Sunday, July 12, 2009

Invisibility (how to do it and solving as many related problems as possible)

I've been thinking a great deal about how to accomplish true invisibility. I read a book when I was about twelve years old that clued me into the greatest problem with accomplishing invisibility - in order for you to be able to see, your eyes must be able to absorb and reflect light, hence the areas of your eyes involved in sight must remain visible. It turns out this isn't strictly true. My original idea - to render the whole body except the eyes invisible was kind of stupid - it's clearly better to use pinhole cameras as even floating corneas would be pretty conspicuous. As long as the type of invisibility we're talking about involves bending light around your person or sensing it and broadcasting it at the exact opposite side, you could have light inside (or at least enough to view a camera through some sort of electronic goggles) and it would be fine.

Angles provoke myriad problems. If we could count on observers only viewing us from one direction the problem would be simpler. Even paint works pretty well from ONE angle, as you can see (link):

Now, first we need to go through what I feel is the best way to accomplish invisibility - a suit or other exoskeleton-like mechanism that takes light pointed at the body and passes it through the exact opposite side with as close to the same hue and intensity as possible. One of the greatest problems we have here is that the reflection mechanism must be incredibly variable - it must be able to accommodate every angle, both vertical and horizontal, that the light source broadcasts, in every possible spectrum including the ultraviolet. I am not technically inclined enough to venture a workable solution here, other than to say that we're getting into pretty detailed nanotechnology here, and must build an incredibly receptive device capable of simultaneously sensing and broadcasting in multiple angles. Neither the broadcast nor the absorption qualities of the material must interfere with each other:

Here's a simple 2D representation of a three-dimensional shape capable of sensing and broadcasting in five directions. Obviously you'd need millions (or billions) of these across a single suit (though it wouldn't be "invisibility" by any stretch of the word - even if it worked to some extent it would only work from five angles). Now, moving to a bulbous design capable of hundreds of degrees of sensitivity seems like a good solution. The more sensitive the design (and the quicker the latency and faster the "framerate"), the less "shimmering" we see, like the Elites in Halo. Something that's "pretty" invisible would do OK against human reaction time but not much against devices programmed to notice the effect.

Still, a bulbous design doesn't solve the myriad problems that crop up:

- The gaps between each of our receptors/broadcasters would create shadows. This is such a complex, small design that I can't tell exactly what this would mean. Obviously the more invisible you are, the more the microscopic bulbs must protrude from your body in order to accommodate more angles, and the bigger the shadow problems created. Maybe making the edges completely transparent going in but not out would solve this problem, and maybe it wouldn't. It's also possible that allowing one receptor/broadcaster in this situation to pass light through to its immediate neighbors which would then broadcast to the other side of the suit would work.

- Invisibility would be near-useless without movement, so the suit must have a very keen idea of where every piece of it is at all times and be able to coordinate hundreds of angles from every single reception point which would require a horrendous amount of processing power.

We see here three angles that a single receptor must be capable of, and the coordination with other sensors required. The pink line, for example, must reflect to the upper thigh, the lower torso and the wrist. Not to do so would create shadows and dark spots. Now this is only three angles of a single receptor. These three angles have seven reflect points. Imagine the difficulty of coordinating only three angles with the entire body while it's constantly moving and I think you'll see the work ahead of us here. Millions (or billions) of receptors/broadcasters would each have to reflect thousands of angles, and most of them would have to communicate with at least two other points for broadcast purposes. And damned if you don't also have to be able to get into the suit, so any fasteners and overlaps would also have to be a part of the invisibility mechanism. And Heaven help you if you want to have access to weapons or other external sensing devices, though I suppose that once we've solved the problem of making a functional suit, integrating other devices into the invisibility mechanism would be comparatively simple.

I'm glad that I'm not actually working on this.

Imagine that we've solved all of these problems. Well, now we have some more:

The Thermal Spectrum - Something like a jet with fewer moving parts and a hard design doesn't have the movement problem discussed above (or at least has it in a smaller measure) but serves as a good demonstration of the problem of hiding yourself in the thermal spectrum, an issue that still exists for human bodies but is more workable there. With the advent of invisibility it is a simple affair to program a turret to shoot at anything that broadcasts in the thermal spectrum but not the visible one. So our device must be built out of a material that can reach and maintain absolute room temperature very quickly and thoroughly to almost the same sensitivity that it reflects light. And there'd still be some thermal "shimmering" due to latency and the fact that each node has to absorb and produce light, so your device wouldn't only have to maintain room temperature, but it would have to maintain the difference between room temperature and whatever heat is emitted by its operation. Which means (you guessed it) it would have to have some mechanism for both producing and absorbing ludicrously precise amounts of heat all over the unit.

Any heat produced inside (and there would be much, both from the device and other devices being used, as well as the human body) would need to be sinked off for wearer comfort into some sort of storage unit that could be cooled down whenever it's safe to do so.

And jets pretty much by definition have to expend heat behind them in order to frickin move. Good luck avoiding any missile guidance systems or sensing devices that rely on heat.

Sound - It isn't a stretch to imagine that sensitive areas, particularly governmental and corporate ones, would begin to incorporate sound surveillance into their surveillance rigmarole, particularly looking out for sound-producing bodies with no apparent visual source. We already have devices that can cancel sound to some extent by broadcasting an opposite but equal wave at the same time, nullifying the vibration. I can't think of anything else that would solve this problem.

Other Reflections - The presence of invisibility technology would essentially create a technology war between invisibility and sensing capabilities. I think that the first wave of invisibility devices would be primarily visual, then begin to incorporate some of the other capabilities discussed above. The final step is to determine every last blasted thing the human body can reflect or absorb and to do whatever you can to make the human body behave like empty air. Some of these would require modeling lag and latency to ludicrously precise degrees in order to represent the amount of time a wave takes to travel through the air but not more concentrated matter - any suspicious quickness would be noticed. I can't even imagine the myriad ways people could develop to find an invisible person or object, and we'd have to think of all of them.

Durability - Oh - and it would have to be strong material. Even if only a tiny portion of the suit were damaged it would essentially nullify our invisibility due to the angle problem discussed above. The device would have to be both incredibly sensitive and incredibly durable.

So it turns out that true "invisibility" is more than just not being seen. It's an incredibly complex problem with hundreds of applications that requires an infinite level of adaptivity across a single surface. We need something that can measure its precise location relative to itself at all times while moving, absorb and emit light in all spectrums and precise amounts of heat, nullify sound and, hopefully, still be fairly light and mobile. Do I have any gargantuan flaws in my reasoning or is this really as hellish a problem as I'm imagining? I'm one to think that being able to discuss how something could work theoretically is a bigger leap than actually working it out, so I'm more alarmed by the fact that we can even discuss the issue in these terms. Also note that I stupidly developed this blueprint without reading anything on current invisibility applications, so much better models probably can and do exist.


  1. Heh. You're a bit of a polymath yourself, thinking about things like this.

    "Never gonna happen" is a risky phrase to use, if you're a widely-read pundit, but fortunately I'm not. Let me qualify:

    I don't think invisibility will ever happen by a "brute computational force" method such as you describe here. That is to say, the hypothetical Romulan cloaking device doesn't work by sensing and re-broadcasting light at all the right angles and frequencies.

    Rather, the particles comprising the object becoming invisible will simply sidestep into a kind of subspace or "hidden dimension" (as string theorists would say).

    Or perhaps, the fundamental electron shell/photon interaction of the object's molecules would be "tweaked off" in such a way that photons simply fly right through the object with no effect, the same way neutrinos fly through us all the time.

    Of course, those methods depend on physics we don't know where yours is confined to what we do know.

  2. I agree with you that real invisibility (practical undetectability) probably relies on some chance discovery rather than a brute force method like the one above. Though I admit that since it's unlikely I'll be involved in the actual discovery of anything world-changing, much of the fun of this line of thinking comes from thinking it through thoroughly rather than practical application.

    Incidentally, I posted this on the XKCD forums where a user suggested a material that when light passes through it one way is turned into radio waves, then converted back into light on the other side. The angle problem is the greatest one, so I imagine that the breakthrough will come from figuring out a method that does most of the work on its own.


Thanks for commenting - I reserve the right only to delete ads, nonsensical spam or comments indistinguishable from such.

Note: Only a member of this blog may post a comment.