Roadmap

orbitr is a work in progress. This page is a running list of features I’m thinking about adding in future versions. Nothing here is promised, and the priority order is loose — it mostly reflects what I personally find interesting or what users have asked for. If any of these sound useful (or terrible), let me know. Suggestions, feedback, and pull requests are all very welcome.

Physics

A `radius` argument on `add_body()`

Right now bodies are treated as point masses. This is fine for most orbital distances — gravity outside a sphere behaves exactly as if all the mass were concentrated at its center (the shell theorem) — but it means there’s no concept of two bodies physically touching. Adding an optional radius parameter would enable collision detection, merging on contact, and more realistic close-encounter behavior.

I’d probably add it to add_body() as an optional parameter with a sensible default, something like:

add_body <- function(system, id, mass, x = 0, y = 0, z = 0,
                     vx = 0, vy = 0, vz = 0, r = NULL)

If r is supplied for any body in the system, the integrator would check for overlaps on each step and handle them according to a user-chosen policy (elastic bounce, inelastic merge, simulation halt, etc.).

In the meantime, the existing softening parameter on simulate_system() partly works around the missing-radius problem by preventing the gravitational force from blowing up at very small separations — see The Physics for details.

Non-gravitational forces

Optional support for forces beyond pure Newtonian gravity:

Atmospheric drag for low orbits around bodies with atmospheres
Radiation pressure for small bodies near a star
J2 oblateness corrections for orbits around non-spherical bodies (Earth’s equatorial bulge measurably perturbs satellite orbits)

General-relativistic corrections

A small post-Newtonian correction term would let orbitr reproduce real GR effects like the precession of Mercury’s perihelion. Probably opt-in via an argument on simulate_system(), since most users wouldn’t need it.

Setup helpers

Construct bodies from Keplerian orbital elements ✅ Added in v0.2.0

Implemented as add_body_keplerian(). See ?add_body_keplerian for details.

A `load_solar_system()` convenience ✅ Added in v0.2.0

Implemented as load_solar_system(). Builds the Sun, all eight planets, the Moon, and Pluto using real Keplerian elements from JPL DE440. See ?load_solar_system for details.

Quality of life

Save and load simulation state

Functions to serialize an orbit_system (or a completed simulation tibble) to disk and reload it later. Useful for long-running simulations and for sharing reproducible setups.

Progress bar for long simulations

A simple progress indicator on simulate_system() for runs that take more than a few seconds, with an option to disable it for scripted use.

Built-in conservation diagnostics

Helpers to compute total energy, total linear momentum, and total angular momentum at every time step, so you can sanity-check an integrator’s behavior over the course of a run. Useful for confirming Velocity Verlet is doing its job, and for spotting cases where the time step is too large.