List of Datasets

Other, Automotive

NTU4DRadLM

Radar/thermal/inertial localization and mapping

Six handcart and vehicle campus recordings pair 4D radar and thermal images with visual/LiDAR/inertial data and derived SLAM reference trajectories.

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
Other, Automotive
Environment
Outdoor
Data origin
Real-world
Sensor count
7
Ground truth
Available
Calibration
Parameters Only
Annotations
Not reported

Sensor overview

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

Cameras

2 reported

IMUs

2 reported

GNSS

1 reported

LiDAR

1 reported

Radar

1 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.
6DoF LiDAR SLAM R2LIVE + g2o reference Not reported Not reported Unknown 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.
Unknown
Reference notes
R2LIVE uses recorded visual/LiDAR/IMU inputs and g2o loop correction. The paper also mentions RTK assistance, but GNSS is present only for vehicle loops; complete independence remains unestablished.

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

Circular-hole and heated spherical-trihedral targets; LiDAR/IMU temporal offset calibration does not establish release-wide multi-sensor clock synchronization.

Known or intentional limitations

Reported constraints and characteristics to check before using the dataset.

Six routes use handcart and vehicle configurations; RTK GNSS is supplied only for the three vehicle loops.

R2LIVE plus g2o produces an input-derived reference. RTK involvement is described inconsistently, so independence and complete coverage remain unknown.

The reported 85 minutes and sum of scan-table durations differ; no aggregate duration is chosen. Raw calibration sessions and common timing uncertainty remain unknown.

The radar supplies Doppler point returns; it does not supply independent platform pose truth. Native component IMU rates are distinct from published topic rates.

Cameras

Camera 1

Model
Vishinsgae SY020HD
Setup
Mono
Count
1
Effective cameras
1
Modality
Rgb
Resolution
640 x 480
Rate
30 Hz
Shutter
Not Reported
Lens type
Not Reported
HFOV
Not reported
VFOV
Not reported

Camera 2

Model
iRay Micro III 640T
Setup
Mono
Count
1
Effective cameras
1
Modality
Thermal
Resolution
640 x 512
Rate
25 Hz
Shutter
Not Reported
Lens type
Not Reported
HFOV
Not reported
VFOV
Not reported

IMUs

Standalone VN-100 IMU

Model
Vector VN-100
Count
1
Accel / gyro
Not reported

Livox Horizon integrated IMU

Model
Not reported
Count
1
Accel / gyro
Not reported

GNSS

u-blox F9P-02B-00

Model or name
u-blox F9P-02B-00
Count
1
Rate
20 Hz
GNSS type
RTK
Position output
Available
Velocity output
Not reported
Heading output
Not reported

LiDAR

Livox Horizon

Model or name
Livox Horizon
Count
1
Dimensions
3
Rate
10 Hz
Channels
Not reported
Range
Not reported

Radar

Oculii Eagle

Count
1
Released data
Point detections / point clouds
Frame rate
12 Hz
Doppler velocity
Available
Elevation angle
Available
Configured range
Not reported
Field of view
120° H × 30° V (per radar)
Notes
Point returns include xyz, Doppler and power; 4D describes radar output, not time plus a pose truth channel.

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
zhang2023ntu4dradlm

BibTeX

@misc{zhang2023ntu4dradlm,
  author = {Jun Zhang and Huayang Zhuge and Yiyao Liu and Guohao Peng and Zhenyu Wu and Haoyuan Zhang and Qiyang Lyu and Heshan Li and Chunyang Zhao and Dogan Kircali and Sanat Mharolkar and Xun Yang and Su Yi and Yuanzhe Wang and Danwei Wang},
  title = {NTU4DRadLM: 4D Radar-centric Multi-Modal Dataset for Localization and Mapping},
  year = {2023},
  url = {https://arxiv.org/abs/2309.00962v1}
}

Dataset corrections

Notice a specific issue after reviewing the page?

Report an inaccuracy