Only a few years ago, it seemed unfathomable that we'd ever see self-driving cars on the road in our lifetime. The countless number of obstacles, safety maneuvers, and hazards -- both expected and unforeseen -- makes the average commute ill-advised for most human motorists. But through constant development and relentless field-testing, Google
(NASDAQ:GOOG) appears to be on the fast track to getting driverless vehicles on the road even before this decade is out.
It's been years since Google first took its self-driving cars out on public roadways
, but it was typically highway cruising, which is far easier and presents fewer complexities and hazard points than city driving. In the intervening years, however, the technology has grown so advanced that Google's driverless vehicles can maneuver around everything from traffic cones and bicyclists to train crossings. Chris Urmson, director of Google's Self-Driving Car Project, lays out some of the obstacles
his team has had to overcome -- including jaywalkers, hidden driveways, and double-parked delivery trucks -- to make driverless transportation viable.
"A mile of city driving is much more complex than a mile of freeway driving, with hundreds of different objects moving according to different rules of the road in a small area," Urmson writes. "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."
Of the advancements the team has made since its last update -- and after tallying more than 700,000 miles of autonomous driving -- Urmson boasts, "[Thousands] of situations on city streets that would have stumped us two years ago can now be navigated autonomously."
But to give a better glimpse at what it's like to be escorted by a self-driving vehicle, he also includes a video showcasing the system the driverless cars use to detect, assess, and maneuver around the typical objects one sees on the road.
The car's surroundings are monitored via radar, video camera, and a spinning "bucket" mounted to the roof that emits 64 lasers, through which a 3-D environment is generated. Like cruise control, an on-off switch on the steering wheel controls the autopilot, but should anything else go wrong, there's also a big red button on the console that disengages the entire system and allows the human to take over. Although project team members never had to use it, software lead Dmitri Dolgov says, "Every robot has a big red button."
In addition to Urmson's post, Dolgov took The Atlantic
's Eric Jaffe for a spin to show off the progress his team has made with complete auto autonomy. Jaffe's blow-by-blow of the test run highlights just how "amazingly smooth" the experience is.
"We go through a yellow light, the car having calculated in a fraction of a second that stopping would have been more dangerous," Jaffe writes. "We push past a nearby car waiting to merge into our lane, because our vehicle's computer knows we have the right-of-way. We change into the right lane for a seemingly pointless reason until, a minute later, the car signals a right turn. We go the exact speed limit because maps the car consults tell it this road's exact speed limit. The car identifies orange cones in the shoulder and we drift laterally in our lane, to give any road workers more space."
The drive is so smooth, in fact, that Dolgov has nothing to log in his notes during the experiment. "Not much interesting stuff is happening," he's quoted as saying.
But while trial runs like this make driverless vehicles even closer to public availability, the Google team realizes there are far more kinks to work out in the system before they're ready to hit the streets. A mapping system far more accurate and advanced than Google Maps is needed, possibly limiting the self-driving system to only a few cities when it launches. There's also the subtle social cues that we give one another when driving -- something as casual as a wave or as subtle as eye contact -- that the system needs to master. "And it still can't understand that universal language of urban traffic: honking," Jaffe writes.
So much is riding on the success of Google's experiment -- not just the safety and convenience of drivers, passengers, and pedestrians, but also the very future of transportation: traffic jams reduced, commutes quickened, parking spots made irrelevant, and potentially thousands of lives saved each year.
Google says its goal is to get the technology to the public by 2017, with Nissan
(OTCMKTS:NSANF) hoping to release an autonomous vehicle by 2020.
If those dates hold, driving could become a futuristic dream come true in just a few years.
More tech stories:
Opinion: 3 Myths About the 'Bad for America' Comcast-Time Warner Cable Merger
The Amazon Fire TV Is Getting Better
3 Next Big Things From Amazon: What Customers Can Expect This Year
No positions in stocks mentioned.