GPU-based high-precision orbital propagation of large sets of initial conditions through Picard-Chebyshev augmentation

TL;DR

This paper introduces a GPU-accelerated method using Picard-Chebyshev augmentation for high-precision orbital propagation of large initial condition sets, significantly improving computational efficiency for space mission analyses.

Contribution

It presents a novel two-level augmentation scheme for efficient, high-accuracy orbital simulations on GPUs, accommodating complex force models within a single algorithm call.

Findings

Achieved high performance with C and CUDA implementations.

Validated method on Solar Orbiter-like orbital case.

Enabled multiple simulations simultaneously with GPU acceleration.

Abstract

The orbital propagation of large sets of initial conditions under high accuracy requirements is currently a bottleneck in the development of space missions, e.g. for planetary protection compliance analyses. The proposed approach can include any force source in the dynamical model through efficient Picard-Chebyshev (PC) numerical simulations. A two-level augmentation of the integration scheme is proposed, to run an arbitrary number of simulations within the same algorithm call, fully exploiting high performance and GPU (Graphics Processing Units) computing facilities. The performances obtained with implementation in C and NVIDIA CUDA programming languages are shown, on a test case taken from the optimization of a Solar Orbiter-like first resonant phase with Venus.