The Takeoff Episode 03: Why Team and Culture Matter for Travel Startups Sponsored This content is created collaboratively with one of our sponsors.
The self-driving cars need to be tested and trained for all possible situations before being released to the public. City street mastery is key problem to solve in the product’s evolution.
It’s been a while since we last heard about one of Google’s most intriguing “moonshots:” the driverless car.
In a video teaser, Google offers a new look at how its self-driving cars perceive a busy city street. The company details that it has up dated its 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.”
In a blog post on the matter, Google shares that its cars have “logged nearly 700,000 autonomous miles” in total. “We still have more work to do,” Google says, “but it’s fun to see how many situations we can handle smoothly and naturally.”
Google’s driverless car project first got its start at Stanford in the 2004 DARPA Grand Challenge.
Meanwhile, Nissan is apparently testing a “self-cleaning” car.