Recently I have read several different articles about developing an orbital transfer vehicle.  The thing is the government doesn't really want to spend the money on it.  However when the last Milstar satellite got stranded in an unusable orbit, funding came about.  Without this satellite or its proposed replacement there will be a gap in air force communications in a few years.  So, instead of launching a very expensive satellite on a very expensive launcher, why not develop the orbital transfer vehicle and make salvaging the Milstar satellite its first mission.  This way instead of making up a new program to fund it, it could just use the money that is going to be spent on the replacement Milstar satellite.  Personally I would be in favor of using some sort of ion propulsion, as seen on Deep Space 1.  However if it proves too tough to adapt this to the mission that I am proposing then it would also work to make something more conventional.  It could be as simple as a centaur stage modified with some kind of mechanism to grab the satellite.  This could also be launched on a smaller rocket because it would only have to launch the upper stage, not an actual satellite.  If we went the route of a centaur then it would probably be an Atlas or Delta 3.  I don’t have any clue what it would be if the ion propulsion system was chosen, it could probably fit on a Delta 2.  However if the plan is to use it more than once it should be supplied with enough fuel for more than one trip to orbit.  This would necessitate a larger launch vehicle.  Exactly how big would depend on how much one would want to reuse it.  A distinct possibility would be to wait for the EELV to come out and use that, remember it wouldn’t have to be launched for a couple years.  As far as costs, the Titan 4, which would probably launch the replacement satellite, costs about $300 million, and the satellite itself costs around $800 Million ( FAS http://www.fas.org/spp/military/program/com/milstar1.htm#N_1_ ).  I’m assuming that another satellite would be a few million cheaper because the design and development is already paid for.  However that still leaves plenty of money to finance this operation.  Deep Space 1, which included the ion propulsion engine, was designed, tested, built, launched, and operated for a couple years for about $150 million.  If it is decided to take the ion propulsion route then the same type of engine could be used.  There would be some small modifications, such as much bigger fuel tanks, and a different bus that it would attach to.  So if you take Deep Space 1, spend about $90 million modifying its fuel tanks, $200 million for some kind of a mating adapter, and $300 million for a launch on the Titan 4 you would still save over half a billion dollars, and those prices are all overestimated.  So, my long-winded point is: since there is a demand for the orbital transfer vehicle and there is also a way to easily pay for it why not go ahead and develop it.