davidhyman / ksp_bodies_graph Goto Github PK

Why is Jool rendered on the left? None of the paths should need to cross!

The controls. They could be better.

Probably swanky popup menus or something. Currently hard to control from a phone, for example. Also currently call their triggers at init (so we can get label text). Probably shouldn't.

Make it more useable for KSP mods etc.

Perhaps if it could use data cascading, people could easily add a few tweaks on top of the base data (e.g. to update only one transition).

As such, 'perfect' aerobraking is disabled.

It should be finding whether the target SOI has atmosphere and use that for 'perfect' braking.

But I also wonder whether we are wrong about values for child soi -> parent orbit? Generally, I am suspicious of the nodes with 'no circularisation' flagged against them. Perhaps we should always circularise endpoints? Or revise the overall calculation so it depends entirely on the route (which is still graph edges) rather than the edges themselves (which is fixed regardless of journey?).

Going from Duna -> Jool, one is routed through Kerbin's SOI.

This also applies to the child planets e.g. Laythe -> Pol (currently it takes us back to a highly elliptical Jool capture orbit).

Once the SOI is at the same depth in the hierarchy (relative to Kerbol) we should transfer directly to an interception in the other SOI.

Can this be solved using just dv averages and a smattering of extra edges on the graph? Or should I be doing something entirely different? (see https://github.com/alexmoon/ksp)

Need planet stats for new universe representations

Will need edge overrides to work properly, and perhaps a way to stop auto-generated edges if a path vai an override exists?