BNO055 Heading Accuracy
We are trying to use the BNO055 for accurate heading (±2 degrees). We want to use the device permanently mounted on a 3-meter mast that extends from a heavy hard iron vehicle. It is important for us to be able to calibrate the device once and store the offset values between power-ons. We intend to initially calibrate it in place on the mast by driving the vehicle around in figure eights. We intend to configure the device in fusion mode. Is there a specific mode that you would recommend for stable and accurate heading? Is there a way to disable auto calibration once the initial offsets are stored? Secondarily, will the sensitivity of the magnetometer be such that the heading would be significantly perturbed by passing vehicles or other metallic interferences? Finally, is there an alternative device such as the BNO080 that would be more appropriate for our application?
Re: BNO055 Heading Accuracy
For BNO055 in this case, you would use NDOF mode. As long as the mast is not ferromagnetic, 3 meters distance to the vehicle should be enough to get a good accuracy. In your validation tests, once gyroscope and magnetometer calibration are 3/3, verify that while in operation the magnetometer calibration stays 3/3. The contrary would indicate distorted magnetic field.
If the device is rotating relatively quickly, then NDOF mode is well suited. In case of very small rotation speeds, COMPASS mode should be used.