The results are in, and Google has just put out some eye-popping numbers for its self-driving automobile.
While it’s long been said that self-driving automobiles like the one Google has created are much safer than ones driven by humans — but the actual difference between the two is nothing short of amazing.
A study commissioned by Google has found that self-driving cars get in one less crash per million miles driven than vehicles driven by humans, a massive 27 percent dip, according to an Engadget report.
Considering more than 30,000 people died in the United States alone in car accidents last year, and such a drop would literally save thousands of lives.
To come to their conclusions, the researchers looked at Google’s self-driving test fleet — a fleet that Google has claimed has never been in at fault in an accident. The study found that while conventionally driven cars experience 4.2 crashes per million miles driven, autonomous vehicles had a rate of just 3.2 crashes per million.
Of course, some might criticize the study for relying on a small sample size. American motorists drive billions of miles per year, allowing for a much deeper data pool to draw from, compared to a few million miles for self-driving cars. But it’s a good sign that self-driving cars in the future could result in safer roads and fewer deaths from accidents.
According to a Google FAQ on self-driving cars, a self-driving car differs from driver assistance in that “with fully self-driving technology, the car is designed to do all the work of driving and the human “driver” is never expected to take control of the vehicle at any time. The National Highway Traffic Safety Administration (NHTSA) calls this full self-driving automation (level 4). This is the type of technology Google is working on.” On the other hand, “driver assistance could include technology like adaptive cruise control or automated parking. With some driver assistance systems, there may be moments when the car is capable of self-driving, but it could also back out of this mode in certain situations and the driver is expected to take over as needed. This type of technology falls between function-specific automation and limited self-driving automation (levels 1-3) by NHTSA’s classification.”