One of the mains goal for autonomous vehicles is to improve human safety, by reducing the number of traffic crashes. But how do you know if autonomous vehicles actually do so, under what conditions, and by how much? A benchmark is needed, and a new white paper offers one that is a result of research and industry collaboration.
Researchers from the University of Michigan Transportation Research Institute (UMTRI) worked with General Motors and the Virginia Tech Transportation Institute (VTTI) and the autonomous vehicle company Cruise on a two-year study that assessed the number and severity of crashes involving ride-hail drivers in San Francisco, and comparing that to crashes involving autonomous vehicles, to come up with a baseline benchmark.
