UGV
ROOAD: RELLIS Off-road Odometry Analysis Dataset
Visual-Inertial Odometry
Six off-road monocular visual-inertial traversals on a Warthog at RELLIS, with RTK position and heading reference and eleven additional camera-IMU calibration sequences.
Access links
Primary places to inspect, download, read about, or reproduce the dataset.
Quick facts
- Vehicle type
- UGV
- Environment
- Outdoor
- Data origin
- Real-world
- Sensor count
- 2
- Ground truth
- Available
- Calibration
- Partial Raw Data
- 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 + heading | RTK GNSS | Ardusimple Differential Heading with RTK kit · simpleRTK2B and simpleRTK2Blite boards with ZED-F9P modules | 3 receiver boards | 10 Hz | Partial · Rt4 Gravel, Rt4 Rim, Rt4 Updown, Rt5 Gravel, Rt5 Rim, Rt5 Updown |
Reference provenance
External
|
|
||||||
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
Eleven camera-IMU recordings vary IMU-x translation or IMU-z rotation. Kalibr estimates camera intrinsics/distortion and spatial/time camera-IMU transforms. Six hours of stationary IMU data are used for Allan analysis, but their release is not established. Manual rail/stage comparison is calibration validation, not platform ground-truth accuracy. Detailed depth results report 47 mm mean error and 10 mm standard deviation, whereas the conclusion says at worst +/-10 mm; no overall accuracy guarantee is proposed.
Known or intentional limitations
Reported constraints and characteristics to check before using the dataset.
Reference supplies position and heading but lacks pitch and roll.
Stereo and LiDAR are future extensions, not released streams documented by this edition.
Cameras
monocular
- Model
- Basler acA1920-50gc
- Setup
- Mono
- Count
- 1
- Effective cameras
- 1
- Modality
- Grayscale
- Resolution
- 1920 x 1200
- Rate
- 30 Hz
- Shutter
- Not Reported
- Lens type
- Not Reported
- HFOV
- Not reported
- VFOV
- Not reported
- Notes
- Printed model suffix -50gc retained; the paper describes B/W acquisition and grayscale output. Lens is 8 mm/F1.8 Edmund Optics; exposure is 1 ms. Pinhole and radial-tangential calibration models are used.
IMUs
Vectornav VN-300 INS
- Model
- Vectornav VN-300 INS
- Count
- 1
- Accel / gyro
- 400 / 400 Hz
GNSS
Not reported
LiDAR
Not reported
Radar
Not reported
Pitot tube
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.
@misc{chustz2022rooad,
title = {ROOAD: RELLIS Off-road Odometry Analysis Dataset},
author = {George Chustz and Srikanth Saripalli},
year = {2022},
eprint = {2109.08228v2},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2109.08228v2}
}
Dataset corrections
Notice a specific issue after reviewing the page?
Report an inaccuracy