SHIRT: Satellite Hardware-In-the-loop Rendezvous Trajectories Dataset

This repository contains the Satellite Hardware-In-the-loop Rendezvous Trajectory (SHIRT) dataset which consists of two rendezvous trajectory scenarios (ROE1 and ROE2) in Low Earth Orbit (LEO) created from two different image sources. One is the OpenGL-based computer graphics renderer to create the synthetic images, and the other is the TRON facility at the Space Rendezvous Laboratory (SLAB) of Stanford University which illuminates a satellite mockup model with the Earth albedo light boxes to create the lightbox images. In ROE1, the servicer maintains the along-track separation typical of a standard v-bar hold point while the target spins about one principal axis, whereas in ROE2, the servicer slowly approaches the target tumbling about two principal axes. The sequential images of the SHIRT dataset can be used to evaluate the robustness of machine learning models and vision-based navigation filters over time across domain gap.

本仓库收录卫星硬件在环交会轨迹（Satellite Hardware-In-the-loop Rendezvous Trajectory，SHIRT）数据集，该数据集包含两类来自不同图像源的近地轨道（Low Earth Orbit，LEO）交会轨迹场景（ROE1与ROE2）。其中一类图像由基于OpenGL的计算机图形渲染器生成，为合成图像；另一类图像源自斯坦福大学空间交会实验室（Space Rendezvous Laboratory，SLAB）的TRON试验设施：该设施通过地球反照光影箱照射卫星样机模型，以此生成光箱图像。在ROE1场景中，服务航天器保持标准V-bar停靠点的沿轨分离姿态，目标航天器则绕单一根主轴自旋；在ROE2场景中，服务航天器缓慢接近绕两根主轴翻滚的目标航天器。SHIRT数据集的序列图像可用于评估机器学习模型与视觉导航滤波器在跨域差异下随时间推移的鲁棒性。

搜集汇总

背景与挑战

背景概述

SHIRT数据集包含两种低地球轨道会合轨迹场景的连续图像数据，分别来自计算机合成和实验室光箱模拟，用于评估机器学习模型和视觉导航系统的跨领域鲁棒性。