I'm envisioning a series of moons in orbit of a super-earth (ideal) or gas giant. If technically workable, the super-earth would have earthlike gravity, the mass and size aren't important. If it's too hard to make it terrestrial, I can use a gas giant. What is important are the moons: 7 earthlike. In addition, there are another 5, smaller, habitable moons/planetoids which themselves orbit the 7 primaries and have roughly earthlike levels of gravity, if not the same density and mass.
Speaking of mass, I'd love to keep all of these moons as small as possible, while keeping earth gravity, but haven't worked out the math on that. I know that this could work in standard orbitals (though the number may be high), but I'd love to have the 7 operate in a horseshoe orbit, and I'm wondering how many objects can share an orbital by using horseshoe orbits.
This is probably a bit complex, so just to recap:
The Star: 'H' A central planet: 'N' 7 earthsized Primary moons & 5 orbiting Secondary moons: D&p V&r, W&i, E, A&z, L&s, F
Can those moons, share a single, complex, horseshoe orbit?
I'm thinking based on the sizes, that the primaries could simply be binary pairs with their secondaries. None of the primaries have more than one secondary, anyway, and it could help make the physics work.
The original idea was simply to have them on different orbitals to the giant, or to have them in a stable ring, spaced out on the same orbital, alongside an asteroid belt, but the horseshoe concept may work better with other things that I won't get into here, and I'd love to know if that high a number is possible.
There are 'magical' elements here, but I'd like to keep them out of the system structure as much as possible. Part of this concept is one original super-earth, which was fragmented into the moon system. I'm thinking maybe a civilization dragged a gas giant inner-system to create a brown dwarf, for a dyson sphere or something, and ended up failing spectacularly, and breaking a large terrestrial planet into fragments later captured by 'N'.
I simply can't find any solar system modelling tools that are complex enough to handle this or tell me if there are stable solutions given the constraints. (P.S. if you know of a tool that can do that, I'd love to know).