group
What campus are you from?
Daytona Beach
Authors' Class Standing
Ashton Steed, Senior Maybelline Flesher, Sophomore, Jarrett Usui, Sophomore Jacob Becker, Junior
Lead Presenter's Name
Ashton Steed
Faculty Mentor Name
Steve Giliam
Abstract
High Precision Orbital Propagators (HPOP) are programs designed to predict the state of a body under the influence of orbital dynamics. For Low-Earth Orbit (LEO) satellites, a variety of perturbation sources are modeled to reflect real world dynamics, such as variations in the Earth’s gravitational field. To isolate the effect of the geopotential model, other perturbations - atmospheric drag and third-body gravitation - are omitted from this analysis. This study examines the speed-accuracy tradeoff for HPOP simulation, quantifying the computational cost required to stay within a given maximum daily position drift. Using ERAU’s VEGA HPC cluster, a reference LEO satellite (500 km circular orbit) is propagated over 24 hours using an adaptive step size integrator. The reference trajectory modeled using the full Earth Gravitational Model 2008 (EGM2008) - complete to degree and order 2159 - serves as the ground truth. Simulations are timed for various maximum orders and degrees of EGM2008, and their position errors with respect to the reference are gathered. Adaptive step size ensures that the runtime holistically represents the compounding cost of fine variations in the Geopotential. Using this timing, the Speed-Accuracy tradeoff for various orders of truncation as well as metrics of calculation efficiency is found and analyzed. A heuristic formula for truncation necessary to achieve a desired maximum daily drift will be established, minimizing unnecessary compute work based on use-case requirements. The resulting formula quantitatively represents the relationship between geopotential fidelity and propagator drift, allowing for optimization of computational resources for future LEO mission analysis.
Did this research project receive funding support from the Office of Undergraduate Research.
No
LEO-HPOP
High Precision Orbital Propagators (HPOP) are programs designed to predict the state of a body under the influence of orbital dynamics. For Low-Earth Orbit (LEO) satellites, a variety of perturbation sources are modeled to reflect real world dynamics, such as variations in the Earth’s gravitational field. To isolate the effect of the geopotential model, other perturbations - atmospheric drag and third-body gravitation - are omitted from this analysis. This study examines the speed-accuracy tradeoff for HPOP simulation, quantifying the computational cost required to stay within a given maximum daily position drift. Using ERAU’s VEGA HPC cluster, a reference LEO satellite (500 km circular orbit) is propagated over 24 hours using an adaptive step size integrator. The reference trajectory modeled using the full Earth Gravitational Model 2008 (EGM2008) - complete to degree and order 2159 - serves as the ground truth. Simulations are timed for various maximum orders and degrees of EGM2008, and their position errors with respect to the reference are gathered. Adaptive step size ensures that the runtime holistically represents the compounding cost of fine variations in the Geopotential. Using this timing, the Speed-Accuracy tradeoff for various orders of truncation as well as metrics of calculation efficiency is found and analyzed. A heuristic formula for truncation necessary to achieve a desired maximum daily drift will be established, minimizing unnecessary compute work based on use-case requirements. The resulting formula quantitatively represents the relationship between geopotential fidelity and propagator drift, allowing for optimization of computational resources for future LEO mission analysis.