Next Generation Simulation (NGSIM) Open Data

ITS DataHub has partnered with the Federal Highway Administration's (FHWA's) Next Generation SIMulation (NGSIM) program to make available detailed vehicle trajectory data and supporting data files along with the raw and processed video files from the NGSIM data collection efforts. Researchers for the NGSIM program collected the specified data on southbound US 101 and Lankershim Boulevard in Los Angeles, CA, eastbound I-80 in Emeryville, CA and Peachtree Street in Atlanta, GA.
This article provides a brief overview of the NGSIM program data collection as well as what types of data are available on ITS DataHub. Some examples of possible uses for the data and information on how to cite the various NGSIM datasets are also included.

NGSIM Program Data Collection Overview

The NGSIM program collected high-quality traffic datasets at four different locations, including two freeway segments (I-80 and US-101) and two arterial segments (Lankershim Boulevard and Peachtree Street), between 2005 and 2006 to support the goals of the FHWA Traffic Analysis Tools Program. Specifically, the subsequent data were collected as part of an effort to develop microscopic behavioral algorithms for modeling and simulation with supporting documentation and validation data that addressed existing traffic data gaps at the time (e.g., the need for detailed trajectory and supporting data to accurately represent multimodal traveler interactions, vehicle and freeway systems, and driver behavior when encountering traffic control devices, congestion, and other environmental features).
The datasets collected and generated for each location include the vehicle trajectory data (primary data), other location-specific primary and support data (e.g., ortho-rectified photographs of the study area, Computer-Aided Design (CAD) drawings of the study area, signal timings, weather data, detector data), raw video files, and processed video files.
In addition to the behavioral algorithms and corresponding reports developed and provided by the NGSIM program using the collected data, the simulation community continues to utilize the NGSIM data to develop, calibrate, and/or validate countless other algorithms and models. 

NGSIM Vehicle Trajectories and Supporting Data

Researchers for the Next Generation Simulation (NGSIM) program collected detailed vehicle trajectory data on the above specified freeway segments of US-101 and I-80 as well as the specified arterial segments of Lankershim Boulevard and Peachtree Street. Data were collected through a network of synchronized digital video cameras. NGVIDEO, a customized software application developed for the NGSIM program, transcribed the vehicle trajectory data from the videos. These vehicle trajectory data provide the precise location of each vehicle within the study area every one-tenth of a second, resulting in detailed lane positions and locations relative to other vehicles. 
For each NGSIM data collection location, location-specified metadata documentation, select video data, and vehicle trajectory data with supporting data (e.g., weather data, CAD files, detector data) are available for download. The algorithms and reports developed by the NGSIM program using the collected data can be downloaded as well. To download the metadata, vehicle trajectory with supporting data, and/or the algorithms and reports, please see the following: Next Generation Simulation (NGSIM) Vehicle Trajectories and Supporting Data.
Please note that the above mentioned data and files are available for download in the "About this Dataset" > "Attachments" section in the provided link above.
Fact sheets are also available for the NGSIM datasets here under "Additional Resources".

Trajectory Data

The trajectory data are considered primary data. There are 25 columns and over 11.8 million rows of trajectory data for the four NGSIM data collection locations. The data can be browsed in the embedded data explorer below or in a new tab.

Other Primary and Support Data

For each of the four NGSIM data collection locations, other relevant location-specific data are included in the respective location .zip files along with the vehicle trajectory data. All of the data collection locations have the following primary and support data:
  • Detector or tube data that provide volume and occupancy (and possibly speed) values at various aggregation levels (e.g., 30-second, 5-minute, or 15-minute) (primary data)
  • Ortho-rectified photographs of the study area (support data)
  • Computer-Aided Design (CAD) drawings of the study area, developed using the ortho-rectified photographs (support data)
  • Signal and/or ramp meter timings (support data)
  • Geographic Information System (GIS) shapefiles that include geometry and other attributes for the study area network (support data)
  • Data analysis files with aggregated vehicle trajectory results that provide common macroscopic traffic flow parameters (support data)
The I-80 NGSIM location has the following additional data provided:
  • Weather data from the San Francisco Airport for the data collection time period (support data)
  • Signage photographs (support data)

NGSIM Video Files

The raw video files and processed video files are available to download for each of the four NGSIM data collection locations. These videos are available in 15-minute segments for each camera used at the site. Either 30 minutes or 45 minutes of total video are provided, depending on the site. The raw videos show the original vehicle movement data and offer users a view of how the section was observed. The processed video files provide videos of the vehicles along with a superimposition of vehicle identification numbers. These videos can be used as stand-alone or they can be used for cross referencing textual vehicle trajectory data provided in the NGSIM trajectory data.
Please note that the raw and processed video files are available for download in the "About this Dataset" > "Attachments" section for each respective location in the links below.

I-80 Videos

Videos were collected on a segment of I-80 located in Emeryville, California on April 13, 2005. 45 minutes of video data are available, segmented into three 15-minute periods: 1) 4:00 p.m. to 4:15 p.m.; 2) 5:00 p.m. to 5:15 p.m.; and 3) 5:15 p.m. to 5:30 p.m. The dataset includes both raw and processed video files from each of the seven cameras for the three time periods available for download. Camera numbering is in order of southern-most (1) to northern-most (7). 
A preview of the raw and processed camera 1 video files from 4:00 p.m. to 4:15 p.m. are shown below on the left and right, respectively. The full video files for all time periods and cameras for I-80 and the corresponding metadata can be browsed and downloaded in a new tab

US-101 Videos

Videos were collected on a freeway segment of US 101 (Hollywood Freeway) located in Los Angeles, California on June 15, 2005. 45 minutes of transcribed data are included in this full data set, segmented into three 15-minute periods representing: 1) 7:50 a.m. to 8:05 a.m., 2) 8:05 a.m. to 8:20 a.m., and 3) 8:20 a.m. to 8:35 a.m. on June 15, 2005. The dataset includes files for both raw and processed video files from each of the eight cameras for the three time periods available for download. Camera numbering is in order of southern-most (1) to northern-most (8). 
A preview of the raw and processed camera 1 video files from 7:50 a.m. to 8:05 a.m. are shown below on the left and right, respectively. The full video files for all time periods and cameras for US 101 and the corresponding metadata can be browsed and downloaded in a new tab

Lankershim Boulevard Videos

Videos were collected on June 16, 2005 on an arterial segment on Lankershim Boulevard located in Los Angeles, California. The data represent 30 minutes of video, segmented into two periods (8:30 a.m. to 8:45 a.m. and 8:45 a.m. to 9:00 a.m.). The dataset includes files for both raw and processed video data from each of the five cameras for the two time periods available for download. Camera numbering is in order of southern-most (1) to northern-most (5). 
A preview of the raw and processed camera 1 video files from 8:30 a.m. to 8:45 a.m. are shown below on the left and right, respectively. The full video files for all time periods and cameras for Lankershim Boulevard and the corresponding metadata can be browsed and downloaded in a new tab

Peachtree Street Videos

Videos were collected on November 8, 2006 on an arterial segment on Peachtree Street located in Atlanta, Georgia. The data represent 30 minutes of video, segmented into two periods (12:45 p.m. to 1:00 p.m. and 4:00 p.m. to 4:15 p.m.). The dataset includes files for both raw and processed video data from each of the eight cameras for the two time periods available for download. Camera numbering is in order of southern-most (1) to northern-most (8).
A preview of the raw and processed camera 1 video files from 12:45 p.m. to 1:00 p.m. are shown below on the left and right, respectively. The full video files for all time periods and cameras for Peachtree Street and the corresponding metadata can be browsed and downloaded in a new tab

Example Uses for the NGSIM Data

As of March 2022, there are over 4,300 Google Scholar Search Results for "NGSIM" and with over 1,500 of those for "NGSIM data". To help potential users start to understand the various uses of the NGSIM data, several open-access research publications were compiled and are listed below. These publications were included in the list based on the high number of citations they have, as well as being open access publications. Please note that the research efforts briefly described in the following list are included strictly for the purposes of illustrating uses of the data in research projects and the U.S. Department of Transportation (U.S. DOT) neither validates the research nor endorses the institutions that conducted the research and their findings.
References
(1) Thiemann, C., Treiber, M., & Kesting, A. (2008). Estimating acceleration and lane-changing dynamics from next generation simulation trajectory data. Transportation Research Record2088(1), 90-101.
(2) Altché, F., & de La Fortelle, A. (2017, October). An LSTM network for highway trajectory prediction. In 2017 IEEE 20th international conference on intelligent transportation systems (ITSC) (pp. 353-359). IEEE.
(3) Herrera, J. C., & Bayen, A. M. (2010). Incorporation of Lagrangian measurements in freeway traffic state estimation. Transportation Research Part B: Methodological44(4), 460-481.
(4) Talebpour, A., Mahmassani, H. S., & Hamdar, S. H. (2015). Modeling lane-changing behavior in a connected environment: A game theory approach. Transportation Research Procedia7, 420-440.
(5) Jiang, F., Yuan, J., Tsaftaris, S. A., & Katsaggelos, A. K. (2011). Anomalous video event detection using spatiotemporal context. Computer Vision and Image Understanding115(3), 323-333.
(6) Schwarting, W., Pierson, A., Alonso-Mora, J., Karaman, S., & Rus, D. (2019). Social behavior for autonomous vehicles. Proceedings of the National Academy of Sciences116(50), 24972-24978.
(7) Kim, J., & Mahmassani, H. S. (2011). Correlated parameters in driving behavior models: Car-following example and implications for traffic microsimulation. Transportation Research Record2249(1), 62-77.

Citing the NGSIM Data

When you use any of the NGSIM trajectory data and supporting data files or the raw and/or processed video files, please ensure to include the appropriate citation. In the below recommended citations, replace YYYY-MM-DD with the date you access the data.
  • For the Next Generation Simulation (NGSIM) Vehicle Trajectories and Supporting Data, the recommended citation is: "U.S. Department of Transportation Federal Highway Administration. (2016). Next Generation Simulation (NGSIM) Vehicle Trajectories and Supporting Data. [Dataset]. Provided by ITS DataHub through Data.transportation.gov. Accessed YYYY-MM-DD from http://doi.org/10.21949/1504477"
  • For the Next Generation Simulation (NGSIM) Program I-80 Videos, the recommended citation is:  "U.S. Department of Transportation Federal Highway Administration. (2016). Next Generation Simulation (NGSIM) Program I-80 Videos. [Dataset]. Provided by ITS DataHub through Data.transportation.gov. Accessed YYYY-MM-DD from http://doi.org/10.21949/1504477"
  • For the Next Generation Simulation (NGSIM) Program US-101 Videos, the recommended citation is: "U.S. Department of Transportation Federal Highway Administration. (2016). Next Generation Simulation (NGSIM) Program US-101 Videos. [Dataset]. Provided by ITS DataHub through Data.transportation.gov. Accessed YYYY-MM-DD from http://doi.org/10.21949/1504477"
  • For the Next Generation Simulation (NGSIM) Program Lankershim Boulevard Videos, the recommended citation is: "U.S. Department of Transportation Federal Highway Administration. (2016). Next Generation Simulation (NGSIM) Program Lankershim Boulevard Videos. [Dataset]. Provided by ITS DataHub through Data.transportation.gov. Accessed YYYY-MM-DD from http://doi.org/10.21949/1504477"
  • For the Next Generation Simulation (NGSIM) Program Peachtree Street Videos, the recommended citation is: "U.S. Department of Transportation Federal Highway Administration. (2016). Next Generation Simulation (NGSIM) Program Peachtree Street Videos. [Dataset]. Provided by ITS DataHub through Data.transportation.gov. Accessed YYYY-MM-DD from http://doi.org/10.21949/1504477"
The ITS JPO Data Program support team would greatly appreciate it if you would let us know if you use the NGSIM data in any publications. Please contact us at: data.itsjpo@dot.gov.