If we first assumme that we are taking about two acoustic instruments being played in a real space, there are a few things we have to do to get this to happen, all of which are hard:
We have to find two instruments that can produce very similar waveforms, and play them with very similar technique such that they produce those very similar waveforms. This is very hard, as all real acoustic instruments have complex behavior, producing waveforms consisting of many harmonic and inharmonic partials that can shift in amplitude and frequency; slight variations in playing technique will change how each of those behaves. Any slight difference in the frequency of a prominent partial between our two instruments may produce an audible chorusing effect.
We have to start the notes at exactly the same time when we play them, by which i mean down to a tiny fraction of a millisecond, otherwise you will get comb filtering effects, and possibly a smearing of the initial transient (to which the ear is very sensitive).
The two instruments have to seem to be in the same place. The auditory system has a variety of ways of locating and separating sounds, and in any real environment, it is likely to have enough clues to separate two sound sources playing the same thing.
If we can somehow overcome those issues, then possibly we could get two instruments to sound like one. It's just that all those things are hard to overcome.
As Tetsujin says in the comment, a computer playing back samples can easily make two notes sound like one - precisely because it can overcome all these issues. It can ensure that both parts play the same waveform; it can start playback at the same point in discrete (sampled) time; and it can mix the waveforms in a way unaffected by acoustic space.