Ballistic capture possible with KSP?

[Constans]

I read the linked article on arXiv about using ballistic capture for a Mars transfer, which uses significantly less delta V than a Hohmann transfer, and I wondered if this would be possible in KSP, i.e. does the game engine model enough of the physics to allow a similar approach to be used in game? For your kind attention, computationally-inclined ladies and gentlemen: arxiv.org/pdf/1410.8856.pdf

[Scotius]

I've seen similiar, spaghettified orbits during Mun\Minmus transfers. Also around Duna, when Ike decided to be a jerk and mess with my ships trajectories. So...maybe?

[Torquemadus]

If I'm reading the article correctly, the idea is to throw the spacecraft towards Mars on a trajectory somewhat similar to a Hohmann transfer, but without enough speed to make it all the way to Mars' orbit. The idea is that if the spacecraft passes close enough to Mars to be considered to be within Mars' gravitational SOI, then Mars' gravity will grab hold of the spacecraft and pull it the rest of the way.

A Hohmann transfer has to be aimed precisely at Mars, while a ballistic capture only needs to be aimed at Mars' SOI, which is a much bigger target. This therefore allows for much greater flexibility in terms of launch windows.

I'm not an expert at mathematics, so if I misunderstood the article. Please correct me!

[Superfluous J]

This won't work in KSP because Duna won't start pulling on your craft until you're actually in the SOI. When you enter Duna's SOI you'll go into "orbit" around it but part of your orbit will fall outside the SOI (because you just came from there). You'll swing around the planet and head back out, passing back out of Duna's SOI to be forever gone.

This works in real life because we don't use patched conics. All bodies in the Solar System (and the Universe, really) are constantly pulling on all other bodies, which creates some very interesting effects that just won't happen in Kerbal Space Program.

[GoSlash27]

Constans,

Since KSP only models one SOI at a time, this technique would not work.

Sorry and Merry Christmas,

-Slashy

[Constans]

Thanks for the quick response guys, and hope you're all having a good Xmas. I'm guessing the computation power required to model multiple gravity wells is too much for home computing with too little pay-off. Perhaps in some far future version of KSP.

[cantab]

You could have a look at the Principia mod, which aims to add n-body gravity to KSP.

I've not tried it myself but it seem to show that it's computationally feasible at least, despite the demands.

I think KSP uses patched conics for simplicity. Simplicity to program, since Keplerian orbits and patched conics are basic geometry really, whereas n-body gravity requires more advanced mathematics. And simplicity to play, since you have perfectly stable orbits and clear SOI transitions.

[Kesa]

there is a mod that compute n body physics in ksp, but only for the active vessel.

Edit : ninja'd. there are at least two modes that aim to simulate nbody physic (orbit manipulator and principia)

Edited December 25, 2014 by Kesa

[RobotsAndSpaceships]

Nope, it requires n-body physics.

[Northstar1989]

Nope, it requires n-body physics.

But there are two mods that simulate N-body physics in KSP:

http://forum.kerbalspaceprogram.com/threads/70881

http://forum.kerbalspaceprogram.com/threads/68502-WIP-Principia-N-Body-Gravitation-and-Better-Integrators-for-Kerbal-Space-Program

Regards,

Northstar