I mean the usual thing by 'measurement' or 'observation'

What about it?

I am not a physicist although I have a math background.

 

I mean the usual thing by 'measurement' or 'observation', a process that sends the wave function to an eigenvector of an observable, which is a Hermitean operator on the Hilbert space of wave functions for the experiment. My question is this: apart from measurement, the wave function dynamics are given by the Schrodinger equation so the wave function evolves by unitary transformations U. Is a measurement in fact an example of such, perhaps in a limiting sense of some complicated quantum process ? Must it be regarded as a distinct type of process? Penrose talks about this a lot in 'Road to Reality', but I have the sense that professional physicists sort of assume measurement is not a fundamentally distinct thing... is that right?