Hacker News Reader

Comment by brianbreslin | Hacker News Reader

It amazes me that we take for granted the sheer complexity of hitting a moving target like a comet at such an incredible distance in space. Navigation in space has to be so incredibly complex because you are moving in so many planes. It's like firing a gun and hitting a quarter that was thrown miles away.

View on HN

If you haven't already, do watch which orbit changes of Rosetta were needed to deploy the lander:

https://www.youtube.com/watch?v=4a3eY5siRRk

Must watch. The video was made by ESA.

That is astonishing. Goes to show how simple stuff can turn into hellish soup once all the requirements are in.

The video is very cool. But I find it hard to imagine how to orbit something that has hardly any gravity - the orbiter must be getting just milligrams of force.[1] The scale isn't clear from the video: are they orbiting really slowly? And it seems like it would be very hard to enter orbit around something that has hardly any pull without flying right past it. On the positive side, the orbit changes in the video would require almost no propulsion, right?

[1] milligrams isn't a unit of force, but hopefully you know what I mean; milli-ounces sounds too weird.

9.8 micronewtons is the correct way of saying one milligram of force.

Milli-Newtons :-)

The comet[1] that Rosetta is orbiting has a mass of 10^13 kg and an orbit semi-major axis of 518,060,000 km. This gives the comet a Sphere of Influence[2] of roughly 62 km.

What this means is that within 62 km or so, the comet is the dominant gravitational body, so orbits of 16, 20, and 30 km (in the video) are perfectly reasonable.

The orbital period[3] of the larger 30 km orbit is around 40 ksec or 11.1 hours.

[1]: http://en.wikipedia.org/wiki/67P/Churyumov%E2%80%93Gerasimen...

[2]: http://en.wikipedia.org/wiki/Sphere_of_influence_(astrodynam...

[3]: http://en.wikipedia.org/wiki/Orbital_period

Your orbital period of 11.1 hours at 30 km looks wrong. The video shows about 7/8 of a 30-km orbit over 14 days (19 Nov - 3 Dec). The orbital period formula gives 14.6 days (2 * 3.14159265 * sqrt(30000^3/(6.67384e-11 * 10^13)) / 3600 / 24).

That's absolutely possible -- I did some in-my-head units conversion.

Maybe another look at the units:

  G - 6.67384e-11  [N (m/kg)^2] = [m^3 / (kg s^2)]
  M - 1.0e13       [kg]
  a - 30000        [m]

  sqrt([m^3] / ([m^3 / (kg s^2)] * [kg]))
  = sqrt([s^2]) = [s]

  2*pi*sqrt(30000^3 / (6.67384e-11 * 1.0e13))
  = 1,263,787 s
  = 14.6 days

Well, how about that :-)

What does a "relay phase manoeuvre" mean, and why are they performed? Those are the weirdest ones, looking very much like "heh, let's go in some random direction that we didn't try yet!"...

These maneuvers look strange because Rosetta is too far from the comet at those points for a real orbit. The probe can change course freely without the gravity of the comet affecting it much.

The reason for the maneuvers is to maintain line of sight with the Philae lander so that communications between Earth and Philae can be relayed, hence "relay phase". Then the probe inserts itself back into an orbit.

Thanks a lot for the explanations!

A better analogy would be firing a bullet that has the ability to correct its course in flight, which is a far less daunting challenge.

Assuming you can, you know, build a vehicle that can correct its course in flight in space, right? :)

I don't think "we" do take it for granted. It's just that those of us who don't can't think of much more to say than "Woah! on a comet?! wow." This is incomprehensibly amazing to me.

It is stunning.

Thinking about navigating in zero gravity became a lot easier for me when I realised that you can arbitrarily redefine “down” as whatever direction is convenient. Gravity assists and orbits, for example, are simply falling around a bend.

The enemy gate is down

Reply on news.ycombinator.com