I guess many of the advances to make mobile DVB happen will be in the receivers, making them more powerful and more selective at the electronic level, given that other factors are reasonably static (if the fat thingy in No 10 sells off the bandwidth when analogue goes, will there be much scope for boosting transmitter power?)
The problem is, DVB-T was designed for a large number of channels at a relatively high bit rate (so the picture doesn't look offensive on a ruddy great TV) but with the data not coded very robustly due to the relatively static nature of most receiving stations and the ability to put up high gain antennae.
The demands of a mobile setup are for a low bit rate (screen is going to be small - so no point in high quality), more robustly coded signal so better error correction and equalisation can be applied by the receiver and it can cope with the multipath you can get on the move.
Whereas with FM radio later innovations like RDS allowed it to be useable in a mobile environment the requirements are so conflicting with DVB-T it seems less likely. However, when FM Stereo was first launched you'd probably have been branded a madman for suggesting listening to stereo radio broadcasts in a car. Where would you put all the valves, for a start?
So who knows?
There are also technologies like HSDPA and LTE which, if they got cheap enough, might allow mobile streaming of video over IP to mobile users.
Kevin