Could some one help to make me my two thoughts clear.
first:
We can only chose (orient to zpoint-xpoint ) to make the dependency of orientation, but never zpoint-ypoint.
Is the reason that the Euler angle should be 313, it means rotation about ZXZ? There may be a singular result if the y-angle in n*180deg .
second:
For example. I want to make the y-direction of the construction frame( cm-marker, inertia_frame marker) as the axial-axis of steering column part with forced. There may be occur error in particular orientation during simulation?
That article is a little bit pessimistic. Particularly if you have the "bad" orientation in your model, Adams will fix that immediately at t=0 and you will never see any problem with it.
That said, when I model systems with many rotating parts (think a chain or tank-track) I make sure that the Z-axis of the parts are all oriented "sideways" so that they never align with global Z during normal simulations.
Go ahead and parameterize the model the way that makes most sense to you and in almost all cases it will work out just fine.