ARPA-I INSIGHTS Dataset

Description

The ARPA-I INSIGHTS Dataset is a large-scale, high-density (~85 pts/m²) geiger-mode aerial LiDAR dataset covering over 1,600 km² across the Salt Lake City, UT and Denver, CO metropolitan regions, including major freeway corridors. Collected in June 2025, the dataset provides dense three-dimensional measurements with approximately 30 cm vertical and horizontal point spacing, enabling detailed representation of transportation infrastructure and surrounding urban environments. The data was acquired as part of a U.S. Department of Transportation ARPA-I-funded project, executed by MIT Lincoln Laboratory, to support state and regional transportation agencies. The dataset is designed to enable transportation digital twins, supporting virtual inspection, measurement, and inventory of infrastructure assets at corridor and regional scales. In addition to raw point clouds, the dataset includes annotated subsets for machine learning, with labeled classes for surface types (e.g., road, sidewalk, trail, driveway) and infrastructure elements (e.g., power lines, traffic signals, light poles). The data is optimized for cloud-based access through a STAC catalog with GeoParquet index, and tiled storage of cloud-optimized point clouds, supporting scalable workflows for geospatial analysis, infrastructure monitoring, and multimodal AI research.

Update Frequency

Ad Hoc

License

CC-BY-4.0

Documentation

https://github.com/VolpeUSDOT/ARPA-I_INSIGHTS_Documentation

Managed By

MIT Lincoln Laboratory

See all datasets managed by MIT Lincoln Laboratory.

Contact

arpa-i-insights-admin@ll.mit.edu

How to Cite

ARPA-I INSIGHTS Dataset was accessed on DATE from https://registry.opendata.aws/arpa-i-insights.

Usage Examples

Tutorials

• Get to know the ARPA-I INSIGHTS Dataset by Jeffrey Liu