Self-Driving Tech Won't Prevent Most Crashes, Study Suggests
June 12, 2020—A study from the Insurance Institute for Highway Safety suggests that self-driving technology might only prevent about a third of crashes. That's based on research into the driver-error causes in current crash data.
“It’s likely that fully self-driving cars will eventually identify hazards better than people, but we found that this alone would not prevent the bulk of crashes,” says Jessica Cicchino, IIHS vice president for research and a coauthor of the study, in a story from IIHS.
IIHS looked into 5,000 police crash reports that involved a medial response and at least one vehicle needing to be towed away. IIHS categorized the top driver error factors in those crashes as sensing and perceiving; predicting; planning and deciding; executing and performance; and incapacitation.
The researchers then looked into how many of those factors could be avoided with self-driving technology. The bottom line is that there are a lot of perceptions that human drivers take into account while operating a vehicle, and it might be more difficult than anticipated for a computer program to replicate that in all road scenarios.