RoboCar Guide / Science
Science · driverless cars
Here are four research papers on driverless cars, all free to read on arXiv. Each summary is ours, written from the paper's abstract.
Our own artist's impression. No figures are taken from the papers.
Every number links to the paper it comes from.
How driverless rides compare with human drivers on crashes, injuries and insurance claims.
What they foundWaymo's rider-only crash rates over 56.7 million miles were compared with human benchmarks across 11 crash types.
Waymo crashes from NHTSA's Standing General Order reports, against human benchmarks matched by place and road type.
What they foundOver 7.14 million rider-only miles, Waymo had 0.6 injury-reported crashes per million miles vs 2.80 for humans, about 80% fewer.
NHTSA crash reports from Phoenix, San Francisco and Los Angeles through October 2023.
Source for: Injury-reported crashes, Waymo rider-only
What they foundCrash rates were compared for Uber trips, supervised test vehicles and driverless Waymo and Cruise service in California.
California CPUC and DMV data, 2020–2023.
What they foundAcross 3.8 million rider-only miles, Waymo had zero bodily-injury liability claims, and it was significantly safer toward other road users than human drivers.
Waymo's liability claims compared with Swiss Re human-driver baselines.
New papers on driverless cars, checked weekly, summarized after a person reads it.
No new papers on driverless cars this week. We check arXiv every week.
We search arXiv for driverless cars papers and keep the ones that add something new, are clearly written up and are relevant to a general reader. A person reads every abstract.
Summaries are written by us from each paper’s abstract, not copied. Diagrams are our own simplified drawings, not figures from the papers.
“Peer-reviewed” means arXiv lists a journal reference for the paper. Everything else is marked “Preprint”: shared openly, possibly before or without journal review.
Paper metadata courtesy of arXiv. Summaries written by us; read the original papers via the links. Paper metadata from arXiv.org, which does not endorse this site.