Robotic Lava Tube Mapping and Multimodal Data Collection Using Quadruped and LiDAR

As part of the Rhizome 2.0 project—an initiative investigating the development of human habitats in Martian lava tubes—we conducted an extensive robotic mapping mission inside the Grotta di Monte Intraleo, a terrestrial lava tube in Sicily serving as an analogue site. This dataset supports the exploration of construction and habitation strategies in similarly structured Martian environments.The dataset includes multi-modal mapping and environmental data collected using manual and robotic scanning techniques. Specifically, the data comprises:<strong>High-resolution floor reference images</strong>: Manually collected photographs of the lava tube floor were taken at a constant height and under controlled illumination to serve as visual calibration and textural references for scale and surface feature analysis.<strong>3D mesh data from mobile scanning</strong>: Phone-based 3D scans were conducted using the Scaniverse app, producing <code>.obj</code> files for rapid spatial documentation.<strong>LiDAR scans</strong>: High-density LiDAR point clouds of the tube interior provide accurate geometric representations.<strong>Robot FPV video footage</strong>: A robotic quadruped equipped with a forward-facing camera collected immersive, first-person video while navigating the cave, providing visual context and documenting terrain traversal.<strong>SLAM-based navigation maps</strong>: Simultaneous Localization and Mapping (SLAM) data recorded during robotic traversal were used to generate autonomous navigation maps.<strong>Ambient environmental data</strong>: Time-synchronized logs from onboard sensors recorded temperature, humidity, and light levels, contributing to environmental characterization of the site.<strong>System interface recordings</strong>: Screen captures of the framework’s user interface during data acquisition sessions offer insights into the control, mapping, and visualization tools used throughout the mission.Data collection adhered to all local regulations, and care was taken to minimize impact on the natural environment of the cave. This dataset is intended to support reproducible research in robotic mapping, autonomous navigation, and extraterrestrial habitat design.

本数据集隶属于根茎2.0（Rhizome 2.0）项目——该项目旨在研究火星熔岩管内人类居住站的建设方案——我们在西西里岛的蒙特因特拉莱奥熔岩洞（Grotta di Monte Intraleo，一处作为地外环境类比研究站点的地球熔岩洞）内部开展了大规模机器人测绘任务。本数据集旨在助力探索结构类似的火星环境中的建造与居住策略。

本数据集包含通过手动与机器人扫描技术采集的多模态测绘与环境数据，具体包括：
1. **高分辨率地面参考影像**：以固定高度、受控光照条件手动拍摄熔岩洞地面的照片，用于视觉校准以及作为尺度与地表特征分析的纹理参照。
2. **移动扫描三维网格数据**：采用Scaniverse应用程序完成基于手机的三维扫描，生成用于快速空间记录的.obj格式文件。
3. **激光雷达（LiDAR）扫描数据**：采集洞穴内部的高密度激光雷达点云，以提供精准的几何形态表征。
4. **机器人第一视角（FPV）视频素材**：搭载前置摄像头的四足机器人在洞穴内导航时，采集沉浸式第一视角视频，用于提供场景视觉信息并记录地形遍历过程。
5. **基于同步定位与建图（Simultaneous Localization and Mapping，简称SLAM）的导航地图**：机器人遍历洞穴时记录的同步定位与建图数据，用于生成自主导航地图。
6. **环境参数数据**：机器人搭载传感器的时间同步日志记录了温度、湿度与光照水平，助力该站点的环境特征刻画。
7. **系统界面录制素材**：数据采集过程中采集的系统用户界面屏幕截图，可用于解析本次任务中使用的控制、建图与可视化工具。

本次数据采集严格遵守当地所有法规，并尽可能降低对洞穴自然环境的影响。本数据集旨在为机器人建图、自主导航以及地外居住站设计领域的可复现研究提供支撑。