Home BlogDataset Teaching Autonomous Vehicles to Understand the World Around Them

Teaching Autonomous Vehicles to Understand the World Around Them

by Michael McLaughlin
Bounding boxes around vehicles.

Argo AI, an autonomous vehicle company, has released two high-definition maps and a dataset that contains information about the trajectories of 300,000 vehicles driving through Pittsburgh and Miami to support advancements in computer vision. The maps include geometric and semantic metadata, such as lane centerlines, lane direction, and the driveable area. Argo AI extracted the vehicle trajectories from over 1,000 driving hours, and the dataset includes left turns and lane changes. Researchers can use this data to teach autonomous vehicles to better anticipate the movements of other vehicles.

Get the data.

You may also like

Show Buttons
Hide Buttons