4

This video, titled "The Missile Knows Where It Is", has become popular as an internet meme. It describes a technique for missile guidance, but it's so confusing that people have started using it as a joke. Since its resurgence, it has amassed many millions of views.

Know Your Meme claims that it originates from a "1997 Air Force training video". It cites a defunct University of Wyoming page that references an issue of the newsletter of the "Association of Air Force Missileers". That newsletter contains a piece of text, titled as "GLCM GUIDANCE SYSTEM", and prefaces it with this information:

Submitted by Colonel (Ret) George Grill, who was with General Dynamics at Greenham Common AB, England - it may not be the first time you have seen this - is seems to apply to all guidance systems.

The missile knows where it is at all times. It knows this because it knows where it isn’t. [...]

The text there deviates somewhat from the video, and it ends with "Simple.", which seems to imply to me that it's a joke.

However, some people claim that it's a correct explanation of feedback loops inside of inertial guidance systems, and that it really was used for training inside of the US Air Force.

I know this is from a 90's Air Force video.. but it sounds like he's reading the script straight off a corporate patent application.

Like someone trying to explain a Kalman filter with zero math.

I'm interested in two things:

  • Was this actually used for training in any military context?
  • Is this vague explanation accurate in a general sense?

Please note that I am not asking whether this is a good explanation, or whether there's a better explanation elsewhere.

8
  • 1
    You say "However, some people claim that it's a correct explanation of feedback loops inside of inertial guidance systems, and that it really was used for training inside of the US Air Force." Please give some examples of that.
    – Oddthinking
    Feb 22 at 10:55
  • 1
    To me this is clearly a spoof. You think it is a joke as well. I am doubting that this is widely believed to be real. As to whether it is an accurate description: how can you determine whether something unintelligible is accurate? It is the Nostradamus problem up close.
    – Oddthinking
    Feb 22 at 11:17
  • I suppose a better Q would be if the audio is genuinely from AF training video. Feb 22 at 13:05
  • 5
    It can be a joke and also accurate. The best technical jokes are.
    – Schwern
    Feb 23 at 0:27
  • FWTW, I've not found a certified vid like that for cruise missiles, but there's an older vid for ballistic ones youtube.com/watch?v=y3lLVet9By4 It actually ends with "so we know exactly where the missile is at all times." So the style at least, even if parodic, is somewhat true to official vids. Feb 23 at 14:21

2 Answers 2

4

Is this a joke?

Most assuredly yes.

Is this accurate?

Yes, just deliberately convoluted and tongue in cheek.

The BGM-109G Ground Launched Cruise Missile (GLCM) uses a combination of an inertial guidance system and terrain contour mapping (TERCOM).

Inertial guidance has been around since the days of sailing ships. It uses gyroscopes and accelerometers to calculate how far it's moved from its origin point. For example, after launch (initial velocity is 0) the missile has been accelerating at 100 m/s^2 (about 10g) for 10 seconds due north, the missile "knows" that it is 5000 m north of its launch point (acceleration * time^2 + initial velocity * time).

Inertial guidance is not 100% accurate, and each inaccurate reading adds "drift". For example, if the accelerometer says 100 m/s^2 for 10 seconds (5000 m) but the missile is actually doing 110 m/s^2 for 10 seconds (5500 m) then there's a "difference or deviation" of 500 m.

TERCOM helps by providing altitude maps of the route. The missile is constantly taking altitude measurements. The missile can check its position via inertial guidance against its TERCOM maps, but because of drift it could be anywhere within 500 m of that point. It can then use its altitude measurements to find a line of elevation on the map within 500m of its position which matches its measurements. Then it knows where it actually is and can adjust its heading to get back on course. For example, if its altimeter measures that it's been flying up a slope, but inertial guidance says its flying over flat terrain, it can look on its maps for a nearby matching slope.

To translate:

The missile knows where it is at all times. It knows this because it knows where it isn’t [it is following inertial navigation, but it knows inertial navigation is inaccurate]. By subtracting where it is [via TERCOM] from where it isn’t [inertial guidance], or where it isn’t from where it is (whichever is greater) [not sure what that means], it attains a difference or deviation [inertial guidance drift]. The guidance system uses deviations to generate corrective commands to drive the missile from a position where it is [where TERCOM confirms it is] to a position where it isn’t [the correct course it has drifted from], arriving at a position where it wasn’t, but is now [the missile moved from being off course to being on course].

Was it ever used as training material?

Probably only as a joke to lighten the mood.

1
  • It's one possible interpretation, but I doubt the original quote was intended to explain how multiple positioning systems interracted (as you have it), as opposed to a more generic description. The newsletter does preface that with "it seems to apply to all guidance systems". Feb 23 at 14:12
0

The text there deviates somewhat from the video, and it ends with "Simple.", which seems to imply to me that it's a joke.

That's how guidance, navigation, and control (GNC) experts talk about how "simple" their job to outsiders is while at the same time poking fun at outsiders (and it is anything but simple; ten years is not enough time to develop expertise in this field).

Is this vague explanation accurate in a general sense?

In a ridiculously general sense, yes, it's kinda-sorta accurate.

Please note that I am not asking whether this is a good explanation, or whether there's a better explanation elsewhere.

A good place to start is Fundamentals of Astrodynamics and Applications by David Vallado, but that will set you back a couple of hundred dollars. A much cheaper option is Fundamentals of Astrodynamics by Bate, Mueller, and White. (This link adds William Saylor as a fourth author, but this is widely known (in the field) as Bate, Miller, and White). Vallado is a graduate level text while Bate, Miller, and White is an undergraduate level text. Both are starter level (for graduate students in the case of Vallado, for undergraduate students in the case of Bate, Miller, and White).

Here's my explanation:

  • Guidance, via a pre-planned trajectory, knows where the vehicle is supposed to go.
  • Navigation, via various navigation sensors and external inputs, knows where it thinks the vehicle is right now. This might or might not be correct. Without external measurements, it almost certainly is incorrect. Sailors of old tossed a piece of wood overboard and used that information to attempt to determine how far they had sailed, and in which direction. This is called dead reckoning. As in, rely too much on dead reckoning and your vehicle will soon be dead. There are some places where dead reckoning by a missile or spacecraft can be augmented by external inputs; GPS is a very good start. However, GPS doesn't quite work beyond Earth geosynchronous orbit. Missiles don't go that high; they can use GPS. (GPS was invented to make missile more accurate; the initial name was NavStar and it was highly classified.) Vehicles that go outside the Earth-Moon system oftentimes do not know where they are.
  • Guidance uses the navigated state to determine where the vehicle should be (now) versus where navigation thinks the vehicle is (now) to determine what to do in order to correct that error. This is not a Kalman filter; it's navigation that uses Kalman filters. Guidance does not. It instead uses a variety of different algorithms that depend on phase of flight (which algorithm is used by guidance strongly depend on phase of flight). Guidance uses these guidance-specific algorithms to eventually get the vehicle back on track by sending commands to control.
  • Control tells various effectors (thrusters, aerodynamic surfaces, etc.) what to do in response to commands from guidance. This varies very much from vehicle to vehicle. Some vehicles have adjustable aerodynamic surfaces (e.g., wing flaps, ailerons); others do not. Some vehicles have orientable thrusters (the relevant term is thrust vector control); others do not (in which case control picks the combination that comes closest to firing in the direction commanded by guidance). Some vehicles have throttleable thrusters; others do not (in which case the thrusters are either on or off, no in between).

Navigation algorithms, guidance algorithms, and control algorithms are quite distinct beasts. Navigation algorithms typically are centered on Kalman filters. Guidance and control algorithms are anything but.

This site requires references, and I truly wish I had better references. The key issue here is that guidance, navigation, and control are highly protected concepts.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .