This project is the result of a $7.5M Federal Highway Administration Automated Driving System Demonstration Grant #693JJ32040003 awarded to the Virginia TechTransportation Institute (VTTI) in 2020. The primary objective of this research project is to demonstrate how Automated Driving System (ADS)-equipped vehicles can interact safely in challenging dynamic scenarios, such as encounters with public safety providers and other public services. These demonstrations are the result of significant collaboration between automotive Original Equipment Manufacturers (OEMs), public safety service providers, Infrastructure Owner Operators (IOOs), and academic ADS developers to identify challenging scenarios, identify technological solutions, and conduct demonstrations for stakeholders to support safety analysis, development of industry standards, and possible rulemaking. This project identified and defined key dynamic scenarios, developed technological concepts for safe interactions of ADS-equipped vehicles in these scenarios, and demonstrated the solutions live on the I-395/I-95 Express Lanes corridor operated by Transurban and the Virginia Smart Roads operated by VTTI in Blacksburg, VA. The I-395 Express Lanes corridor was optimized for automation through the integration of applications that exchange data with Transurban’s Traffic Operations Center (TOC). The research was led by VTTI and the demonstration team included representatives of OEMs [through Crash Avoidance Metrics Partners (CAMP), LLC], IOOs [the Virginia Department of Transportation (VDOT) and Transurban], Center for Automotive Performance Simulation (GCAPS)].

The team defined six tasks to accomplish the goals and objectives for the project. Task 1 focused on optimizing the infrastructure capabilities on the I-395/I-95 Express Lanes for demonstrating safe ADS operation in dynamic scenarios. In Task 2, we identified and defined the scenarios, generated ADS-equipped vehicle interaction requirements, and finalized the candidate technological concepts using a structured process with significant stakeholder involvement. The results of Task 2 then informed the design of the SAE Level 4 (L4) ADS-equipped reference platform vehicle to be built and tested in Task 3. The reference platform provided the necessary capabilities and unique features to collaboratively demonstrate safe ADS operation in challenging scenarios involving public services. In Task 4, the team developed the TOC-based data services and cooperative automation applications necessary to provide real-time operational design domain information and suggested operating parameters (speed, headway, lane selection, etc.) to the ADS to support safe operation in the selected scenarios. Within Task 5, we conducted high-profile demonstrations on both the I-395 Express Lanes in Northern Virginia and the Virginia Smart Roads in Blacksburg, VA. Each demonstration spanned several days, with participants representing specific groups of key ADS stakeholders necessary for successful deployment. Demonstrations showcased the L4 ADS-equipped reference vehicle operating safely through a combination of staged and naturally occurring dynamic scenarios while collecting a wide variety of high-definition data. Finally, in Task 6, the data collected during the demonstrations were processed into datasets along with simulations and will be published for public consumption through this web portal to support additional safety analyses, ADS development, and standards development and as a media source for public services training materials.

Safety and innovation are two of USDOT’s (2024) top strategic goals,^[1] which closely align with the goals of this project. In addition, the project objectives emphasize safety, collaboration, and gathering data for safety analysis, with the intent to generate solutions with significant public benefits. Not only has the team implemented advanced technical solutions that meet these goals, but the data collected from the testing will be widely available for use by future researchers of L4 ADS-equipped vehicles. With the emergence of automated fleets of automated taxi fleets and the increase in driverless vehicles on the roads, it is imperative that those vehicles can navigate these frequent edge case scenarios. Ultimately, this project has produced the basis for standard protocols covering complex interactions with ADS, allowing ADS to improve safety beyond what is possible with human drivers.