Crash rates caused by autonomous cars are probably lower than records from crashes caused by conventional vehicles, according to a new report commissioned by Google and published by the Virginia Tech Transportation Institute. Researchers analyzed the national crash data from more than 3,300 vehicles which were driven by more than 34 million miles.

According to the findings published in a press release on Friday, the national crash rate registers 4.2 crashes per million miles. When comparing it to autonomous vehicles, it seems the latter have a smaller number, which reaches of 3.2 crashes per million miles, even considering all levels of severity.

According to a new report, crash rates caused by autonomous cars are lower than records from crashes caused by conventional vehicles. Credit:

It was stressed that all accidents caused by self-driving cars are mandatorily reported, however, it appears that not all car crashes caused by common vehicles must be reported in all states. In order to obtain more precise results, researchers took data from strictly monitored common vehicles. Also, the data from more than 50 self-driving cars from Google, which registered 1.3 million miles in Texas and California, were analyzed.

According to registers of crashes caused by autonomous cars, just 17 crashes were recorded in the last six years, and it appears they were caused by humans who were distracted with the complex system the cars have in its pedals.

It was reported in 2013 by the National Highway Traffic Safety Administration, that 2.31 million people were injured as a consequence of vehicle crashes. In the same year, police reported approximately 5.5 million of motor vehicle traffic crashes that caused the death of 32,719 people. The main causes of car accidents in the U.S. are related to alcohol consumption, speeding and red light running.

“Current data suggest that conventional vehicles may have higher rates of more severe crashes than self-driving cars, but given the small overall number of crashes for the self-driving car at these levels, there is insufficient data to draw this conclusion with strong confidence. However, there is statistically-significant data that suggest less severe events may happen at significantly lower rates for self-driving cars than conventional vehicles,” wrote researchers from the Virginia Tech Institute in a press release.

That being said, the Virginia Tech Institute, which declared it seeks to save lives, time and money, while protecting the environment, explained that with the boom of autonomous cars, more data will be analyzed and less crashes will occur, since many self-driving makers are developing systems that function in real time with artificial intelligence, in order to continuously update the driving algorithms to prevent accidents.

Last year, it was announced by the California state that regulations for self-driving cars are going to be published soon. Among the first measures that were mentioned, it is mandatory that the high-tech vehicles need to have a steering wheel, also, they must be equipped with a system capable of recognizing and stopping cyber attacks.

It is calculated by car analysts that autonomous cars will dominate the roads in a few years, since companies such as Google, Tesla, Toyota, Ford and Uber, are already investing in it. By this moment, there are several cars from Tesla Motors that are already circulating with a self-driving system that appears to have a pristine performance. However, Toyota is the company that owns most of the self-driving patents.

“We wish to enhance the safety of automobiles with the ultimate goal of creating a car that is incapable of causing a crash, regardless of the skill or condition of the driver. Most of what we have collectively accomplished with intelligent cars to date has been relatively easy because most driving is relatively easy. Where we need to help is not where driving is easy. We need to solve driving when it’s hard,” Gill Pratt, CEO of the company remarked on Thursday at the Consumer Electronics Show in Las Vegas, that artificial intelligence is fundamental when thinking about the function of self-driving vehicles.

Source: Virginia Tech Institute