empyrean 0.0.1

High-performance astrodynamics in Python. Orbit propagation, ephemeris generation, and orbit determination powered by automatic differentiation.

Empyrean

by Empyrean Dynamics

High-performance astrodynamics in Python, powered by automatic differentiation.

Orbit propagation, ephemeris generation, and orbit determination.

Architecture

Empyrean is built on a stack of Rust libraries with Python bindings via PyO3.

• nolan — K-jet hyperdual numbers for automatic differentiation. Tracks arbitrary-order derivatives through physical parameters (position, velocity, time) without finite differences or hand-coded partials.

• villeneuve — Orbital propagator built on nolan. Two-body and n-body dynamics, relativistic force models, coordinate transforms, and covariance propagation.

• scott — Orbit determination. Differential corrections, initial orbit determination, and residual fitting. Uses the same AD pipeline for exact partials through the measurement model.

The full Empyrean Dynamics platform also includes:

• kubrick — REST API server exposing the propagation and orbit determination engines. Available as a hosted service at www.empyrean-dynamics.com, not included in the Python package.

• spielberg — Interactive web dashboard for orbit visualization, close approach analysis, and impact probability. Available at www.empyrean-dynamics.com, not included in the Python package.

Much more to come. Stay tuned.