{"success":true,"data":{"author":"Institute for Photogrammetry and Geoinformatics, University of Stuttgart (Zhang, Ress, Skuddis, Soergel, Haala)","bucketAccessType":"public","bucketName":"Willyzw/rtk-slam-dataset","bucketPrefix":"","category":"Robotics","cloudProvider":"huggingface","createdAt":"2026-11-21T22:15:32.861Z","description":"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.","displayTitle":"RTK-SLAM Dataset: Handheld LiDAR-Visual-Inertial Recordings with RTK GNSS for Absolute Accuracy in GNSS-Degraded Environments (University of Stuttgart)","huggingFaceUrl":"https://huggingface.co/datasets/Willyzw/rtk-slam-dataset","id":"d1bbac85-8c2d-5f09-aa77-e607266085d9","license":"CC-BY-4.0","licenseEvidenceUrl":"https://huggingface.co/datasets/Willyzw/rtk-slam-dataset","objectCountEstimate":18,"probeStatus":"ok","referenceUrl":"https://huggingface.co/papers/2604.07151","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"Willyzw/rtk-slam-dataset\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"Willyzw/rtk-slam-dataset\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=Willyzw, license:cc-by-4.0, size_categories:10K<n<100K, format:imagefolder, downloads=1116"],"format":"mixed","generatedAt":"2026-11-21T22:31:25.029Z","layout":{"example":"euroc/construction_seq1_euroc.zip","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"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","encoding":"hex","key":"euroc/construction_seq1_euroc.zip","truncatedAt":512}],"sampleObjects":["euroc/construction_seq1_euroc.zip","euroc/construction_seq2_euroc.zip","euroc/stadtgarten_seq1_euroc.zip","euroc/stadtgarten_seq2_euroc.zip","ros1/construction_seq1.bag"],"sidecars":[{"bytes":"---\nlanguage:\n- en\nlicense: cc-by-4.0\nsize_categories:\n- 100G<n<1T\ntask_categories:\n- robotics\npretty_name: RTK-SLAM Dataset\ntags:\n- slam\n- rtk\n- lidar\n- imu\n- gnss\n- localization\n- georeferencing\n- ros\n---\n\n# RTK-SLAM Dataset\n\n**An RTK-SLAM Dataset for Absolute Accuracy Evaluation in GNSS-Degraded Environments**\n\n> Wei Zhang, Vincent Ress, David Skuddis, Uwe Soergel, Norbert Haala  \n> Institute for Photogrammetry and Geoinformatics, University of Stuttgart, Germany\n\n[[Project Page]](https://rtk-slam-dataset.github.io/) | [[Paper]](https://huggingface.co/papers/2604.07151) | [[Code]](https://github.com/Willyzw/rtk-slam-eval)\n\n---\n\n## Overview\n\nThis dataset is designed for evaluating the **absolute global positioning accuracy** of RTK-SLAM systems in GNSS-degraded and GNSS-denied environments. A key design principle is that the RTK receiver is used exclusively as a **system input**, while ground truth is established independently via a geodetic total station. This separation is absent from all existing datasets.\n\nStandard SLAM benchmarks evaluate accuracy using SE(3)-aligned ATE, which absorbs global drift and can underestimate absolute errors by up to **76%**. This dataset enables direct evaluation without SE(3) alignment.\n\n---\n\n## Sensor Platform\n\nA handheld device comprising:\n\n| Sensor | Model | Rate |\n|---|---|---|\n| LiDAR + IMU | Livox MID360 | 10 Hz / 200 Hz |\n| Camera | 2 MP global shutter | 20 Hz |\n| GNSS receiver | Unicore UM980 | 10 Hz |\n\nRTK corrections provided by the German **SAPOS** service (centimeter-level accuracy under open sky).\n\n---\n\n## Sequences\n\n| Sequence | Duration | Length | RTK Fix | Checkpoints | Type |\n|---|---|---|---|---|---|\n| Stadtgarten Seq. 1 | 26 min 42 s | 1.04 km | 54% | 36 | Outdoor park |\n| Stadtgarten Seq. 2 | 14 min 36 s | 0.46 km | 40% | 19 | Outdoor park |\n| Construction Hall Seq. 1 | 12 min 21 s | 0.48 km | 25% | 16 | Outdoor + Indoor |\n| Construction Hall Seq. 2 | 9 min 59 s | 0.39 km | 23% | 16 | Outdoor + Indoor |\n\n**Stadtgarten**: Public park in Stuttgart with open-sky areas, tree cover, building obstruction, and a 30 m GNSS-denied underpass.\n\n**Construction Hall**: Large construction site (IntCDC, University of Stuttgart) with short outdoor segments and a GNSS-denied indoor hall spanning 400+ s of travel.\n\n---\n\n## Repository Structure\n\n```\nrtk-slam-dataset/\n├── ros1/                          # ROS1 (.bag) format\n│   ├── construction_seq1.bag\n│   ├── construction_seq2.bag\n│   ├── stadtgarten_seq1.bag\n│   └── stadtgarten_seq2.bag\n├── ros2/                          # ROS2 (.db3) format\n│   ├── construction_seq1/\n│   ├── construction_seq2/\n│   ├── stadtgarten_seq1/\n│   └── stadtgarten_seq2/\n└── euroc/                         # Extended EuRoC format (compatible with OKVIS2-X)\n    ├── construction_seq1_euroc.zip\n    ├── construction_seq2_euroc.zip\n    ├── stadtgarten_seq1_euroc.zip\n    └── stadtgarten_seq2_euroc.zip\n```\n\n### ROS Topics\n\n| Topic | Type | Rate |\n|---|---|---|\n| `/livox/lidar` | `livox_ros_driver/CustomMsg` | 10 Hz |\n| `/livox/points` | `sensor_msgs/PointCloud2` | 10 Hz |\n| `/livox/imu` | `sensor_msgs/Imu` | 200 Hz |\n| `/camera/image_raw/compressed` | `sensor_msgs/CompressedImage` | 20 Hz |\n| `/gnss/fix` | `sensor_msgs/NavSatFix` | 10 Hz |\n\n> **Note on camera timestamps:** The camera has a hardware trigger delay of −20.6 ms relative to the IMU clock (estimated by Kalibr). This offset is already compensated in all released formats — camera image timestamps in the ROS bags and EuRoC files reflect the corrected capture time. No additional time shift is needed when using this data.\n\n---\n\n## Privacy\n\nAll camera images were anonymized prior to release by blurring human faces using [deface](https://github.com/ORB-HD/deface).\n\n---\n\n## Citation\n\n```bibtex\n@article{zhang2025rtkslam,\n  title   = {An RTK-SLAM Dataset for Absolute Accuracy Evaluation in GNSS-Degraded Environments},\n  author  = {Zhang, Wei and Ress, Vincent and Skuddis, David and Soergel, Uwe and Haala, N","encoding":"utf8","key":"README.md","name":"README.md","size":4337,"truncated":true},{"bytes":"rosbag2_bagfile_information:\n  compression_format: ''\n  compression_mode: ''\n  custom_data: {}\n  duration:\n    nanoseconds: 741316797019\n  files:\n  - duration:\n      nanoseconds: 741316797019\n    message_count: 185014\n    path: construction_seq1.db3\n    starting_time:\n      nanoseconds_since_epoch: 1764589611286919116\n  message_count: 185014\n  relative_file_paths:\n  - construction_seq1.db3\n  ros_distro: rosbags\n  starting_time:\n    nanoseconds_since_epoch: 1764589611286919116\n  storage_identifier: sqlite3\n  topics_with_message_count:\n  - message_count: 14826\n    topic_metadata:\n      name: /camera/image_raw/compressed\n      offered_qos_profiles: ''\n      serialization_format: cdr\n      type: sensor_msgs/msg/CompressedImage\n      type_description_hash: RIHS01_15640771531571185e2efc8a100baf923961a4d15d5569652e6cb6691e8e371a\n  - message_count: 7413\n    topic_metadata:\n      name: /gnss/fix\n      offered_qos_profiles: ''\n      serialization_format: cdr\n      type: sensor_msgs/msg/NavSatFix\n      type_description_hash: RIHS01_62223ab3fe210a15976021da7afddc9e200dc9ec75231c1b6a557fc598a65404\n  - message_count: 148203\n    topic_metadata:\n      name: /livox/imu\n      offered_qos_profiles: ''\n      serialization_format: cdr\n      type: sensor_msgs/msg/Imu\n      type_description_hash: RIHS01_7d9a00ff131080897a5ec7e26e315954b8eae3353c3f995c55faf71574000b5b\n  - message_count: 7286\n    topic_metadata:\n      name: /livox/points\n      offered_qos_profiles: ''\n      serialization_format: cdr\n      type: sensor_msgs/msg/PointCloud2\n      type_description_hash: RIHS01_9198cabf7da3796ae6fe19c4cb3bdd3525492988c70522628af5daa124bae2b5\n  - message_count: 7286\n    topic_metadata:\n      name: /livox/lidar\n      offered_qos_profiles: ''\n      serialization_format: cdr\n      type: livox_ros_driver/msg/CustomMsg\n      type_description_hash: RIHS01_94041b4794f52c1d81def2989107fc898a62dacb7a39d5dbe80d4b55e538bf6d\n  version: 8\n","encoding":"utf8","key":"ros2/construction_seq1/metadata.yaml","name":"metadata.yaml","size":1930,"truncated":false},{"bytes":"rosbag2_bagfile_information:\n  compression_format: ''\n  compression_mode: ''\n  custom_data: {}\n  duration:\n    nanoseconds: 599509111166\n  files:\n  - duration:\n      nanoseconds: 599509111166\n    message_count: 149558\n    path: construction_seq2.db3\n    starting_time:\n      nanoseconds_since_epoch: 1764590367894611597\n  message_count: 149558\n  relative_file_paths:\n  - construction_seq2.db3\n  ros_distro: rosbags\n  starting_time:\n    nanoseconds_since_epoch: 1764590367894611597\n  storage_identifier: sqlite3\n  topics_with_message_count:\n  - message_count: 11990\n    topic_metadata:\n      name: /camera/image_raw/compressed\n      offered_qos_profiles: ''\n      serialization_format: cdr\n      type: sensor_msgs/msg/CompressedImage\n      type_description_hash: RIHS01_15640771531571185e2efc8a100baf923961a4d15d5569652e6cb6691e8e371a\n  - message_count: 5993\n    topic_metadata:\n      name: /gnss/fix\n      offered_qos_profiles: ''\n      serialization_format: cdr\n      type: sensor_msgs/msg/NavSatFix\n      type_description_hash: RIHS01_62223ab3fe210a15976021da7afddc9e200dc9ec75231c1b6a557fc598a65404\n  - message_count: 119819\n    topic_metadata:\n      name: /livox/imu\n      offered_qos_profiles: ''\n      serialization_format: cdr\n      type: sensor_msgs/msg/Imu\n      type_description_hash: RIHS01_7d9a00ff131080897a5ec7e26e315954b8eae3353c3f995c55faf71574000b5b\n  - message_count: 5878\n    topic_metadata:\n      name: /livox/points\n      offered_qos_profiles: ''\n      serialization_format: cdr\n      type: sensor_msgs/msg/PointCloud2\n      type_description_hash: RIHS01_9198cabf7da3796ae6fe19c4cb3bdd3525492988c70522628af5daa124bae2b5\n  - message_count: 5878\n    topic_metadata:\n      name: /livox/lidar\n      offered_qos_profiles: ''\n      serialization_format: cdr\n      type: livox_ros_driver/msg/CustomMsg\n      type_description_hash: RIHS01_94041b4794f52c1d81def2989107fc898a62dacb7a39d5dbe80d4b55e538bf6d\n  version: 8\n","encoding":"utf8","key":"ros2/construction_seq2/metadata.yaml","name":"metadata.yaml","size":1930,"truncated":false},{"bytes":"rosbag2_bagfile_information:\n  compression_format: ''\n  compression_mode: ''\n  custom_data: {}\n  duration:\n    nanoseconds: 1602929974556\n  files:\n  - duration:\n      nanoseconds: 1602929974556\n    message_count: 400136\n    path: stadtgarten_seq1.db3\n    starting_time:\n      nanoseconds_since_epoch: 1762267489872663021\n  message_count: 400136\n  relative_file_paths:\n  - stadtgarten_seq1.db3\n  ros_distro: rosbags\n  starting_time:\n    nanoseconds_since_epoch: 1762267489872663021\n  storage_identifier: sqlite3\n  topics_with_message_count:\n  - message_count: 32058\n    topic_metadata:\n      name: /camera/image_raw/compressed\n      offered_qos_profiles: ''\n      serialization_format: cdr\n      type: sensor_msgs/msg/CompressedImage\n      type_description_hash: RIHS01_15640771531571185e2efc8a100baf923961a4d15d5569652e6cb6691e8e371a\n  - message_count: 16028\n    topic_metadata:\n      name: /gnss/fix\n      offered_qos_profiles: ''\n      serialization_format: cdr\n      type: sensor_msgs/msg/NavSatFix\n      type_description_hash: RIHS01_62223ab3fe210a15976021da7afddc9e200dc9ec75231c1b6a557fc598a65404\n  - message_count: 320460\n    topic_metadata:\n      name: /livox/imu\n      offered_qos_profiles: ''\n      serialization_format: cdr\n      type: sensor_msgs/msg/Imu\n      type_description_hash: RIHS01_7d9a00ff131080897a5ec7e26e315954b8eae3353c3f995c55faf71574000b5b\n  - message_count: 15795\n    topic_metadata:\n      name: /livox/points\n      offered_qos_profiles: ''\n      serialization_format: cdr\n      type: sensor_msgs/msg/PointCloud2\n      type_description_hash: RIHS01_9198cabf7da3796ae6fe19c4cb3bdd3525492988c70522628af5daa124bae2b5\n  - message_count: 15795\n    topic_metadata:\n      name: /livox/lidar\n      offered_qos_profiles: ''\n      serialization_format: cdr\n      type: livox_ros_driver/msg/CustomMsg\n      type_description_hash: RIHS01_94041b4794f52c1d81def2989107fc898a62dacb7a39d5dbe80d4b55e538bf6d\n  version: 8\n","encoding":"utf8","key":"ros2/stadtgarten_seq1/metadata.yaml","name":"metadata.yaml","size":1933,"truncated":false}],"sourceObjectKey":"euroc/construction_seq1_euroc.zip","starterCell":"from huggingface_hub import hf_hub_download\npath = hf_hub_download(repo_id=\"Willyzw/rtk-slam-dataset\", filename=\"euroc/construction_seq1_euroc.zip\", repo_type=\"dataset\")\nprint(path)","starterSamplePath":"euroc/construction_seq1_euroc.zip","totalObjects":18},"summary":"A benchmark for the absolute (georeferenced) accuracy of SLAM and sensor fusion when satellite positioning is partly blocked, from the Institute for Photogrammetry and…","title":"RTK-SLAM Dataset: Handheld LiDAR-Visual-Inertial Recordings with RTK GNSS for Absolute Accuracy in GNSS-Degraded Environments (University of Stuttgart)","updatedAt":"2026-11-22T00:48:29.758Z","visibility":"public","websiteUrl":"https://rtk-slam-dataset.github.io/"}}