List of Datasets

Handheld, UGV

Hilti SLAM Challenge 2023

multi-session SLAM; multi-device SLAM

Construction recordings from a handheld scanner and heavy drilling robot with sparse surveyed control-point evaluation.

Metadata: Complete Metadata source documented Released 2023

Primary places to inspect, download, read about, or reproduce the dataset.

Supplementary

Not reported

Quick facts

Vehicle type
Handheld, UGV
Environment
Indoor, Outdoor
Data origin
Real-world
Sensor count
17
Ground truth
Available
Calibration
Partial Raw Data
Annotations
Not reported

Sensor overview

High-level sensor availability before the detailed sensor records below.

Cameras

13 reported

IMUs

2 reported

GNSS

Not reported

LiDAR

2 reported

Radar

Not reported

Magnetometers

Not reported

Barometers

Not reported

UWB

Not reported

Ground truth

Reference scope, method, rate, coverage, and provenance.

Scope Method Reference system Count Rate Coverage Reference provenance
External
Measured using equipment independent of the platform’s normal onboard sensor pipeline.
Derived
Computed primarily from the dataset’s own recorded sensors.
Hybrid
Combines independent reference equipment with onboard measurements.
3D position Surveyed markers Phasma spike contact with TLS-surveyed GCP · Trimble X7 Not reported Not reported Partial · Handheld Recordings Reference provenance
External
Measured using equipment independent of the platform’s normal onboard sensor pipeline.
Derived
Computed primarily from the dataset’s own recorded sensors.
Hybrid
Combines independent reference equipment with onboard measurements.
External
Reference notes
Sparse handheld physical-contact position constraints from surveyed control points; not a dense 6-DoF trajectory.
3D position Surveyed markers Trailblazer LiDAR-observed surveyed GCP · Trimble X7 + Robosense Bpearl Not reported Not reported Partial · Tracked Robot Recordings Reference provenance
External
Measured using equipment independent of the platform’s normal onboard sensor pipeline.
Derived
Computed primarily from the dataset’s own recorded sensors.
Hybrid
Combines independent reference equipment with onboard measurements.
Hybrid
Accuracy notes
Marker detection residuals are reported in millimetres but exclude absolute sensor bias; they are not a full trajectory accuracy bound.
Reference notes
Surveyed external control-point coordinates are located using recorded LiDAR. Extrinsic uncertainty and sensor bias affect these sparse comparisons.

Calibration and synchronization

Reported calibration level, reproducibility signals, and supporting notes.

Level
Partial Raw Data
Processed parameters
Available
Raw calibration data
Available
Calibration targets
Not reported

Reproducible calibration data

Raw calibration data is reported, so users can reproduce or inspect the calibration process.

Notes

Official archive lists parameter files, one handheld calibration sequence, one robot calibration sequence and a calibration-board YAML. Complete raw sessions for every calibration are unverified.

Known or intentional limitations

Reported constraints and characteristics to check before using the dataset.

Fifteen recordings comprise twelve primary challenge sequences and three additional handheld sequences. Handheld and tracked-robot sensor configurations are alternatives.

Reference files contain sparse 3D surveyed control-point constraints, not dense six-degree-of-freedom trajectories. Robot control-point extraction uses the recorded LiDAR and is only partly independent.

Maintained official documentation supplies handheld sensor details and calibration recordings for both platforms. Timing figures are subsystem-specific and are not unconditional bounds for every measurement.

Published marker-localization residuals exclude absolute sensor bias and do not establish universal trajectory accuracy.

Cameras

Camera 1

Model
Luxonis OAK-D
Setup
Stereo
Count
4
Effective cameras
8
Modality
Other
Resolution
Not reported
Rate
Not reported
Shutter
Not Reported
Lens type
Not Reported
HFOV
Not reported
VFOV
Not reported
Notes
Tracked robot only: four OAK-D stereo systems. Separate RGB/depth delivery and common native cadence remain unestablished.

Camera 2

Model
Sevensense Alphasense Core
Setup
Mono
Count
5
Effective cameras
5
Modality
Other
Resolution
Not reported
Rate
Not reported
Shutter
Global
Lens type
Not Reported
HFOV
Not reported
VFOV
Not reported
Notes
Handheld alternative only: five 0.4 MP global-shutter cameras. Exact released dimensions and cadence remain unknown.

IMUs

XSens MTi-670

Model
XSens MTi-670
Count
1
Accel / gyro
Not reported

Handheld associated inertial sensing

Model
Not reported
Count
1
Accel / gyro
Not reported

GNSS

Not reported

LiDAR

Robosense Bpearl

Model or name
Robosense Bpearl
Count
1
Dimensions
3
Rate
Not reported
Channels
32
Range
Not reported

Hesai PandarXT-32

Model or name
Hesai PandarXT-32
Count
1
Dimensions
3
Rate
Not reported
Channels
Not reported
Range
Not reported

Radar

Not reported

Additional sensors

Not reported

Annotations

Reported annotation availability and task support.

Availability
Not reported
Format
Not reported
Class count
Not reported
Semantic segmentation Not reported
Instance segmentation Not reported
Object detection Not reported
Optical flow Not reported
Depth ground truth Not reported

Citation

Citation key and BibTeX kept at the end of the page for reference.

Citation key
nair2024hiltislam2023

BibTeX

@misc{nair2024hiltislam2023,
  author = {Ashish Devadas Nair and Julien Kindle and Plamen Levchev and Davide Scaramuzza},
  title = {Hilti SLAM Challenge 2023: Benchmarking Single + Multi-session SLAM across Sensor Constellations in Construction},
  year = {2024},
  url = {https://arxiv.org/abs/2404.09765v2}
}

Dataset corrections

Notice a specific issue after reviewing the page?

Report an inaccuracy