
Hi Tony,
Happy New Year, and thank you for your previous guidance. We have successfully resolved the software update and vehicles setup issues.
We are now planning to deploy the BlueBoat for an upcoming mission. As part of our preparation, we conducted initial testing in a swimming pool using Auto mode. However, we keep encountering a warning that states “Prearm failed: compass inconsistency.”
We have attempted to recalibrate the compass several times, but the issue persists. This is a bit puzzling, as all calibrations were completed successfully and the system status indicators are showing green.
Could you please advise if there are any additional steps we might have overlooked, or if you have any suggestions on how to resolve this issue?
Thank you in advance for your support.
Hi @Nurul -
It’s likely the pool you’re testing in has strong magnetic fields, from iron rebar in the walls or strong electrical currents. Pools are generally too small for testing auto mode…
To override the pre-arm checks, simply remove the check next to compass, under autopilot parameters, searching arming.
Hi Tony,
Thank you for your support.
We have moved our testing location to a lake; however, the issue still persists. When we disable all compass modules, the boat is able to start moving.
However, a new issue arises: although the boat moves, it does not follow the waypoint directions that we have set.
We are currently trying to test the auto mode, but the process keeps failing due to these compass-related and navigation issues.
We would appreciate your guidance on how to resolve this so that the boat can follow waypoints correctly in auto mode.
Hi @Nurul -
If you disable all the compass modules, then the vehicle doesn’t know where it is pointing - thus auto, guided, acro and loiter modes would all fail to function!
I suggested you eliminate the pre-arm checks, not disable the compass completely…
You should run a fresh compass calibration, and “do the dance” away from any large metal objects. Generally I only disable a compass if the calibration score is in the red - if yellow or green, it contributes to the EKF ok. You can calibrate the compass, gyro, and accelerometer from the BlueOS vehicle setup page, with QGC and Cockpit closed.
Hi Tony, thanks for the clarification.
I followed your recommendation to run a fresh compass calibration and “do the dance” well away from large metal objects, and repeated the calibration after relocating to an open coastal environment. Despite this, I am still encountering persistent Pre-Arm: Magnetic Field errors (e.g. XY diff exceeding threshold).
You mentioned eliminating the pre-arm checks rather than disabling the compass entirely. Could you please advise which specific pre-arm checks or parameters should be adjusted on the BlueBoat / BlueOS platform, and the recommended way to do this safely?
Our immediate goal is to demonstrate reliable waypoint following and record the mission execution for validation purposes.
Hi @Nurul -
It’s unusual that a fresh calibration didn’t fix things. Were you inside when you performed the calibration? Repeating it on location should not be required.
I provided instrucitons on the parameter already - the specific term to search and edit is Arming_Check:
Hi Tony, thanks for the suggestion.
I just want to clarify, if we uncheck the compass pre-arm check under ARMING_CHECK, wouldn’t the underlying issue still remain and we’re essentially just bypassing or ignoring the compass problem rather than actually resolving it?
In your experience, is it acceptable to operate the BlueBoat in AUTO / waypoint mode with the compass pre-arm check disabled, or is the pre-arm check considered mandatory for safe and reliable autonomous operation?
I’d really appreciate your advice on whether this should only be used as a temporary workaround or if it’s an accepted practice in certain use cases.
Hi @Nurul -
In most situations, you would not want to disable this check. However, when launching from locations with large ferrous metal objects or large engines / electrical currents present, disabling the check may be necessary if on-site location does not overcome the issue.
In such situations, I manually control the BlueBoat some distance from the launch point / interference source, and then verify the heading of the vehicle in the GCS against reality. If everything looks good, I tell the vehicle to loiter or go to a location with Guided mode. If this behaves normally, normal autonomous operation can be expected. You could apply the failsafe again at this point, but that risks the system refusing to arm at a critical moment during recovery, assuming the magnetic interference is still present !
