Slashdot Mirror
Smartphone-Enabled Replicators Are 3-5 Years Away, Caltech Professor Says
merbs writes: In just a few years, we could see the mass proliferation of DIY, smartphone-enabled replicators. At least, Caltech electrical engineering professor Ali Hajimiri and his team of researchers thinks so. They've developed a very tiny, very powerful 3D imager that can easily fit in a mobile device, successfully tested its prowess, and published the high-res results (PDF) in the journal Optics. Hajimiri claims the imager may soon allow consumers to snap a photo of just about anything, and then, with a good enough 3D printer, use it to create a real-life replica "accurate to within microns of the original object."
Bad headline: "Smartphone-Enabled Replicators Are 3-5 Years Away, Caltech Professor Says"
Good headline: "Smartphone-Enabled 3-D Scanners Are 3-5 Years Away, Caltech Professor Says"
I *can* imagine the possibility that within 5 years, we'd have portable enough 3d imagers and powerful enough phones to both stick the hardware in a phone-sized device and have a phone-sized device run the required software. I have no real understanding of how the physics of that would really work, granted, but it doesn't seem totally outside the bounds of possibility.
But that's just the input. I *can't* really imagine the possibility that within 5 years, we'd have powerful enough *printers* to take the output of such a precise scanner, and recreate it anywhere near so precise, even if you're only talking about an object made entirely out of one or two kinds of plastic, which is unlikely to be the sort of object people would really want to "replicate". "Just about anything"? Yeah right.
Wake me up when we can replicate food, say, and have it taste the same as the original. Will we see that in my lifetime? Maybe, if I'm lucky. Will we see that in 3-5 years? I'd bet quite a lot of money on "no", and I'm not a betting sort of person.
Great, you can scan something and then print it in crappy plastic. Big whoop.
Seriously, 3D printing has been around for a while now, and I am still waiting to see anything beyond the Gee-Whiz level of cool or useful. You can only make so many money clips, pencil holders, and miniature busts before it becomes clear it is just a toy. Industrial ones that can print in metal are a different story, but the crappy plastic extruders are never going to take over the world or replace China's factories.
Perhaps it only peripherally relates, but... I'm interested in the team that recycles all these damn plastic wrappers, bags, bottles and endless polymer stuff into usable 3D printer raw material. Now let's se THAT breakthrough!
Please have respect for people with different abilities, especially children.
that the "accurate within microns" part is only applicable if you feed it some scaling information.
Otherwise, it's going to only be as accurate as the person guessing the size of the original.
AFAICT, the technique used by this imager is FMCW (frequency modulated continuous wave) which basically give you a very accurate time-of-flight measurement. In this type of system, the received optical frequency difference from the transmitted frequency is measured by optical-coherent mixing and sensing the resultant beat signal frequency. Apparently this groups contribution to this technique is to measure both the phase and amplitude of this beat signal digitally so multiple algorithms can be deployed to analyze the beat signal.
In any case, given a very accurate distance to an object, the solid angle projection to the imaged object, and some basic optical system calibration data, it is presumably fairly straightforward compute the actual size of the object w/o guessing.
How often do you indulge your inner need to tell the world just how forward thinking you believe you are?