Upping the volts will make hybrid cars much cheaper

Power delivered by the battery is this equation:

P = I * V

So you can have (1 volt @ 100 amps), or (100 V @ 1 amp) and still get (100 watts). If both are the same, wouldn't the low voltage solution be safer, and more desirable?

Massaging the same equation a little bit, we can model our losses along the power lines. Substituting V = IR you get...

P = (I^2)R

It's difficult to make conductors less resistive without using expensive metals or unconventional techniques (like cryogenically cooled wires). So to cut our losses, we'll go with high voltage, low current.

The ICE is connected to the drive shaft via the 'hybrid synergy drive' which is a gear box and electric drive in one. It can drive the vehicle either under electric mode with the ICE off, with both electric and ICE, or in electric mode while the ICE charges the battery.

When you brake it cuts power to the ICE (either just cutting the fuel, or stopping it completely) and charges the battery. There are also normal brake disks if you push the peddle extra hard.

I like it because it's relatively mechanically simple, which means less things can break. The ECU does a lot of work, but try and find a car that doesn't nowadays. Also it only needs electric motors in one place - the rest of the axle and drive system is exactly the same as a normal car.

The only thing really letting it down is the low capacity battery pack, but I guess that's for cost savings. There is also a plug-in version which can do 26 miles on electricity alone. Toyota offer this as an option in most of their cars, even the Rav4 SUV, so I expect in the next 10 years it will become standard across their range.

The reason you don't see this kind of system everywhere is most likely cost; these systems cost extra. I have the Mazda system on my car; there's a pricey supercapacitor for storing power, plus a pricey DC-to-DC converter under the driver's seat, plus a special high-voltage alternator, plus a special battery. It improves fuel economy by 1-2mpg, which is the other reason you probably don't see them that much: an extra $500 or more in the new price of a car is a lot for an extra 1-2mpg when gas is $2/gallon and everyone's buying SUVs.

Such systems could be done better, but that's likely going to cost even more; to get a more significant improvement you really need to go full-hybrid with a large traction motor, which of course is going to make the vehicle cost even more. Take a look at the Chevy Volt for a good example of what you can achieve here: it's a serial hybrid with a small gas engine with generator plus a traction motor that actually moves the car. It can go 50+ miles without using any gas, which is pretty cool, but when the gas engine comes on (like for long trips) it can only get fuel economy in the low 40s because of the lossiness in the system. On top of that, having all that hardware in the car (engine, generator, motor, battery pack) limits the useful space, and the cost of the vehicle is rather high. Or you can get a Toyota Prius which is a parallel hybrid (engine drives wheels) and gets much better highway economy, but it can't go very far without burning gas and it's pretty expensive for an economy car with very anemic performance.

I've long wondered about that given so many trains and ships run that way using electrical transmission from the prime mover ("that's what "diesel-electric" is), I know little about the concerns so I must be missing some significant drawback on the scheme given so few hybrids are "series" and most manufacturers go for either parallel or split topologies.

Of note, an other potential advantage of a series hybrid is you can run alternative ICE topologies e.g. with high but very short efficiency bands.

Now, when you add the cost of the ABS/TC/DSC system module, maybe it works out better, I dunno, but the amount of copper you'd need in 4 separate electric brake motors would be pretty high, so I don't see how they'd be cheap.

The real advantages for a 48V system are weight and size of existing electric systems. Modern cars have a lot of motors, including now the power steering system (run by a motor attached to the steering shaft). These can all be downsized with higher voltage, saving copper ($$$) and weight. There's also smaller motors in the car, like those in the seats, and in luxury cars there's all kinds of motors (for opening and closing doors and liftgates, adjusting the steering, etc.). Finally, the wiring weight can be reduced all over the car, as well as the wiring cost (since again, copper is very expensive). Your statement about shielding is BS: yes, higher-voltage wire needs thicker insulation, but not that much going from 12V to only 48V, and insulation jackets weigh very little compared to the copper inside them. Furthermore, with a reduced diameter of the copper conductor, the diameter of the insulation will also decrease, so the insulation weight will probably be the same, if not a bit less. Besides, I'm pretty sure most wiring used on cars these days can handle hundreds of volts, and is sized not for electrical insulation properties, but instead mechanical properties (they make it thicker and tougher to handle being handled during assembly and repair, and also so it doesn't get worn through from vibration over the car's lifetime). If they actually sized wire insulation for only 12V, the wires in cars would barely have any insulation on them at all.

Safety is of course a huge component. 48V is about the limit before special isolation and insulation is required.

There actually is an existing sweet spot for 60V FETs, optimized for traditional 48v power systems. I've got to wonder if they're going to use -48v to allow simple driving of N FETs, or if the ubiquity of modern high-side drivers has made that a moot point.

Yes, actually they do. Power MOSFETs are usually in the 30-60V range. When you go higher than that, costs go way up and you generally need to switch to IGBTs. MOSFETs capable of switching 30V loads at extremely high loads (like 100A) are dirt cheap. Power devices capable of switching 100V are not.