United States Department of Transportation
i
A Framework for Validating Traffic Simulation Models at the Vehicle Trajectory Level
-
2017-03-01
-
[PDF-4.20 MB]
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Edition:Final Report
-
Contracting Officer:
-
Corporate Publisher:
-
NTL Classification:NTL-INTELLIGENT TRANSPORTATION SYSTEMS-INTELLIGENT TRANSPORTATION SYSTEMS
-
Abstract:Based on current practices, traffic simulation models are calibrated and validated using macroscopic measures such as 15-minute averages of traffic counts or average point-to-point travel times. For an emerging number of applications, including connected vehicles, the realism of simulated driver dynamics at the second-by-second or sub-second trajectory level plays an important role. A framework to validate the realism of simulated vehicle dynamics at the trajectory level is presented in this report. Trajectory measures related to safety, comfort, vehicle kinematics, and traffic flow are presented. Example validation measures include time to collision, lane change urgency and rate, acceleration range, jerk, and root mean square of acceleration. Insights on the distribution and characteristics of each of these measures have been assembled from naturalistic driving studies, the American Association of State Highway and Transportation Officials (AASHTO), the International Organization for Standardization (ISO), the Federal Motor Carrier Safety Administration (FMCSA), the National Highway Traffic Safety Administration (NHTSA), and the analysis of trajectory data collected in this project. Practitioners can use the validation framework to assess the realism of the simulated vehicle dynamics in a model. Realistic vehicle dynamics at the sub-second level are required in the modeling of many emerging and technologically advanced applications that involve autonomous vehicles or vehicle-to-vehicle communications. In more traditional modeling applications, realistic vehicle dynamics result in realistic traffic properties at the aggregate level and the ability to reliably emulate a wide range of traffic phenomena. The proposed framework can be used by practitioners, researchers and software developers to document and improve the capabilities of traffic microsimulation models with respect to vehicle dynamics.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type:
