Between November 14 and 18, two CERMICS researchers participated in the world’s largest collaborative autonomous vehicle experiment on the highway during rush hour. A hundred connected autonomous vehicles were deployed on the highway between 6am and 10am with control algorithms aiming to reduce traffic jams and in particular stop-and-go waves.
The goal is to reduce fuel consumption and CO2 emissions of the whole traffic by acting only on a small proportion of connected vehicles. The vehicles were looped on a 6km stretch of I-24 south of Nashville, Tennessee.
This is the culmination of a three-year project led by the CIRCLES consortium, which includes Amaury Hayat, a researcher at CERMICS, and Nathan Lichtlé, a PhD student at CERMICS and ITS Berkeley. Amaury Hayat and Nathan Lichtlé co-designed the control algorithms used in this experiment. These algorithms use either classical mathematical techniques or AI models obtained by reinforcement learning.