Hi,

I'm new to Robot localization package. After reading through the documentations, I'm still a little vague about how can I properly fuse the sensors, so here are some questions and I hope to get as much information as possible. These are the sensors I'm fusing:

1. Wheel encoder odom: frame ID odom and publishing at 100 Hz
2. IMU in ENU convention: frame ID imu_link and publishing at 50Hz
3. Stereo Camera's odom: frame ID odom and publishing at 15Hz
4. Stereo Camera's IMU NOT in ENU convention: frame ID zedimulink and publishing at 350Hz

These sensors have different frame, different initial values and different publishing rate. How can I make sure all the sensors are aligned (such as having the same absolute heading or pose)? I realised that if I fuse the yaw from different sensors the result is oscillating. Do I need to do a sensors calibration? If yes how can I do that?

Secondly, I am not sure that if I should publish the odom TF in this case since I have a lidar localizer node that performs NDT scan matching and provides the pose and TF straight from map to base_link. How can I make use of robot localization along with NDT matching?

Any help or guidance in regards of this would be much appreciated. Thank you.