In the 1970s when gigantic in-space colonies similar to the one depicted in the recent movie Elysium were proposed, the obvious question was how could such behemoths possibly be affordable when it had cost billions just to send a handful of people to the Moon. That question still remains up front today for anyone proposing such free flying islands in space.
[]

Those who were working on such colony designs in the 70s were quite aware of that question and were counting on three factors to make the colonies affordable:

1. A big drop in space transport costs once the Space Shuttle started flying
2. Get most materials from the Moon whose low gravity allows for relatively easy launch into space, e.g. can literally throw material into orbit with a mass driver
3. Earn income from solar power stations built by colonists to supply power to earth.

Unfortunately, the Shuttle, which was far from the robust and fully reusable vehicle initially proposed, failed to lower the cost of getting to orbit by even a modest amount.

Without low cost launch, it has been impossible get back to the Moon in an affordable manner and to build up an infrastructure there that could one day support construction of habitats in orbit.

Space based Solar Power remains a popular cause among many space advocates but it also remains a distant dream as long as space transport costs are so high. (I also expect that it’s more likely that space settlers will earn their keep via other commercial services such as off-shored digital services (e.g. software development) and space tourism, which is what many island economies on earth rely on.)

Fortunately, launch costs may finally start to fall significantly as NewSpace companies like SpaceX begin to make progress towards fully and rapidly reusable launch vehicles. (E.g. see a SpaceX design here and a prototype here.)  The propellant cost for a typical expendable rocket are less than half a percent of the total cost of putting an object into orbit. The rest comes from throwing away the rocket each time. So cost reductions of a factor of 50 to 100 are not out of the question.

However, even such huge reductions in launch costs leave tremendous challenges to creating a construction infrastructure in space that can build something as big as a habitat to provide a comfortable environment for thousands of people. Another factor is essential to making large scale space construction feasible – advanced robotics.

It just isn’t practical due to radiation, logistics, and other issues to expect to see crews of humans assembling large structures in space. Most such work will have to be done by robots working autonomously as well as via  teleoperation by humans who could reside nearby in radiation protected habitats.

[]

Rapid progress is being made in robotics in general and specifically for space as well. See, for example, the satellite servicing demonstrations on the ISS of teleoperated robotic systems. And just last week, the company Tethers Unlimited won a NASA grant to continue development of the SpiderFab, a robotic system that also employs 3D printing to build kilometer size structures in space.

SpiderFab here assembles a large solar panel. (Tethers Unlimited)

(See as well this discussion of using a system like SpiderFab to build large space structures: Inflatable space stations, Reusable rockets and Spiderfab for the 2020s – NextBigFuture.com.)

It can be amazing to watch a construction project go from a muddy foundation to a brilliant tower reaching the sky. On a given day the progress looks so meager. Yet over a finite time we see that it really is possible to build a gigantic structure just by many small incremental steps by a lot of workers.

Similarly, it is in fact possible for hundreds or thousands of robots working 24/7 to build a really big structure in space in a finite time. Once such an infrastructure of robotic construction is established, we could see rapid development of the Moon and the whole cislunar system.