We always opt for wired where possible, even if wireless exists. We are paranoid about secure and safe cable runs. When wireless is necessary, we transmit on multiple frequencies with smart frequency-hopping algorithms so that if two of them get interference, the content still comes through on the third.
We have a visceral distrust of rechargeable batteries. We swap out AAs and 9-volts (only the brands we trust) obsessively. A good sound engineer is made very uncomfortable by (and thus obsessively monitors) battery-powered devices in the critical path.
In short: there's no way you're going to get respectable production companies to trust this thing. One, it flies, Two, it's battery. Three, it's RF. Under these circumstances, we proceed with EXTREME caution. We'd much rather build truss to get a camera where it needs to go, run cables to it, mat and tape them as appropriate. Automation is okay, but there is existing, battle-tested camera automation and it's not even that widespread yet.
Film production, maybe. A blown take isn't free, but it's pretty cheap. A flying camera at WWDC hitting battery starvation, motor failure, RF interference, etc. is the end of the world in that line of work.
And it would be a use case for the defunct(?) Microsoft Kinect 2.
Think TED Talks, Apple Keynotes, anything fairly big-budget where chasing the random motions of the speaker just doesn't look on par with the money going in & out of the presentation.