G'day,
I've read through this document from STMicroelectronics, regarding the calculation and application of compensation values/matrices to pre- or post-fused accelerometer/gyroscope data to account for incorrect module installation (with respect to the desired reference orientation).
I don't require absolute position in my application, so hadn't planned on using a magnetometer (acceleration, pitch, roll and yaw rate are all I'm after). Can anyone confirm for me whether the Euler angle/quaternion approaches listed in the above link (either pre- or post-fusing in the DMP) will work for a 6-axis module, or will I need yaw angles as well (and hence a full 9-axis module). The introduction of the document suggests accelerometer and gyroscope are required, but then uses yaw (psi) throughout the calculations, so I'm somewhat at a loss.
I can follow up with STM directly, but I plan on using an InvenSense chip so figured I'd ask here first. I have next to zero experience with IMUs, so this may well be a relatively straight forward answer.
For simplicity, assume my module is installed with the correct yaw, but with incorrect pitch and roll.