Has anyone idea what is mathematical/physical interpretation quaternion accuracy?
quaternion.accuracy() method is available and works but I do not really understand how it should be interpreted. The only thing I know is that after some movement of nicla, autocalibration works and it gets lower, what means results are more reliable.
In reality, accuracy starts at 3.14 after power on, and after autocalibration of Nicla it goes to 0.44 But I have no idea what does it mean...
One very simple test would be to convert the quaternion to Euler angles, and determine when yaw = 0 degrees, whether the sensor accurately points to magnetic or true North.
Then rotate the sensor in steps of 90 degrees in the horizontal plane and see if it accurately points East, South and West.
After autocalibration quaternion accuracy is 0.44 and directions are correctly measured. But what this magic 0.44 means? J
The code should be open source, so look through that to see how it is calculated.
directions are correctly measured
Define "correctly measured".
To make a meaningful assessment of the measurement error, you could, for example, calculate the root mean square difference in degrees between the true and measured 0, 90, 180 and 270 degree compass directions.