(c)iStock.com/Jason Doiy

There’s an old joke about the elderly driver who’s never had an accident in their 60 years on the road – but they’ve seen hundreds of them. Is that ancient gag to be re-applied for self driving cars? Google has admitted that its autonomous cars had been in 11 minor traffic accidents over a six year period, yet was adamant their cars had not been the cause of the accident.

Chris Urmson, principal engineer and software lead for Google’s self-driving car unit, wrote in a Medium post: “If you spend enough time on the road, accidents will happen whether you’re in a car or a self-driving car. All the crazy experiences we’ve had on the road have been really valuable for our project.”

Urmson noted that, of the 11 accidents, eight were caused by city driving instead of three on the freeways, with the majority of crashes caused by rear-ending, although Google’s cars have also been side-swiped and hit by a car rolling through a stop sign.

The Google exec explains the difficulties between machine learning, and predicting user behaviour, as opposed to what would actually happen. For instance, on intersections, the cars have been programmed to pause briefly after a light turns green, to more likely avoid drivers who barrel through. In other words, it’s the nature of human error which is the biggest roadblock.

Urmson added: “We’ll continue to drive thousands of miles so we can all better understand the all too common incidents that cause many of us to dislike day to day driving – and we’ll continue to work hard on developing a self-driving car that can shoulder this burden for us.”

A patent won by Google was recently unveiled which enables autonomous vehicles to understand cyclists’ hand signals. The company said last year that “what looks chaotic and random on a city street to the human eye is actually fairly predictable to a computer.”

Read Urmson’s post here.