FANTASY: User-friendly Symplectic Geodesic Integrator for Arbitrary Metrics with Automatic Differentiation

We present FANTASY (Finally A Numerical Trajectory Algorithm both Straightforward and sYmplectic), a user-friendly, open-source symplectic geodesic integrator written in Python. FANTASY is designed to work “out of the box” and does not require anything from the user aside from the metric and the ini...

Full description

Saved in:
Bibliographic Details
Published inThe Astrophysical journal Vol. 909; no. 1; pp. 67 - 75
Main Authors Christian, Pierre, Chan, Chi-kwan
Format Journal Article
LanguageEnglish
Published Philadelphia The American Astronomical Society 01.03.2021
IOP Publishing
Subjects
Online AccessGet full text
ISSN0004-637X
1538-4357
1538-4357
DOI10.3847/1538-4357/abdc28

Cover

More Information
Summary:We present FANTASY (Finally A Numerical Trajectory Algorithm both Straightforward and sYmplectic), a user-friendly, open-source symplectic geodesic integrator written in Python. FANTASY is designed to work “out of the box” and does not require anything from the user aside from the metric and the initial conditions for the geodesics. FANTASY efficiently computes derivatives up to machine precision using automatic differentiation, allowing the integration of geodesics in arbitrary space(times) without the need for the user to manually input Christoffel symbols or any other metric derivatives. Further, FANTASY utilizes a Hamiltonian integration scheme that doubles the phase space, where two copies of the particle phase space are evolved together. This technique allows for an integration scheme that is both explicit and symplectic, even when the Hamiltonian is not separable. FANTASY comes prebuilt with second- and fourth-order schemes, and is easily extendable to higher-order schemes. FANTASY also includes an automatic Jacobian calculator that allows for coordinate transformations to be done automatically.
Bibliography:AAS27219
High-Energy Phenomena and Fundamental Physics
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0004-637X
1538-4357
1538-4357
DOI:10.3847/1538-4357/abdc28