NASA Spaceflight has an interesting report on the status of the study.

It sounds about right to me. Retire Atlantis and make it a parts queen or a launch-on-need vehicle, and fly the other two vehicles once each per year. But at that low a flight rate, I wonder if the processing teams lose their “edge” and start to screw up? There’s an optimal flight rate for both cost and safety. Too fast and you make mistakes because of the rush, but too slow, and you get out of practice. And of course each flight would cost over two billion bucks, assuming that it costs four billion a year to keep the program going.

And as noted numerous times in the past, this doesn’t solve the problem of leaving US crew on the station. They still need a lifeboat of some sort. They discuss this as a “COTS-D Minus”:

…several companies have noted the ability to make available a lifeboat vehicle from 2012 (names and details currently embargoed due to ongoing discussions).

Clearly, one of those companies has to be SpaceX.

But this idea seems to never die:

‘There is some interest now in developing this (RCO) into a full mission capability, thus enabling unmanned shuttles to launch, dock to ISS, undock and land in 2011 and beyond.’

‘While that’s an interesting idea and would be a fun development project, we are working to understand the level of effort the program desires for this study.’

It’s not an “interesting idea.” It’s a monumentally dumb idea. There is little point in flying Shuttle without crew. The ability to fly crew is its primary feature. It’s far too expensive to operate to act as a cargo vehicle. If the point of the idea is to not risk crew, then we have no business in space.