To produce a Ph.D. astronaut takes about 28 years plus ten more years to become experienced in the trade. Even though an astronaut is necessarily a bit of a jack of all trades, their knowledge is still very limited. Transported to space this human being needs 7 pounds of oxygen, food, and water per day, plus a whole community of support services. The investment needed to keep a single human being alive and working in space is tremendous. A space colony would have thousands of humans. Over coming that huge cost in developing the basic infrastructure of life in space is the major restriction on extending human life into the solar system. What is really needed is a seed, a small cheap easily transported self contained robotic system that will build all not only the human required infrastructure, but even the infrastructure needed to build the equipment that will build the space colony. It will be the equivalent of packing the 500 year history of the industrial revolution and all of our scientific knowledge into a one meter 1000Kg cube that is expected to not only fend for itself on a hostile planetary surface, but build the means to reproduce itself and then build a home for man in space. Such a device would be the ultimate human achievement. The smaller it is the cheaper it will be to transport it to it's new home. We're looking for that minimum of basic tools, a set of fingers and two opposable thumbs. It would mean a trillion to one reduction in the cost of getting a human civilization into space.
Let's use the Moon as an example. It's not too hard to get to and only 1 1/2 seconds away by radio. Our little robot can always ask questions if there's a problem, and the main brains of the operation can be on earth. The first couple of attempts at self reproducing robots is sure to have problems. The first step is to survey the Moon for resources. The Clemintine probe has already done this, so yes we can already build that technology. Once our robot has landed on the chosen site it must prospect for materials. Then mine, refine, smelt, and machine those materials. Then produce a small factory capable of building a duplicate of itself. The robot does not have to reinvent the entire industrial revolution, that knowledge will be stored in it and it can take advantage of high technology solutions, but it will have to build a complete factory system, capable of building precision metal parts, electrical equipment, solar cells, optical equipment, and the latest in computer integrated circuits, all from raw material. It will have to build furnaces, forges, machine tools, everything that is needed to build a miniature factory that will build a robot exactly like itself. The time frame from there being one robot to there being two robots is critical as a fatal error will result in the whole process coming to a halt. Once there are a large number of robots the odds of failure are greatly reduced, any one robot could start the process all over again. When there are many then they can build a much larger industrial base and the much larger robots that will build the infrastructure needed to support human life.
To review, the steps to civilization are; Explorer, Prospector, Miner, Refiner, Forger, Machinist, Engineer, Cyberneticist, Farmer.
A human being has been called a self replicating learning organism, the vision of John Von Neummann was to build a machine that would do the same thing. One big step in breaking down the barrier to space will be to develop Von Neummann technology and to employ it to build space colonies and colonies on the Moon.