What’s Out There
← Back to swiping
TransportationFreeNational government

Next Generation Simulation (NGSIM) Vehicle Trajectories

U.S. Department of Transportation Intelligent Transportation Systems Joint Program Office

Second-by-second trajectories for every vehicle on stretches of US 101, I-80, and two arterials.

Detailed vehicle trajectory data from the NGSIM program, collected by cameras over southbound US 101 and Lankershim Boulevard in Los Angeles, eastbound I-80 in Emeryville, California, and Peachtree Street in Atlanta. Each row is a vehicle observed in a frame with position, lane, and preceding vehicle. The standard data set for car-following and lane-change research.

Access

Cost
FreeNo charge; Socrata app tokens are optional and only raise API rate limits.
License
Open licenseCreative Commons Attribution Share Alike 3.0 Unported
Sign-up
No sign-up
How you get it
API and Bulk downloadCSV, JSON

What's in it

One row is
One row per vehicle per time frame
Covers
US 101 and Lankershim Boulevard (Los Angeles), I-80 (Emeryville), Peachtree Street (Atlanta)
Years
Mid-2000s collections (record updated 2018)
Updated
One-time release
Size
Not stated
vehicle IDframe IDlocal Y positionglobal X and Ylane IDvehicle classpreceding vehicledirection

Quality

4 of 5

A benchmark for traffic modeling, with documentation attached, though it is old and short-duration.

  • Only a few hours per site
  • Known measurement noise in positions

Connects to

  • Highway Performance Monitoring System: Annual traffic counts for the same corridors
  • Waymo Open Dataset: Modern driving trajectories from autonomous vehicle sensors

Questions it could answer

  • How do headway and speed relate on congested US 101?
  • How frequent are lane changes on I-80?

Similar data sets

  • Argoverse

    Self-driving datasets pairing sensor data with detailed maps for tracking and forecasting.

    TransportationArgoverse (Argo AI research datasets)Free

  • Vehicle Miles of Travel by Functional System and State (VM-2)

    Annual miles driven in each state, split by road type, back to 1980.

    TransportationFederal Highway Administration (U.S. Department of Transportation)Free

  • INRIX Traffic and Trip Data

    Road speeds and trips pieced together from GPS pings of connected cars, fleets, and phones.

    TransportationINRIXPaid

  • Boreas Autonomous Driving Dataset

    A year of repeated drives with lidar, radar, and camera in sun, rain, and snow.

    TransportationAutonomous Space Robotics Lab, University of TorontoFree

  • PhysicalAI Autonomous Vehicles

    About 1,700 hours of multi-sensor driving data from diverse geographies for end-to-end driving research.

    TransportationNVIDIAFree

  • nuScenes

    1,000 annotated driving scenes from Boston and Singapore with a full self-driving sensor suite.

    TransportationMotionalFree