Throw out your speculations, predictions, guesses, and expectations.
To keep it focused, wishful thinking and way-out-there ideas should be kept to a minimum.
I'll start as the devil's advocate:
[I'm not expecting much, TBH. I am not convinced thus far that PGI is up to it in terms of game design.]
There will be a central base of some sort which both teams try to conquer somehow. I guess by getting into it and standing in a square. While your team holds the cap, the enemy bar ticks down. The base will be a relatively boring smallish structure with some automated defense turrets and maybe barricades that need to be removed. Around the map there are control points that can be captured by standing in a square. Holding a control point magically has some effect on the base, like disabling a certain turret or opening a certain door. It won't take long for a certain small set of effective tactics to appear and the whole exercise will turn out not much more interesting than the current pick-a-path and wack-a-mech.
Looking forward to being proven wrong!

PS. Of course, "Conquest" will prove to be just as much an over-the-top misnomer as "Assault".