Google develops self-driving cars

Google researchers have developed 'unmanned' cars that can drive on public highways.

The cars incorporate technology developed through participation in a series of autonomous vehicle races sponsored by the Defense Advanced Research Projects Agency (DARPA). The search giant's self-driving Toyota Prius has already clocked 140,000 hours driving on roads between the company's Mountain View, California, headquarters and its office in Santa Monica, California, with near-zero human intervention.

The robot cars take after Google's Street View vehicles and instead of a camera on the roof, the cars have an optical LIDAR (Light Detection and Ranging) sensor on top with additional radar sensors mounted on chassis.

In a blog post Google software engineer, Sebastian Thrun explained that the automated cars used video cameras, radar sensors and a laser range finder to 'see' other traffic as also detailed maps which Google collects using manually driven vehicles to negotiate the road ahead. He added, "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 researchers involved in Google's project participated in DARPA's challenges including Chris Urmson, a Carnegie Mellon robotics scientist, Mike Montemerlo, senior research engineer in Stanford's Artificial Intelligence Lab, and Anthony Levandowski, a product manager at Google who is noted for modification of a Prius to enable it to deliver pizza without a person inside.

Google's pursuit of self-driven cars stems from a desire to save humans from themselves and to also help the environment. Thrun says that as many as 1.2 million people are killed every year in road accidents. He adds that automated vehicle technology could reduce the car-related mortality by as much as 50 per cent.