Google self-driving cars can now spot cyclist signals, level crossings and more
Google's self-driving cars are now street smart for cities.
Up until recently, the autonomous vehicles from Mountain View were best suited to wide-open roads like highways (or as Californians would say, freeways). But they've now logged nearly 700,000 miles during testing, which has given Google the opportunity to hone its self-driving technology and software. And sometime last year, the company even started tackling city street driving.
"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," explained Google in a blog post. "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."
The video below provides a demonstration of how Google's self-driving cars can now pay attention to all the distractions on city streets in Mountain View. The vehicle's technology and software can predict what we view as random chaos - without getting tired or distracted like us. After learning and recognising "thousands of different situations," Google's cars can do things like pause for cyclists, yield at train tracks, and spot crossing pedestrians, etc.
"We still have lots of problems to solve, including teaching the car to drive more streets in Mountain View before we tackle another town, but thousands of situations on city streets that would have stumped us two years ago can now be navigated autonomously," Google added.