> In addition, you have to jump through a lot more hoops with the FAA getting an automated
> vehicle approved for flight than a manned one.

This is true, and for the Armadillo vehicles, it makes for a fairly ridiculous situation:

Since we are completely fly by wire, the computer doesn't care if it is getting its control input from a joystick plugged into the electronics box, or over telemetry from a laptop. Prudence would dictate testing each new change remotely many times before putting someone in it. However, the regulations for unmanned rockets (they don't discuss guidance systems) specifies a 15 second burn limit before you need a launch license. That isn't long enough to do a hop to any kind of altitude followed by a powered landing. If we had it classed as an experimental aircraft, it could fly as long as we wanted, but would require a pilot on board for every flight.

The FAA won't let us do things the "safe" way with standard regulations.

It does look like we are going to be able to avoid the issues by doing the development testing in some specially delimited areas in Oklahoma. I'll worry about deployment issues another year...

John Carmack