Maglev Catapult Launch Systems

Actually, I think something like a plasma shield is possible, although not with plasma, but with air. If you could have some kind of a fan blowing wind in front of the vehicle, you might be able to decrease the air pressure enough to do some good (faster moving air, less pressure).
Also, the idea of throwing out air in front of a rocket isn't new - It's used in rocket torpedoes, which are much faster than regular torpedoes.

http://en.wikipedia.org/wiki/Shkval_torpedo

Air isn't fast or thin enough. Plus you would simply move the high pressure region (and drag) in front of your low pressure cushion, and still have to shove the whole thing along. In atmosphere, the equivalent to those "supercavitation torpedoes" would ideally be a complete vacuum, with a plasma the next best thing.

The idea is to use the evaporated plasma from the leading edge to insulate the rest of the craft, it dies so the rest is reusable...

Ok newest understanding of this so far,

Stage 1 Maglev launch sled on open air track, accelerates to well past the speed of sound, hydrogen scramjet engages,

Stage 2 Front heat shield begins to evaporate into plasma, behind nose large electromagnet fires up, this repells the plasma just off the surface of the craft enough to keep the incoming pressure from forcing it into a point further down the ship, you could have a chain of them, to keep the supercavity in place. Front heat shield is designed to he replaced.

Stage 3 Acccelerating using the plasma envelope as a reaction mass, the ship reaches over escape velocity and powers out, using a unknown power supply,

M.h.d generators would make a great stearing and breaking system for the controls in the supercavity. As no air means control surfaces a moot point.

Somebody probably mentioned this already but you could try and design a small enough version of your plasma supercavity generator that could fit inside a bullet shot by those hydrogen gas guns that can achieve high mach numbers. Then again you also could try and put together a cubesat sized demo that was meant to deorbit to test the concept. These potentially wouldnt be too expensive especially if quicklaunch gets its project off the ground. With their orbital cannon you could do tests both ways, into orbit and while deorbiting. Who knows this techknology might even make their bullets more efficient and enable them to deliver more mass to orbit for a given amount of energy input.