LCDC: Bridging Science and Machine Learning for Light Curve Analysis

TL;DR

The paper addresses the lack of reproducible, comparable light-curve analyses for artificial space objects by introducing LCDC, a Python toolkit for preprocessing, dataset creation, and ML-ready analysis, along with the RoBo6 benchmark dataset sourced from MMT. It demonstrates the utility of LCDC through diverse use cases, including standardized rocket-body classification (RoBo6), surface-property inference for Atlas 2AS, and rotational-dynamics analysis of the Delta 4 rocket body, supported by the publicly accessible MMT_snapshot. The approach emphasizes standardized preprocessing, large-scale data handling, and transparent benchmarking to enable fair comparisons across methods and datasets, thereby advancing space debris characterization and sustainable space exploration. The work contributes an open-source framework, a publicly documented snapshot dataset, and concrete scientific demonstrations that showcase both AI-driven insights and physical-interpretation capabilities of space objects.

Abstract

The characterization and analysis of light curves are vital for understanding the physical and rotational properties of artificial space objects such as satellites, rocket stages, and space debris. This paper introduces the Light Curve Dataset Creator (LCDC), a Python-based toolkit designed to facilitate the preprocessing, analysis, and machine learning applications of light curve data. LCDC enables seamless integration with publicly available datasets, such as the newly introduced Mini Mega Tortora (MMT) database. Moreover, it offers data filtering, transformation, as well as feature extraction tooling. To demonstrate the toolkit's capabilities, we created the first standardized dataset for rocket body classification, RoBo6, which was used to train and evaluate several benchmark machine learning models, addressing the lack of reproducibility and comparability in recent studies. Furthermore, the toolkit enables advanced scientific analyses, such as surface characterization of the Atlas 2AS Centaur and the rotational dynamics of the Delta 4 rocket body, by streamlining data preprocessing, feature extraction, and visualization. These use cases highlight LCDC's potential to advance space debris characterization and promote sustainable space exploration. Additionally, they highlight the toolkit's ability to enable AI-focused research within the space debris community.

Figures (9)

• Figure 1: Example of Falcon 9 light curve from MMT database mmtWeb.
• Figure 2: Reconstructed rotation phase of spared Ariane 5 upper stage published in the SDLCD. For this specific light curve, the apparent rotation period of the object was found to be 12.1 seconds. Source: https://www.sdlcd.space-debris.sk
• Figure 3: Visualisation of a part of the light curve with ID 5231905 of a CZ-3B rocket body with NORAD ID 29517.
• Figure 4: Coefficient $a_{0,st}$ calculated from standard magnitudes of object ATLAS 2AS CENTAUR R/B as a function of phase angle.
• Figure 5: Coefficient $a_{0,red}$ calculated from $a_{0,st}$ coefficients of object ATLAS 2AS CENTAUR R/B as a function of phase angle (grey points). Plotted is the fit assuming a diffuse/specular sphere (blue dashed line) and 95$\%$ prediction interval (grey area).