UGV, Quadruped
Socially CompliAnt Navigation Dataset (SCAND)
Socially compliant robot navigation
Human-teleoperated navigation demonstrations with lidar, cameras, inertial information, odometry, joystick commands and trajectory-level social interaction labels, collected on wheeled Jackal and legged Spot robots at UT Austin.
Access links
Primary places to inspect, download, read about, or reproduce the dataset.
Quick facts
- Vehicle type
- UGV, Quadruped
- Environment
- Indoor, Outdoor
- Data origin
- Real-world
- Sensor count
- 15
- Ground truth
- Not reported
- Calibration
- Parameters Only
- Annotations
- Available
Sensor overview
High-level sensor availability before the detailed sensor records below.
Ground truth
Reference scope, method, rate, coverage, and provenance.
Not reported
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
Sensor-to-body frame transforms are included for both robots. Intrinsic calibration, raw calibration recordings and synchronization accuracy are not established.
Known or intentional limitations
Reported constraints and characteristics to check before using the dataset.
Collection is limited to Austin, Texas, and may encode local conventions such as keeping right and overtaking on the left.
Some novel social interactions may be infrequent.
Human detection and tracking annotations are explicitly not provided.
Cameras
azure-kinect-rgb
- Model
- Azure Kinect
- Setup
- Mono
- Count
- 2
- Effective cameras
- 2
- Modality
- Rgb
- Resolution
- Not reported
- Rate
- 20 Hz
- Shutter
- Not Reported
- Lens type
- Not Reported
- HFOV
- Not reported
- VFOV
- Not reported
- Notes
- Front-facing RGB camera on each robot. Depth availability is unresolved; the comparison table says RGB-D, but the released-visual description specifies RGB imagery.
jackal-stereo
- Model
- Not reported
- Setup
- Stereo
- Count
- 1
- Effective cameras
- 2
- Modality
- Not Reported
- Resolution
- Not reported
- Rate
- 20 Hz
- Shutter
- Not Reported
- Lens type
- Not Reported
- HFOV
- Not reported
- VFOV
- Not reported
- Notes
- Front-facing stereo system on Jackal only; exact model, channels and image dimensions are not given.
spot-body-cameras
- Model
- Not reported
- Setup
- Mono
- Count
- 5
- Effective cameras
- 5
- Modality
- Not Reported
- Resolution
- Not reported
- Rate
- 5 Hz
- Shutter
- Not Reported
- Lens type
- Not Reported
- HFOV
- Not reported
- VFOV
- Not reported
- Notes
- Five body-mounted monocular cameras on Spot, publishing at 5 Hz; front left, front right, left, right and rear placement shown in Figure 3.
IMUs
Jackal 6D IMU
- Model
- Not reported
- Count
- 1
- Accel / gyro
- Not reported
Spot 6D IMU
- Model
- Not reported
- Count
- 1
- Accel / gyro
- Not reported
GNSS
Not reported
LiDAR
VLP-16 Velodyne
- Model or name
- VLP-16 Velodyne
- Count
- 2
- Dimensions
- 3
- Rate
- 10 Hz
- Channels
- Not reported
- Range
- Not reported
Radar
Not reported
Pitot tube
Not reported
Additional sensors
Odometry
- Name
- Jackal wheel odometry
- Count
- 1
- Rate
- 30 Hz
- Notes
- Jackal-only wheel odometry stream; physical wheel encoder details are not established.
Odometry
- Name
- Spot visual odometry
- Count
- 1
- Notes
- Body-frame visual odometry recorded using the Boston Dynamics APK as printed; it is an input estimate, with no external accuracy reference established.
Other sensor
- Name
- Spot leg joint angles
- Count
- 1
- Notes
- Recorded joint-angle information; the paper does not establish the physical measurement count or rates.
Annotations
Reported annotation availability and task support.
- Availability
- Available
- Format
- textual captions per trajectory
- Class count
- 12
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.
@article{karnan2022scand,
title={Socially CompliAnt Navigation Dataset (SCAND): A Large-Scale Dataset Of Demonstrations For Social Navigation},
author={Karnan, Haresh and Nair, Anirudh and Xiao, Xuesu and Warnell, Garrett and Pirk, S{\"o}ren and Toshev, Alexander and Hart, Justin and Biswas, Joydeep and Stone, Peter},
journal={arXiv preprint arXiv:2203.15041v2},
year={2022}
}
Dataset corrections
Notice a specific issue after reviewing the page?
Report an inaccuracy