Handheld, UGV
Hilti SLAM Challenge
Multi-sensor SLAM in construction and building environments
Original 2021 twelve-sequence camera, dual-lidar and IMU collection with sparse total-station position and lab/arena motion-capture references.
Access links
Primary places to inspect, download, read about, or reproduce the dataset.
Quick facts
- Vehicle type
- Handheld, UGV
- Environment
- Indoor, Outdoor
- Data origin
- Real-world
- Sensor count
- 10
- Ground truth
- Available
- Calibration
- Parameters Only
- Annotations
- Not reported
Sensor overview
High-level sensor availability before the detailed sensor records below.
Ground truth
Reference scope, method, rate, coverage, and provenance.
| Scope | Method | Reference system | Count | Rate | Coverage |
Reference provenance
|
|---|---|---|---|---|---|---|
| 3D position | Total station | Hilti total station · PLT 300 | Not reported | Not reported | Partial · Stationary Total-Station Samples in Ten Recordings Outside the Two Motion-Capture Areas; Current Release Links Their 3DoF Truth Files. |
Reference provenance
External
|
|
||||||
| 6DoF | Motion capture | Optical motion capture | Not reported | 200 Hz | Partial · Lab and RPG Drone Testing Arena; Current Release Links Their 6DoF Truth Files. |
Reference provenance
External
|
|
||||||
Calibration and synchronization
Reported calibration level, reproducibility signals, and supporting notes.
- Level
- Parameters Only
- Processed parameters
- Available
- Raw calibration data
- Not reported
- Calibration targets
- Not reported
Parameters only
Processed calibration parameters are reported, but raw calibration data is not reported.
Notes
Manufacturer sensor intrinsics, camera checkerboard calibration, CAD extrinsics and refined camera/IMU calibration; parameters/CAD/noise files released.
Known or intentional limitations
Reported constraints and characteristics to check before using the dataset.
Total-station truth is sparse and position-only.
Motion-capture/logger clock difference is 1–3 ms.
Frames can drop under controller load.
Historical challenge references were partly withheld; the current 2021 page links truth files for all twelve recordings. Total-station coverage remains sparse, rather than continuous six-DoF truth.
Reference coordinates may be in the prism frame; use the supplied sensor transforms before comparing a body-frame trajectory.
This entry describes the 2021 collection, excluding later Hilti/Trimble challenge datasets.
Cameras
Camera 1
- Model
- Sevensense AlphaSense
- Setup
- Multi Camera
- Count
- 5
- Effective cameras
- 5
- Modality
- Not Reported
- Resolution
- Not reported
- Rate
- 10 Hz
- Shutter
- Global
- Lens type
- Not Reported
- HFOV
- Not reported
- VFOV
- Not reported
- Notes
- Five physical image channels include one stereo pair; represented once as a five-camera array. Approximately 1.3 MP per camera; exact pixel dimensions and color modality are unestablished.
IMUs
Analog Devices ADIS16445
- Model
- Analog Devices ADIS16445
- Count
- 1
- Accel / gyro
- 800 / 800 Hz
Bosch BMI085
- Model
- Bosch BMI085
- Count
- 1
- Accel / gyro
- 200 / 200 Hz
InvenSense ICM-20948
- Model
- InvenSense ICM-20948
- Count
- 1
- Accel / gyro
- 100 / 100 Hz
GNSS
LiDAR
Ouster OS0-64
- Model or name
- Ouster OS0-64
- Count
- 1
- Dimensions
- 3
- Rate
- 10 Hz
- Channels
- 64
- Range
- 50 m
Livox MID70
- Model or name
- Livox MID70
- Count
- 1
- Dimensions
- 3
- Rate
- Not reported
- Channels
- Not reported
- Range
- 200 m
Radar
Additional sensors
Annotations
Reported annotation availability and task support.
- Availability
- Not reported
- Format
- Not reported
- Class count
- Not reported
Citation
Citation key and BibTeX kept at the end of the page for reference.
@misc{helmberger2022hilti,
title = {The Hilti SLAM Challenge Dataset},
author = {Michael Helmberger and Kristian Morin and Beda Berner and Nitish Kumar and Giovanni Cioffi and Davide Scaramuzza},
year = {2022},
eprint = {2109.11316v3},
archivePrefix = {arXiv}
}
Dataset corrections
Notice a specific issue after reviewing the page?
Report an inaccuracy