Autonomous cars are better city drivers than humans, Google says

Google's self-driving car effort moves from highway trials to the streets of Mountain View

Google has taken on a new challenge -- city streets -- as it works to develop a self-driving car.

That's right. A Google autonomous car has taken on hectic and unpredictable city streets and all they entail - from jaywalking pedestrians, avoiding other drivers circling in search for parking spaces or parallel parking.

In a blog post today, Chris Urmson, director of Google's self-driving car project, said the team has taken a new step with its autonomous car project -- moving from highway to far more complex in-city driving.

"Cars lurching out of hidden driveways. Double-parked delivery trucks blocking your lane and your view," Urmson wrote. "At a busy time of day, a typical city street can leave even experienced drivers sweaty-palmed and irritable. That's why over the last year we've shifted the focus of the Google self-driving car project onto mastering city street driving."

He also noted that since late in the summer of 2012, Google has "logged thousands of miles" driving on the streets of Mountain View, Calif., where Google is based. In total, Google's autonomous cars have driven more than 700,000 miles.

"A mile of city driving is much more complex than a mile of freeway driving," wrote Urmson. "We've improved our software so it can detect hundreds of distinct objects simultaneously -- pedestrians, buses, a stop sign held up by a crossing guard, or a cyclist making gestures that indicate a possible turn. A self-driving vehicle can pay attention to all of these things in a way that a human physically can't -- and it never gets tired or distracted."

According to Google, computers, and thus driverless cars, are simply better at dealing with what humans would find unexpected.

"As it turns out, what looks chaotic and random on a city street to the human eye is actually fairly predictable to a computer," said Urmson. "As we've encountered thousands of different situations, we've built software models of what to expect, from the likely (a car stopping at a red light) to the unlikely (blowing through it)."

Google's autonomous cars aren't ready for official release, though.

Urmson said Google engineers still have a lot of work to do. For instance, they still need to have their cars drive on more streets in Mountain View before they move on to driving in other cities.

He did add, though, that the team has successfully tackled "thousands" of city driving situations that would have stumped it just two years ago.

Google has made several moves recently that analysts say are likely aimed, at least partially, at boosting the autonomous car efforts. For instance, analysts note that Google's Android Wear mobile operating system can help tie together several of the company's different ventures, including autonomous cars, search tools, robotic devices and Google Glass.

Also, Google's acquisition of robotics heavyweight Boston Dynamics late last year could lead to a bolstering of the cars' autonomous software.

"With every passing mile, we're growing more optimistic that we're heading toward an achievable goal -- a vehicle that operates fully without human intervention," wrote Urmson.

Google's self-driving car takes on the streets of Mountain View, Calif., navigating near the company's business complex. (Video: Google)

Sharon Gaudin covers the Internet and Web 2.0, emerging technologies, and desktop and laptop chips for Computerworld. Follow Sharon on Twitter at @sgaudin, on Google+ or subscribe to Sharon's RSS feed . Her email address is

See more by Sharon Gaudin on

Copyright © 2014 IDG Communications, Inc.

Bing’s AI chatbot came to work for me. I had to fire it.
Shop Tech Products at Amazon