RTK-SLAM Dataset: Handheld LiDAR-Visual-Inertial Recordings with RTK GNSS for Absolute Accuracy in GNSS-Degraded Environments (University of Stuttgart)
- Provider
- Institute for Photogrammetry and Geoinformatics, University of Stuttgart (Zhang, Ress, Skuddis, Soergel, Haala)
- License
- CC-BY-4.0 (source)
- Size
- 18 files
- Stored at
- Hugging Face (https://huggingface.co/datasets/Willyzw/rtk-slam-dataset)
- Access
- Free and public. No account with the data provider is needed.
- Category
- Robotics
- Source
- https://huggingface.co/papers/2604.07151
About this dataset
A benchmark for the absolute (georeferenced) accuracy of SLAM and sensor fusion when satellite positioning is partly blocked, from the Institute for Photogrammetry and Geoinformatics at the University of Stuttgart. A handheld rig carried a Livox MID360 LiDAR with built-in IMU (10 Hz scans, 200 Hz inertial), a 2 MP global-shutter camera at 20 Hz and a Unicore UM980 GNSS receiver at 10 Hz with real-time kinematic (RTK) corrections from the German SAPOS service. Four sequences were walked: two in Stuttgart's Stadtgarten park (1.04 km and 0.46 km, with tree cover, building shadowing and a 30 m underpass) and two at the IntCDC construction hall (0.48 km and 0.39 km, with a GNSS-denied indoor hall of more than 400 s), with 16 to 36 surveyed checkpoints per sequence and RTK fix rates of 23 to 54 percent. Every sequence is provided as a ROS 1 bag, a ROS 2 db3 bag and an extended EuRoC-format zip compatible with OKVIS2-X; camera timestamps are already corrected for the trigger delay. About 182 GB (measured 2026-11-21). Released under CC BY 4.0; cite Zhang et al., ISPRS Annals 2025.
How to load this dataset in Python
from huggingface_hub import hf_hub_download
path = hf_hub_download(repo_id="Willyzw/rtk-slam-dataset", filename="euroc/construction_seq1_euroc.zip", repo_type="dataset")
print(path)Data structure
Files
Example file: euroc/construction_seq1_euroc.zip
Analyze this dataset without downloading it
On Scigantic, RTK-SLAM Dataset: Handheld LiDAR-Visual-Inertial Recordings with RTK GNSS for Absolute Accuracy in GNSS-Degraded Environments (University of Stuttgart) opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.
This page is also available as Markdown and as JSON.
Related Robotics datasets
- DexWild: Human and Robot Dexterous Hand Demonstrations in the Wild
- PhysicalAI Robotics Mindmap GR 1 Stick In Bin
- PhysicalAI GR 00 T Tuned Tasks
- Ego-Tactile Manipulation: Egocentric Video with 160-Channel Two-Hand Tactile Glove Pressure and Touch-Grounded Action Labels (OpenGraph Labs)
- T-Rex: Tactile-Reactive Bimanual Dexterous Manipulation Dataset
- Monado SLAM Datasets (MSD): Egocentric Visual-Inertial Recordings from Valve Index, HP Reverb G2 and Samsung Odyssey+ VR Headsets (Collabora)
- Video to Data: Robot Dexterity Task Library and Dataset
- PhysicalAI Robotics GR 00 T Teleop GR 1