On a stretch of I-24 in Tennessee that is newly equipped to serve as a testbed for intelligent transportation technologies, 100 vehicles with AI-driven adaptive cruise control will join the morning rush hour next week so that researchers can assess whether cars directed by AI can help to ease stop-and-go traffic jams by altering the behavior of human drivers.
The testing is being conducted by the CIRCLES Consortium, which consists of researchers from partners including Vanderbilt University, the University of California, Berkeley, Rutgers University-Camden and Temple University, plus participation from Nissan North America, Toyota, General Motors and the Tennessee Department of Transportation. It will take place from November 14-18 in the Nashville-Davidson County area on a four-mile section of I-24 outfitted with around 300 ultra-high-definition cameras mounted on poles every 600 feet, which produce images that are “converted into a digital model of how every vehicle behaves with unparalleled detail” (yet are anonymized via algorithm), according to a description of the I-24 MOTION testbed, which is meant to take automated vehicle tech out of its usual closed-track test environment and put vehicles into the flow of real traffic conditions. The highway testbed was just completed this fall.
The experiments are scheduled to take place between 5 a.m.-10:30 a.m. on Monday through Friday of next week. Adaptive cruise control (ACC), which is already readily available on newer vehicle models, enables a vehicle to speed up and slow down while maintaining a safe distance relative to the vehicle in front of it. In a previous closed-track experiment with 20 cars, CIRCLES Consortium researchers found that “just one vehicle equipped with the AI system changed the driving behavior of the other 20 cars, alleviating the stop-and-go dynamic that often leads to traffic jams with no obvious cause,” which both smoothed out the driving experience and led to a “sizable fuel savings” compared to driving in regular traffic jams. Now, they’ll try that out with up to 100 Nissan Rogues, Toyota RAV4s and a Cadillac XT5, each of which are equipped with the same AI-equipped ACC that was used in the earlier experiment.
“As researchers increase the scale of the testing and introduce real world driving conditions, they will investigate whether the improved traffic and fuel-economy outcomes measured in the smaller study continue to hold,” Vanderbilt University said in a release about the project. The research is being funded by the National Science Foundation and the U.S. Departments of Transportation and Energy, with additional support from Toyota North America and GM.
