Safe-D: Safety through Disruption

Characterizing Level 2 Automation in a Naturalistic Driving Fleet


Abstract

In the ongoing NOVA fleet data collection effort, vehicles with Society of Automotive Engineers (SAE) Level 2 (L2) features will be instrumented with sophisticated data acquisition systems that are able to collect multiple video views, sensor data, and vehicle network data. These data acquisition systems are the most sophisticated systems available at VTTI and provide the capability to easily add or remove various sensors from the fleet. The instrumented vehicles will mostly be drivers’ personal vehicles. The effort to collect these data will be funded from other sources and will be used as cost share for this project. That is, without the collection of these data, this analysis project would not be possible. The data collection is expected to occur in the Northern Virginia area.

For this Safe-D project, dash video from the NOVA fleet collection effort will be analyzed using machine vision to, combined with additional approaches that offer some redundancy, determine the frequency, timing, and characteristics of L2 feature activations and deactivations. These data will be used to address research questions associated with L2 automation feature usage. These questions relate to what features are activated, when they are activated, where they are activated, when they are deactivated, and what leads to system takeover requests. Takeover requests in particular will be used to examine the effectiveness of L2 systems in handling diverse roadway features and environmental conditions. The combination of these efforts will result in the creation of a framework to characterize the real-world operational domain of L2 features using naturalistic driving data. A de-identified dataset that includes summary data for the activations and deactivations used in the proposed analyses will also be generated as a product of this research effort. The project will provide funding for two master’s student for two years; the students will use the project as the topic of their graduate theses.

Project Highlights

Coming Soon!

Final Report

Coming Soon!

EWD & T2 Products

Coming Soon!

Presentations/Publications

Coming Soon!

Research Investigators (PI*)

Miguel Perez (VTTI)*
Jon Hankey (VTTI)
Nick Britten (Student-VT)

Project Information

Start Date: 2019-08-10
End Date: 2020-08-09
Status: Active
Grant Number: 69A3551747115
Total Funding: $869,677
Source Organization: Safe-D National UTC
Project Number: VTTI-00-024

Safe-D Theme Areas

Big Data Analytics
Automated Vehicles

Safe-D Application Areas

Driver Factors and Interfaces
Vehicle Technology

More Information

RiP URL
UTC Project Information Form

Sponsor Organization

Office of the Assistant Secretary for Research and Technology
University Transportation Centers Program
Department of Transportation
Washington, DC 20590 United States

Performing Organization

Virginia Polytechnic Institute and State University
Virginia Tech Transportation Institute
3500 Transportation Research Plaza
Blacksburg, Virginia 24061
USA