USA Banner

Official US Government Icon

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure Site Icon

Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

U.S. Department of Transportation U.S. Department of Transportation Icon United States Department of Transportation United States Department of Transportation

Analytics and Artificial Intelligence in a Federal Framework That Encourages Transportation Innovation

Thursday, June 21, 2018

A “growing torrent” of technology advances are poised to fundamentally change transportation, according to U.S. DOT Under Secretary of Transportation for Policy Derek Kan, who kicked off the Volpe Center’s 2018 speaker series, Transportation in the Age of Artificial Intelligence and Predictive Analytics.

“Three technology areas are garnering significant investment, testing, and deployment,” Kan said. “They are data integration and analytics, automated vehicles, and unmanned aerial systems.”

Safety is propelling U.S. DOT programs, and the safety impetus is real. Road vehicle crashes caused more than 37,000 traffic fatalities in 2016, and were up 14 percent from 2014.

Watch video highlights from Under Secretary Derek Kan’s talk kicking off Transportation in the Age of Artificial Intelligence and Predictive Analytics. Scroll down to watch his full remarks.

Drawing on New Data Analytics to Prevent Road Fatalities

The department has access to a vast amount of data. Looking at that data in new ways may help turn the fatal crash trend south, Kan said. One way is to rethink how U.S. DOT modal administrations collect and organize road data.

“The department’s data is often siloed and it comes at different cadences,” Kan said. “Various data sources are analyzed separately, housed in different modes, and many are only made available at an annual basis. Much of this data has been collected and organized in the same way for years, and maybe even decades. Recent innovations in data science provide the opportunity to do so much more.”

A data-rich analytic environment is blooming. Automated vehicles use artificial intelligence, and big data sources provide previously unseen amounts of information on roadway and operating conditions. This data can help transportation professionals assess fatal crash risk at increasingly granular levels.

“This is one of the big pushes Secretary [Elaine L.] Chao has given us: use the latest technology to prevent traffic fatalities,” Kan said.

U.S. DOT established its Safety Data Initiative to do that. Because speed is a contributing factor in many traffic fatalities, one pilot project will integrate established data on crashes and highway design with anonymous data from GPS devices. For the first time, U.S. DOT will be able to directly analyze how speed—and speed differentials—and roadway characteristics interact to affect the likelihood of crashes.

Another pilot project will integrate crash data with data on hazards and conditions from the crowd-sourced app Waze. This effort will determine if it is possible to use a crowd-sourced application as a reliable, timely indicator of traffic crashes, and to estimate crash risk.

“The vision has always been, ‘Let’s use new countermeasures—let’s deploy capital to install countermeasures, broader roads, traffic circles,’” Kan said. “But there’s a whole other way to bring down traffic fatalities, and that’s using 0s and 1s—bits and bytes.”

Frameworks, Not Prescriptions, for Deploying Automated Vehicles

Volpe Center analysts were closely involved in developing Preparing for the Future of Transportation: Automated Vehicles 3.0, released October 2018, which offered a holistic, multimodal framework to accelerate the safe testing and integration of surface automated driving systems. AV 3.0 builds on Automated Driving Systems 2.0: A Vision for Safety, which encourages best practices and prioritizes safety for automakers deploying advanced driver assistance technologies. 

As U.S. DOT continues to refine its approach to automated vehicle deployment, it will only pursue regulations that focus on the capabilities those vehicles should have, without prescribing the technologies to achieve those capabilities, Kan said.

“We hope to prepare for the future and encourage innovation without compromising safety,” Kan said. “Under this approach, we will not pick winners and losers among technology innovations. We will remain tech-neutral and let the quality of safety performance and market interest drive the evolution of innovative technologies.”

Integrating Drones into the National Airspace

Similar to U.S. DOT’s approach to automated vehicles, regulations related to unmanned aerial systems (UAS) need to strike a careful balance between ensuring safety and allowing the public and private sectors to boldly experiment with UAS technologies and operations, Kan said.

“It’s been only 18 months since the small UAS rule became effective, yet the rule—part 107—is the first comprehensive set of performance-based rules for routine small UAS operations in the United States,” Kan said. “Today, we have 50,000 new commercial drone pilots.”

Drone technology and public acceptance are still developing, but U.S. DOT already has several efforts that are striking that balance between public safety and UAS integration. The UAS Integration Pilot Program is bringing together state, local, and tribal governments with private industry to understand public response to expanded UAS operations. FAA’s B4UFLY app is helping UAS operators understand restrictions or requirements in areas where they want to fly. And U.S. DOT is coordinating UAS cybersecurity challenges with partners at the Departments of Defense, Homeland Security, and Justice.

“It’s exciting to me to be here,” Kan said. “It’s an exciting time to be at Volpe because all of you will be playing a critical role in helping form these regulations and usher in new technologies.”

Speaker series logo: Transportation in the Age of Artificial Intelligence & Predictive Analytics.