Modular pipeline for small bodies gravity field modeling: an efficient representation of variable density spherical harmonics coefficients

TL;DR

This paper introduces a modular pipeline for efficiently generating variable density gravity field models of small bodies, aiding autonomous guidance, navigation, and control in space missions by providing accurate and computationally efficient simulations.

Contribution

It presents a novel, robust, and faster framework for modeling small bodies' gravity fields using spherical harmonics, validated against various models and analytical solutions.

Findings

Demonstrates high accuracy in gravity field modeling

Shows computational efficiency over existing methods

Validates robustness across diverse small body shapes

Abstract

Proximity operations to small bodies, such as asteroids and comets, demand high levels of autonomy to achieve cost-effective, safe, and reliable Guidance, Navigation and Control (GNC) solutions. Enabling autonomous GNC capabilities in the vicinity of these targets is thus vital for future space applications. However, the highly non-linear and uncertain environment characterizing their vicinity poses unique challenges that need to be assessed to grant robustness against unknown shapes and gravity fields. In this paper, a pipeline designed to generate variable density gravity field models is proposed, allowing the generation of a coherent set of scenarios that can be used for design, validation, and testing of GNC algorithms. The proposed approach consists in processing a polyhedral shape model of the body with a given density distribution to compute the coefficients of the spherical harmonics expansion associated with the gravity field. To validate the approach, several comparison are conducted against analytical solutions, literature results, and higher fidelity models, across a diverse set of targets with varying morphological and physical properties. Simulation results demonstrate the effectiveness of the methodology, showing good performances in terms of modeling accuracy and computational efficiency. This research presents a faster and more robust framework for generating environmental models to be used in simulation and hardware-in-the-loop testing of onboard GNC algorithms.