作者: Noah J. Goodall
DOI: 10.1016/J.AAP.2021.106056
关键词: Crash 、 Motor vehicle crash 、 Mode (statistics) 、 Economic impact analysis 、 Naturalistic driving 、 Economic cost 、 Computer science 、 Aeronautics
摘要: Abstract Automated vehicle developers in California are required to submit records of crashes and distances traveled autonomous mode for all vehicles their fleets. Several studies have investigated this database compare automated crash rates with national rates. Although struck from behind 73 % crashes, is the first study struck-from-behind using equivalent definitions. Rear-end collisions substantial public health economic impacts, representing a third $3.9 B annual costs. In study, were found be at 4.8 times rate human-driven naturalistic driving study. When controlling environment, AVs 5.0 higher urban not significant business/industrial driving, although these results different manufacturers, complicating results. more likely when stopped than moving compared vehicles, suggesting that vehicles’ decisions about where stop or remain intersections plausible contributing factors unexpected deceleration.