A couple of years ago similar articles were saying that Chrome OS was going to replace Android, so I'm a bit more cautious about reading the tea leaves than this.
And some of the things he says don't make much sense: the "fragmentation" problem isn't because of anything to do with the Android OS, it's what happens when you allow many manufacturers use it on many different devices with it up to them to build and distribute updates. But that's not going to be solved by a different OS base unless Google also undertake building and distributing updates for all devices out there (including removing carrier's fingers from the software), and that's not going to happen. So unless they cut the OEMs out (suicidal) I don't see that changing significantly. Similar comments apply to performance issues - even forbidding OEM skins (which would probably make Samsung walk) will still leave a plethora of hardware and the OEMs responsible for optimisation for that.
So I reckon that for "Android" to go the Apple way, which seems to be what the author is saying should happen, it really would have to go the Apple way: a closed OS (so Google can control who uses it and on what devices), very constrained hardware designs, possibly very centralised updates. Anything less than that won't be very different in practice. But what incentive would the OEMs have to become assemblers for Google's designs? So if Google went that way you would probably see Samsung switch fully to their own OS, and other manufacturers (or maybe a consortium of them - a real Open Handset Alliance) could take Android itself and continue from there. Leaving Google playing their own game by themselves.
So even if the base of the OS does shift (and become more closed) my feeling is that it will take longer than the author thinks, and won't make as much difference as he thinks either. But given that Android is successful now, and Google have an extensive track record for starting and dropping projects, I think he is reading far too much into Fuschia.