Categories: M2MNetworks

Google Self-Driving Car Report Reveals 13 ‘Near-Misses’

Google has released a full report detailing the results of its initial spate of testing for self-driving cars.

The search giant revealed the during nearly 425,000 miles of testing on the roads around its California headquarters, its vehicles were never at fault in any recorded crashes, although it did admit that engineers riding in the cars had to assume manual control several times.

The data covers Google’s testing between September 24 2014 and November 30 2015, and was released as the company begins to ramp up development for a public launch soon.

Safety first

Overall, Google’s engineers had to assume control of an autonomous vehicle 341 times during the period detailed.

Of these, 272 incidents stemmed from an issue affecting the “overall stability of the autonomous driving system”, meaning a breakage or mechanical failures affecting components such as communication and computing systems.

The other 69 are “related to safe operation of the vehicle,” – i.e. occasions when the vehicle was set to make a bad decision, forcing the passenger to take the wheel. This including 13 such incidents over the last two months of testing, which, according to computer simulations carried out afterwards indicated that there would have been a crash if they had not taken control.

Of these thirteen incidents, two would have involved hitting a traffic cone, with the other 11 noted as potentially having been more serious.

The company is also keen to state that “driver disengagements” dropped significantly over the past year.

The cars involved were modified versions of the Lexus RX450h and electrically-powered Google Koala test cars.

“These events are rare and our engineers carefully study these simulated contacts and refine the software to ensure the self-driving car performs safely,” Google’s report stated, “We are generally driving more autonomous miles between these events.”

“From April 2015 to November 2015, our cars self-drove more than 230,000 miles without a single such event.”

The period included in the results covered May 2015, when Google revealed that its prototype vehicles had been involved in 11 accidents over six years of testing – however, all of these were the result of errors by other (human) drivers.

What do you know about transport technology? Try our quiz!

Mike Moore

Michael Moore joined TechWeek Europe in January 2014 as a trainee before graduating to Reporter later that year. He covers a wide range of topics, including but not limited to mobile devices, wearable tech, the Internet of Things, and financial technology.

Recent Posts

OpenAI, Broadcom In Talks Over Development Of AI Chip – Report

Rebelling against Nividia? OpenAI is again reportedly exploring the possibility of developing its own AI…

2 days ago

Microsoft Outage Impacts Airlines, Media, Banks & Businesses Globally

IT outage causes major disruptions around the world, after Crowdstrike update allegedly triggers Microsoft outages

2 days ago

GenAI Integration Efforts Hampered By Costs, SnapLogic Finds

Hefty investment. SnapLogic research finds UK businesses are setting aside three-quarters of their IT budgets…

3 days ago

Meta Refuses EU Release Of Multimodal Llama AI Model

Mark Zuckerberg firm says European regulatory environment too ‘unpredictable’, so will not release multimodal Llama…

3 days ago

Synchron Announces Brain Interface Chat Powered by OpenAI

Brain implant firm Synchron offers AI-driven emotion and language predictions for users, powered by OpenAI's…

3 days ago