Xianfeng “Terry” Yang, associate professor of civil and environmental engineering at the University of Maryland and director of the Maryland Transportation Institute, is working to make autonomous vehicle technology roadworthy, using artificial intelligence to maximize how efficiently AVs travel while addressing gaps where AVs lack training for split-second decisions. Yang said AVs may eventually communicate directly with traffic-management systems, and his team uses reinforcement learning, a type of AI in which algorithms improve through trial and error, to teach AVs adjustments such as slowing down so a light turns from red to green by the time the vehicle reaches an intersection. With support from the U.S. National Science Foundation, projects in his lab seek to better simulate AV performance on snowy or icy roads and allow AVs to detect hazards using information from roadside sensors. The team also focuses on rare, possibly deadly scenarios; graduate student Dianwei Chen uses large language models to identify potential edge-case scenarios and verifies and replicates them in CARLA, an open-source 3D driving simulator built on the Unreal Engine, looking with reinforcement learning for ways to prevent the situation and, if a crash is unavoidable, to minimize injury and damage.