However, it might seem like a trivial task to measure rotation knowing rotational speed from sensor of iPhone, it's actually not so easy. Integration in time gives errors and result to drift of measured angle. Over easy to cope with the drift is to account for gravity direction. More details in my post in Towardsdatascience.
I guess you’re the author and not a native English speaker, so here’s a language tip. Nice article, but harder to understand than needed because it uses “rotation” for both the direction the phone is pointing at and changes to it.
The article would be easier to read if it used “orientation” for the former (https://en.wikipedia.org/wiki/Rotation: “Rotation is the circular movement of an object around an axis of rotation” and (under “See also” “Orientation (geometry) – Notion of pointing in a direction”)
