2025-07-24 –, Main Room 5
Accurate and efficient astrodynamics models are central to a wide range of space flight applications. The conventional approach to implementing these astrodynamics models in software often struggles with scalability and real-time performance, limiting their utility in modern space applications. In this talk, we introduce AstroForceModels.jl and AstroPropagators.jl, a suite of open-source Julia libraries designed to provide high-performance and differentiable solutions for orbit propagation.
The Julia programming language is increasingly recognized for its ability to provide high-performance numerical computing while maintaining a clear and expressive syntax. This makes it an ideal platform for astrodynamics, where computational efficiency, accuracy, and flexibility are paramount. AstroForceModels.jl and AstroPropagators.jl together form a comprehensive ecosystem that enhances Julia’s capabilities for space applications, offering modularity and differentiability—key features that set them apart from traditional tools.
AstroForceModels.jl
Accurate force models are critical for orbit propagation, as small perturbations accumulate over time, affecting long-term predictions. AstroForceModels.jl provides a structured and extensible approach to defining perturbative forces acting on a spacecraft. It currently includes: gravitational models, high-fidelity atmospheric drag models, solar radiation pressure, third-body perturbations, and relativistic force models.
The library is designed to integrate seamlessly with AstroPropagators.jl, ensuring that force models can be easily composed and customized for a wide range of mission scenarios. It is built off of the SatelliteToolbox ecosystem, leveraging a validated and extensive set of libraries and extending the capabilities for space mission design and analysis inside of Julia.
AstroPropagators.jl
Orbit propagation is a core requirement for many astrodynamics applications. AstroPropagators.jl provides integration with DifferentialEquations.jl supporting a wide range of solvers, including symplectic and adaptive methods. The combination of these libraries enables differentiable propagation which facilitates a number of applications, including machine learning, optimal control, and variational methodologies. These features make AstroPropagators.jl an essential tool for scalable and high-precision orbit propagation, outperforming legacy software in both flexibility and computational efficiency.
The Julia astrodynamics community has seen rapid growth, with tools like Tempo.jl, OrbitalTrajectories.jl, and SatelliteToolbox.jl providing essential functionalities. However, prior to these packages, Julia lacked a unified, high-performance framework for orbit propagation and force modeling. By integrating seamlessly with Julia’s numerical computing stack, AstroForceModels.jl and AstroPropagators.jl significantly enhance the astrodynamics ecosystem.
Jordan Murphy is a Senior Astrodynamics Engineer at Slingshot Aerospace and a Ph.D. candidate in Aerospace Engineering at the University of Colorado Boulder under Professor Dan Scheeres. His expertise lies at the intersection of astrodynamics, scientific machine learning, numerical computing, and optimization. With experience spanning NASA’s Jet Propulsion Laboratory, Lawrence Livermore National Laboratory, CU Aerospace, and SpaceX, Jordan has contributed to mission design, machine learning and reinforcement learning applications, and high-performance computing for space systems. A strong advocate for the Julia programming language, he integrates cutting-edge computational techniques into astrodynamics research and development. Jordan’s current work focuses on enhancing space situational awareness, data-driven simulation methodologies, and optimal control.