Google announced this weekend that it's been building robotic cars that have been driving themselves around California -- down curvy Lombard Street in San Francisco, across the Golden Gate Bridge, along the Pacific Coast Highway, around Lake Tahoe and from Google's Mountain View headquarters to Santa Monica (a 350-mile trip). So far, the cars have logged over 140,000 miles.
The company hasn't said yet what it's going to do with the cars, but they should be a good business for Google, because they draw heavily on Google's data centers.
From Google Distinguished Software Engineer Sebastian Thrun:
Our automated cars use video cameras, radar sensors and a laser range finder to “see” other traffic, as well as detailed maps (which we collect using manually driven vehicles) to navigate the road ahead. This is all made possible by Google’s data centers, which can process the enormous amounts of information gathered by our cars when mapping their terrain.
The New York Times has a detailed story on the cars here, including a description of what it's like to ride in one.
Tech blogger Robert Scoble, meanwhile, interviewed one of the engineers who works on the cars -- Mike Montemerlo, formerly of Stanford's DARPA challenge team -- back in 2007. Google hired several veterans of DARPA challenge teams.
Montemerlo "thinks about the act of driving in a much different way than you or I do," Scoble writes. "This is a fascinating discussion where he talks about the car, the sensors used, the algorithms he’s developing, and the approach he’s using to get a car through an intersection at the same time as another robotic car is there."
Here's a video of the car in action -- there is a person behind the wheel, but he or she only takes control if something goes wrong. (The picture above is from Zee News).