Slashdot Mirror
Towards Self-Replicating Rapid Prototypers
Neil Halelamien writes "Researchers at the University of Bath are developing a rapid prototyping machine capable of making copies of itself and other products, reminiscent of the Universal Constructor proposed by von Neumann. The so-called Replicating Rapid-Prototyper (or RepRap) would produce items from raw materials and small components like microchips. If successful, this could make rapid prototyping cheap enough for regular in-home usage, especially since the project's lead, Dr. Adrian Bowyer, will be releasing his project's designs under the GNU GPL. It's previously been proposed that a similar system would be useful for space exploration and industrialization."
Black holes mentioned earlier, now self-replicating robots. We're screwed.
Such a machine would have a number of interesting characteristics, such as being subject to Darwinian evolution, I have always been interested in applying evolution to computer chips - the randomness and efficiency of evolution is going to find better ways of doing things than our current methods, and is also just damn cool to know your computer chip is analogous to a living 'species'.
This will be GPL'd. I don't know how the copyright cartels would react if a machine could make illegal copies of itself.
What?
If the good doctor were to suddenly die in the next four years, I'd start lining my baseball caps.
Oh darn... the editors cut out my link to the Wikipedia article on von Neumann's Universal Constructor (i.e. clanking replicator). Here it is:
...
http://wikipedia.org/wiki/Universal_Constructor
From the article:
A clanking replicator is an artificial self-replicating system that relies on conventional large-scale technology and automation. The term evolved to distinguish such systems from the microscopic "assemblers" that nanotechnology may make possible.
Such a machine violates no physical laws, and we already possess the basic technologies necessary for some of the more detailed proposed designs.
A self-replicating machine would need to have the capacity to gather energy and raw materials, process the raw materials into finished components, and then assemble them into a copy of itself. It is unlikely that this would all be contained within a single monolithic structure, but would rather be a group of cooperating machines or an automated factory that is capable of manufacturing all of the machines that make it up. The factory could produce mining robots to collect raw materials, construction robots to put new machines together, and repair robots to maintain itself against wear and tear, all without human intervention or direction. The advantage of such a system lies in its ability to expand its own capacity rapidly and without additional human effort; in essence, the initial investment required to construct the first clanking replicator would have an arbitrarily large payoff with no additional cost.
On a completely different note, does anyone else remember the Slylandro probes from Star Control 2?
A rapid prototyping machine that can reproduce itself is a kind of holy grail. When we get that, we have something that can bootstrap itself. This would be the Santa Claus machine that we have long dreamed of. One of the major proponents of this has been Don Lancaster. His site is the link below.
www.tinaja.com/santa01.html
Once everyone has a machine in their basement, the economy of the world will be turned on its ear. Consumer goods will cost only the price of their materials. The cheap labor advantage of India and China will vanish. The nature of products will change. Right now, it makes no sense to make something repairable. It is cheaper to build something that can't be fixed and throw it away. When we get very distributed manufacturing however, things will be built with only one or two raw materials. Things will be built so they are easy to assemble. It would make sense to build a new heating element for your coffee pot. Waste would go down. Recycling would become much more immediate and local. People would share designs the way we now share open source software. Quite a different world would result.
Apparently there already is a self-replicating system out there; the system is capable of manufacturing virtually any kind of tool, machine or technology known. To make new copies of itself, it uses only common materials from the environment- water, oxygen, vegetable matter, protein, and that kind of thing. Unfortunately, top computer scientists and engineers are having trouble figuring out the self-replication process. Apparently, it involves some mysterious mechanism known as "sex", which takes place with a "woman". Anyone else know any more about this?
- Structure: The core parts of the device need a strong, stable material that can hold everything together.
- Motion: The device needs materials that convert energy into mechanical motion. These materials might include electromagnetics, electrostatics, piezoelectrics, shape-memory alloys, chemo-dynamic protein muscles, thermodynamic cycle systems, etc. Each of these types of motion-creating materials has special needs/chemicals that might require special handling devices that, in turn, must be made out of the materials in the self-replicating device. Motion is often tricky because it requires specialized assemblies of materials (think of the complexity of a simple DC electric motor or the gears and linkages in a robotic arm).
- Control: The device needs some form of logic that can read some analog of a blue-print, ROM, DNA, etc. and direct the fabrication process. If based on standard electronics, this would include materials that act as insulators, conductors, and semiconductors.
- Power: This may be the trickiest because creating sufficient power requires purified, highly engineered materials. Self-replicating a modern alkaline battery would be quite a feat. Perhaps the semiconductor technology of the control materials could be leveraged for solar panels.
I suspect that one of the trickiest part of all this is in handling and converting bulk materials (usually a liquid, powder, or solid ingot) into a shaped and controlled component or assembly. The replicator must interface with raw materials supplies, move bulk materials to a fabrication point, and convert the bulk material into a usable component in its offspring. Space exploring self-replicators face an even greater challenge of processing raw space materials (moon rock, asteroidal metals, etc.) into refined feed-stocks for replication.Its a tricky problem, but one that we will eventually solve.
Two wrongs don't make a right, but three lefts do.
While the idea of a 3D printer cheap enough for personal use /is/ going to revolutionize the world by making certain real items as cheap as software, the part about it being a von Neumann machine is overrated. The article just mentions it in passing and there is no evidence that he's actually figured out how to do that. That's been one of the holy grails of engineering since it was proposed. The article doesn't mention whether the materials used will be recyclable. Since everyone and their grandmother will start spitting out objects if they have this and since it would probably be cheaper to build a new object rather than repairing an old one, mass use of UCs will produce tons of waste. Imagine if you could never delete any file on your computer but could create more easily. You would run out of space very quickly.
BTW, for a good book on the social implications of cheap universal constructors, I suggest the Stephenson's book Diamond Age.
--
Want a free iPod?
Or try a free Nintendo DS, GC, PS2, Xbox. (you only need 4 referrals)
Wired article as proof
All this was done in a very crude way, as if they were developing a process for home use. Their metal casting technique is scary. They used "Wood's Metal", which is a solder-like alloy of tin, lead, cadmium, and bismuth. All of which are toxic. Lead and cadmium cause heavy-metal poisoning, and the body won't clear either of them. No serious precautions seem to have been taken against inhalation - they just used gloves. At one point they tried powdered metal, which is much more of an inhalation hazard than molten liquid. They need to run their people through the usual checks for heavy-metal poisoning.
There are rapid prototyping machines that deposit metal, and that's probably a more useful direction.
All this is a long, long way from self-replication.
I'm not sure what the big deal is about this particular rapid prototyping machine at Bath. Hod Lipson's lab at Cornell, for instance, has been able to create a solid freeform fabrication system which can print plastic, metal, circuits, actuators, and even batteries! They are, in my opinion, much further along than the referenced article. Other related projects of include Chrikjian's work at Johns Hopkins, and Jordan Pollack's DEMO Lab at Brandeis University.
Actually what you really want to do is to build a set of evolving self-assembling robots that get their parts by disassembling other robots. That way there is evolutionary pressure to evolve faster and faster self-assemblers.
Some drink at the fountain of knowledge. Others just gargle.