Computer-controlled car has logged more than 700,000 miles as Google's research continues
Google has declared a partial victory as it attempts to get its computer-controlled car to understand complex and dangerous city roads.
The company said that its self-driving car could now understand a huge range of distractions and unpredictable events to safely navigate city and suburban roads.
“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,” Google explained.
In order to understand more complex roads and behaviour, Google’s self-driving car has been given a software upgrade. This allows it to understand and detect “hundreds of distinct objects simultaneously”, according to the company.
The car is able to spot pedestrians, buses, stop signs held up by traffic marshals and gestures made by cyclists. Google said that a self-driving vehicle can track all these things in a way that a human can’t, adding that the vehicle “never gets tired or distracted”.
Google explained how the system works in a video, which shows both the movements of the car and what the computer sees:
“As it turns out, what looks chaotic and random on a city street to the human eye is actually fairly predictable to a computer. 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 said.
A number of problems still need to be solved, including tackling the roads of other towns and cities outside of Google’s Mountain View headquarters.
Google added that its self-driving cars had now covered more than 700,000 miles around Mountain View in California.