IMU Brick 2.0 accuracy compared to iPhone

[jones5322]

I'm trying to understand my IMU Brick 2.0.  What should the euler angle H be when X is pointing to 360 degrees?  Mine is showing 244. (see attached)  My firmware appears to be 2.0.9.  I have not run any calibration processes.

[borg]

I think it should show ~0/360° if in your image the USB connector goes to the bottom (or maybe to the top? Not 100% sure).

 

I would expect that there will still be a discrepancy of a few degrees between the phone and the Brick. The earth magnetic field is quite weak and there are lots of small magnetic fields generated in the phone as well as the Brick.

 

Do you have a real compass for comparison?

[jones5322]

Using a compass and confirming I was pointing North the euler H reads 259 when IMU x axis is pointing North.  Brick View is reporting green on the calibration page.

 

What is the process for doing a manual calibration?  I only see a button to save calibration.

 

I want to use you IMU in ROS like https://github.com/nxdefiant/ros_wild_thumper/issues/1  Matt Droter and I are working on this together to understand the mounting position and any needed offsets.

imu_testing.pdf

[borg]

The BNO055 uses an automatic continuous calibration. The chip tells us when it thinks it is currently calibrated well. Then all of the calibration parameters turn green (in Brick Viewer) and you can press the "save calibration" button. The current calibration is then saved and will be used on next startup as a starting point for the automatic continuous calibration.

 

If you leave the IMU running for a while (if possible move it around a lot during the time), does the orientation then change or is it still 45° off?

 

For the exact specifications and description of the calibration you can take a look at the datasheet: https://github.com/Tinkerforge/imu-v2-brick/raw/master/datasheets/BNO055.pdf

 

See 3.10 for the automatic calibration description.