The measure of aerial over watch is probably only viable in more open ground with nowhere to hide, A heavily defended modern city would post a nightmare for AFVs and infantry alike, seems like the only tried and true way of armored push into cities is by reducing it to rubble first through artillery fire/air power. APS would allow for a lot more survivability in instances where a city would need to be minimally damaged.
In fact an effective and well stocked APS would be crucial on landing AFVs if China want to land on a contested beach without wasting time for defenses to be sufficiently degraded through long range fire or suffer heavy casualties.
APS in urban environments would have its effectiveness compromised and reaction time reduced, because the sensors needed to cue the APS itself will be in a high clutter and overlapping background environment.
In an amphibious assault against a beach, whether you had APS or not, you still need to use substantial long range fire and ISR/recce with on call strikes to deal with enemy defenses and counter attacks.
And don't forget the money and maturity factor again.
I understand there is a certain attractiveness to the idea of greatly reducing the effectiveness of ATGMs through APS, but it just isn't that simple.
Perhaps one day in the medium term future for the PLA, APS will be sufficiently mature and affordable to be employed among some primary AFVs. But that isn't happening any time soon.
So until then, the PLA will have to deal with ATGMs through combined arms warfare, superior recce and fires, and the ability to absorb losses while still achieving their objectives.
And even once the PLA does have some type of proliferated APS, it would still make sense to rely on combined arms warfare, superior recce and superior fires to degrade enemy ATGM capabilities and ATGM defensive options -- in such a situation, the APS would be a final last ditch defense rather than the primary means of defending against ATGM teams.