The problem with hydrogen is one of efficiency.
Present commercial hydrogen production is from natural gas. However, significant energy losses occur in the process such that it would be more efficient (in energy terms) and cheaper to simply run the vehicle on natural gas.
The alternative is to produce hydrogen from electricity. Again, efficiency is the problem. About 65% of world electrcity production (and about 90% of Australia's electricity) is derived from fossil fuels - coal, gas and oil in that order of importance. The remainder is mostly from hydro and nuclear in that order. Virtually all of the non-fossil fuel electricity in Australia is from hydro, and 60% of that is in Tasmania (30% from the Snowy, rest mostly in Vic, NSW and Qld with a bit in WA).
From an efficiency perspective, there is already nearly a 70% loss of energy in fuel by the time it reaches your home as electricity. Most of that loss occurs at the power station (contrary to popular belief, transmission losses aren't really that high although they are significant). So, you only have 30% of the energy left and then you lose about 30% of the remainder (leaving 21%) converting that to hydrogen. Then you lose a bit of the hydrogen itself to leaks and other losses (leaving 20%). Then you lose at least another 40% of what remains, leaving 12%, with a fuel cell. Then you lose another 10% or so converting that electricity into mechanical power. So an overall efficiency of about 11% from fuel in the ground to power at the wheels.
This is about half the efficiency of a petrol engine so switching to hydrogen from fossil fuel electricity would simply double the quantity of fossil fuels burned. Not a worthwhile objective. Even getting hydrogen from natural gas still isn't efficient. And the efficiency of a fuel cell is even worse when compared to a hybrid petrol (or better still diesel) engine.
So it only makes sense to switch to hydrogen if we have first replaced fossil fuel electricity with something else (renewable, nuclear). It wouldn't make sense to build renewable power generation just to make hydrogen whilst still using fossil fuels for other electricity. It would make more sense to simply use the renewable electricity to reduce the use of fossil fuels for electricity rather than producing hydrogen. Hydrogen is thus a technology for
after we already have a predominantly non-fossil fuel power supply from which to produce it.
Whilst there's an absolutely critical need to move away from oil, efficiency wise it is better to make liquid fuels from gas or coal than to generate electricity with them and then use that power to produce hydrogen. That doesn't mean hydrogen won't happen commercially, but it's nowhere near the most efficiency use of resources. A petrol powered fuel cell would make more sense if the ultimate source of energy is going to be coal, oil or gas (which it will be for quite some time yet). There is some research in this area.
That doesn't mean that hydrogen is a dud though. If you already have a renewable source for electricity then it makes a lot of sense. And of course now is the time for research even if commercial use is still some time off.
To this end Hydro Tasmania / UTAS (University of Tas) have a hydrogen powered Toyota Corolla up and running. It's in this year's Targa Tasmania (a 2000km road rally in Tas) which is on now. The car is entered in the touring section (not the racing section) however. Targa Tasmania ends this Sunday in Hobart. This isn't Hydro's first venture into non-oil powered transport, but it's certainly a big step up from their previous battery powered cars, hydrogen scooters and postie bikes.
Next step is to start a small scale hydrogen vehicle roll-out, set up filling stations etc. The longer term plan is for a massive upscaling of wind power generation to feed a large hydrogen production industry. Present idea is to produce the hydrogen on-site at service stations. On a large scale this would involve a reasonably sophisticated linking of hydrogen production with wind generation output and having adequate on-site storage. The latter shouldn't be too much of a problem - service stations already store petrol rather than making it constantly and this isn't really that different.
The actual conversion of the Toyota Corolla uses the standard petrol engine and is basically a hydrogen conversion "kit" that, with some modification, could be fitted to any car just as LPG is done now. And of course the engine can still run on petrol if need be. In fact it can run on hydrogen and petrol at the same time (or can run solely on hydrogen or solely on petrol). Whilst the car also has various other non-standard systems (notably the unique traction control system) they aren't directly related to the hydrogen fuel system and aren't necessary for its operation.
From a cost and efficiency perspective, at present the advantage lies with hydrogen-powered hybrid internal combustion engines rather than fuel cells. The latter are still too expensive, too unreliable and aren't in practice more efficient than the hybrid.
So why the conversion of a simple petrol engine rather than a hybrid? Well, the batteries in the hybrid would cause a bit of suspicion as to whether the hydrogen really works... Hydro / UTAS don't want to be accused of building fake "perpetual motion machies" etc. And of course most of the cars that would be converted in any large scale roll-out will be conventional petrol powered engines.
More info on the Hydro Tasmania / UTAS hydrogen car in Targa is here.
http://www.utas.edu.au/prue/Media Releases/2006/0407hydrogen_targa_car.pdf