I am using the BNO055 sensor to measure the YAW in a vehicle in order to determine the turning radius (to identify whether the car is turning right or left). The sensor works correctly while driving, but I am facing a specific problem when the vehicle is stopped with the engine running.
When the car is stopped and subjected only to engine vibration, the YAW value starts to increase gradually, without returning to the correct value until the vehicle is turned off. This leads to an incorrect reading because I would like the value to remain the same and only change when turning or something like that.
In this case, I used the NDOF Mode: It works very well to calculate the turning radius while the car is moving, but the problem of the YAW value increasing occurs when the car is stopped and only the engine is vibrating.
Fusion M4G Mode: In this mode, the YAW does not increase gradually but oscillates slightly with the engine vibration, which is expected (I believe that the gyroscope is responsible for this gradual increase since it is disregarded in this mode). However, in turning radius tests, the M4G's performance is not as good as that of the NDOF mode, which offers more accurate results.
I would like to know if there is any way to mitigate this problem or any specific configuration/calibration recommendations to deal with engine vibration when the vehicle is stopped.
