We are using the ADIS16210 for the sake of pitch and roll measurement on a stationary platform.
1. The datasheet indicates that the device can be set up for inclinometer mode or for accelerometer mode (default is inclinometer mode).
It specifically says that selecting accelerometer mode "disables a nonlinear correction term that is only relevant in the angle computation".
2. Even though I need the device for measuring roll and pitch, I intend to use accelerometer readings (XACCL_OUT, etc.) rather than inclination readings (XINCL_OUT, etc.) since:
a) I need to measure the roll and pitch of the platform rather than that of the accelerometer itself, which is most easily done using the gravity vector components measured by the accelerometer, transforming them to platform body frame and then computing pitch and roll from the results.
b) Inclination data provided by the device does not include a usable roll computation (or perhaps I am missing something). Specifically, assuming, for example, the common aerospace body frame definition:
pitch = atan(-aX / sqrt(aY * aY + aZ * aZ))
roll = atan2(aY / aZ)
The ADIS16210 does something akin to the the first calculation, but not the second.
3. I have the following questions:
a) What is the "nonlinear correction term that is only relevant to the angle computation" and can I safely use the device in accelerometer mode even though I intend to use it for measuring pitch and roll?
It is possible/reasonbale to set up the device to inclinometer mode and still use acceleration readings (XACCL_OUT, etc.) for the sake of calculating pitch and roll?
b) Is it possible to extract roll from inclination data provided by the device (XINCL_OUT, etc.), or should I compute it from acceleration data, as outlined above?
Thank you very much for your assistance.