The S3LI Vulcano Dataset: A Dataset for Multi-Modal SLAM in Unstructured Planetary Environments

2026-01-27

AI Summary

This work addresses the lack of publicly available multimodal benchmarks for simultaneous localization and mapping (SLAM) and place recognition in unstructured planetary-like environments. To bridge this gap, the authors present S3LI Vulcano, the first multimodal dataset collected in a real-world planetary-analog setting on Vulcano Island, Italy, featuring diverse terrains, rocks, and vegetation with synchronized visual and LiDAR data. The dataset is accompanied by an open-source toolchain that enables ground-truth pose generation and sample annotation, establishing a unified evaluation platform for SLAM and place recognition algorithms. This contribution fills a critical void in the field by providing a high-quality, real-world benchmark tailored to the challenges of non-structured, extraterrestrial-like environments.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

We release the S3LI Vulcano dataset, a multi-modal dataset towards development and benchmarking of Simultaneous Localization and Mapping (SLAM) and place recognition algorithms that rely on visual and LiDAR modalities. Several sequences are recorded on the volcanic island of Vulcano, from the Aeolian Islands in Sicily, Italy. The sequences provide users with data from a variety of environments, textures and terrains, including basaltic or iron-rich rocks, geological formations from old lava channels, as well as dry vegetation and water. The data (rmc.dlr.de/s3li_dataset) is accompanied by an open source toolkit (github.com/DLR-RM/s3li-toolkit) providing tools for generating ground truth poses as well as preparation of labelled samples for place recognition tasks.

Problem

• SLAM
• place recognition
• multi-modal
• planetary environments
• dataset

Innovation

• multi-modal SLAM
• planetary environments
• LiDAR-visual dataset
• place recognition
• ground truth generation