1356: Orbital Mechanics

Explain xkcd: It's 'cause you're dumb.
(Redirected from 1356)
Jump to: navigation, search
Orbital Mechanics
To be fair, my job at NASA was working on robots and didn't actually involve any orbital mechanics. The small positive slope over that period is because it turns out that if you hang around at NASA, you get in a lot of conversations about space.
Title text: To be fair, my job at NASA was working on robots and didn't actually involve any orbital mechanics. The small positive slope over that period is because it turns out that if you hang around at NASA, you get in a lot of conversations about space.


Randall roughly plots how high school physics, undergraduate-level physics and a job at NASA somewhat increased his knowledge of orbital mechanics. But this learning was apparently nothing compared to the "direct" experience of playing Kerbal Space Program, a rocket building and piloting sandbox game.

Orbital mechanics can be somewhat counterintuitive. The art of changing orbits involves relative velocities, positions, and times in complex interactions. As soon as you try deviating from a perfectly regular orbit, or start having to deal with N-body problems and orbital resonances, you have to coordinate your movements in possibly counterintuitive ways. One example is that if you want to reach an object ahead of you, on the same orbit, you actually have to 'brake' to reach a lower orbit. Once at that lower orbit, your angular velocity is faster, and you can start to overtake your target. After that manoeuver, you then have to accelerate to increase your orbital altitude again, which will end up reducing your angular speed so that you intercept your target.

At the title text Randall admits that at the time when he did work at NASA he was not involved in orbital mechanics—which is also true for most other NASA employees—but everybody was talking about this which in the end did increase his knowledge a little, as can be seen in the curve after the Job at NASA arrow.


How Well I Understand Orbital Mechanics:
[There is a graph with the x-axis labeled as 'time' increasing to the right. The y-axis is unlabeled but it can be assumed that it could be labeled 'understanding'.]
The line stays flat until the first bump, which is annotated "Took High School Physics".
The line decreases, then grows to a higher maximum with the annotation "Got Physics Degree".
There is a small amount of decrease until a gradual bump in the graph labeled "Actual Job At NASA".
It gradually decreases to pre-NASA levels, but then begins to increase exponentially. The annotation reads "Started Playing Kerbal Space Program".


comment.png add a comment! ⋅ comment.png add a topic (use sparingly)! ⋅ Icons-mini-action refresh blue.gif refresh comments!


Ksp actually doesn't model n body problems as referenced in the explanation. Ksp orbits are always perfectly stable unless the craft enters an atmosphere or changes sphere of influence. There are mods that change this but in the core game you only have to deal with the gravity of the main body you are orbiting.

I just put in a first attempt at the explanation. Could do with links to pages regarding KSP, etc, etc. (Or rewrite entirely how it ought to be done, of course.) Also, if anyone knows for sure that "aim nose at destination, fire retros", as seen in the film Gravity, would or would not give the desired effect, that'd be useful to clarify or dismiss. From my own experience with the Kerbals, it wouldn't (never mind all the other broad assumptions made in that otherwise spectacular film ), but KSP also rather fudges away the N-body problem, artificially. 05:38, 16 April 2014 (UTC)

You are correct that "aim nose, fire retros" doesn't work in reality. KSP fudges the n-body problem by putting the planets and moons on tracks, and then changing you into a different 2-body problem when you cross into a smaller hill sphere than the one you were in. The maneuver node system does a little bit of n-body work when you get a maneuver close to another body, but you'll notice that when the ship actually crosses into the other hill sphere the trajectory for the maneuver goes weird. It's a rather clever optimization for a simulator like KSP. lcarsos_a (talk) 06:21, 16 April 2014 (UTC)

I really wish there was a downwards curve for "I saw Armageddon". 06:26, 16 April 2014 (UTC)

There will be a huge upwards curve on "how much I think i know about orbital mechanics" - See Dunning-Krueger effect for more info. -- 14:40, 16 April 2014 (UTC)
In my opinion, the Dunning-Krueger effect is already present in the graph, at the downwards slope after every local maxima. As in, someone starts learning about something, they think they know all about it, and then they start realizing how much they don't know about it. 01:15, 6 November 2015 (UTC)
Future mission failure due to discrepancies in Kerbal Space Program

I note some differences in KSP (from wikipedia):

The game simulates trajectories and orbits using patched conic approximation instead of a full n-body simulation, and thus does not support Lagrange points and halo orbits.
The celestial bodies in the Kerbal solar system are about 1/10 the radius of their real-universe equivalents yet have comparable surface gravity, implying that they have unrealistically high densities. This change to scale makes many tasks considerably easier. For example, a surface to low-Kerbin-orbit launch requires a delta-v of about 4.5 km/s, compared to 9.5 km/s for a low-Earth-orbit launch. In particular, because of the game also having unrealistically efficient and flexible (in terms of speed and altitude) turbojet engines, this means it is much easier to make a single-stage-to-orbit vehicle using jet engines to accelerate a vehicle to orbital speed on only a small fraction of its mass in jet fuel, then give a tiny boost with rockets to reach orbit, whereas in real life, a highly efficient but powerful and lightweight scramjet would be necessary to do the same with several times the amount of fuel.

So I predict some probability that, after reading this comic, some NASA person will make the mistake of designing real missions using notions or designs from it, which will fail in real life (or at least be ridiculed at mission design review time). And then Randall will have to write a really challenging comic about it..... Nealmcb (talk) 13:12, 17 April 2014 (UTC)

Maybe Randall should add a horizontal line well over the curve, labelled: Level of knowledge required for a successful mission in real life - 10:32, 25 April 2014 (UTC)
Do you really think that a trained, qualified person at NASA, who had to go to school and study physics to plan missions, will be stupid enough to revert from his physics degree to Kerbal Space Program, thus reenacting 1244:_Six_Words? I think that unlikely. Jetman123 (talk) 07:53, 26 April 2014 (UTC)

Once at that lower orbit, your velocity is faster ... really? I though that on lower orbit, your velocity is slower BUT your ANGULAR velocity is faster, which is the reason you start to overtake your target ... but I never played Kerbal, so I may be wrong. -- Hkmaly (talk) 23:37, 24 April 2014 (UTC)

According to the Kepler's Equation a lower orbit means faster speeds. The Kerbal program is much more sophisticated and I even still did not figure out how to use my German keyboard on that Demo. Nevertheless, orbital mechanics are simple in general and then look at Neil Armstrong at Gemini 8 — moving around in weightlessness is not easy. --Dgbrt (talk) 21:09, 25 April 2014 (UTC)

He was born with innate knowledge of orbital mechanics equivalent to roughly freaking 20-25% of high school physics! THAT is Randall Munroe, ladies and gentlemen! 00:40, 7 June 2014 (UTC)

Not to put too fine a point on it, and I am in no way anything even close to resembling a physicist, engineer, etc., but I expect that most people are probably "born" (i.e. gain from birth to HS physics) with a similar level of understanding. I am happy to be corrected on this point, but I would imagine that the 20-25% level cited probably involves a basic conceptual understanding of Newtonian physics that we all gain from our life experience from birth to 17/18 years. From what I recall, HS physics clarified some of these principles and revealed the mathematical structures behind them. With that in mind, I'm pretty comfortable saying that my knowledge of physics pre-HS was about 1/4 of the final. And now that it's been about 15 years, I'm probably back down to the same level. Orazor (talk) 10:46, 7 October 2014 (UTC)
That, or the x axis isn't at y = 0, to make it more readable (my Economics Engineering teacher would death glare at me now...). 01:15, 6 November 2015 (UTC)