Google has been secretly building and testing autonomous vehicles that have successfully driven over 1,000 miles without any human intervention. The fleet of robot Priuses has logged 140,000 miles across California with only minimal input from a driver. The vehicles use cameras, radar and lasers to detect other cars, obstacles and pedestrians, and the vast databases of images in Google Maps and Street View to navigate. During the testing, the autonomous cars were involved in only one accident. And, in that case, the Google Carbot was rear-ended by a human driver while stopped at a red light. (Somehow, California law enforcement was aware of the tests, yet managed to keep them secret; consider us impressed.)
The project is the passion of Sebastian Thrun, the director of the Stanford Artificial Intelligence Laboratory and creator of Google’s Street View. He has been working on building a self-piloted car since at least 2005, when he led a team of Stanford students and faculty to victory in a DARPA-funded contest for building autonomous vehicles. Proponents of these cars, like Thrun and Larry Page, argue that self-driving cars could make streets safer, reduce fuel consumption, and move more passengers more efficiently. Such technology is years away from being commercially viable, and laws will have to catch up to the technology, but it’s clear that Google wants to be intimately involved in this potentially revolutionary change to how we drive. Or, more accurately, how we don’t.
Continue reading Google Cars Drive Themselves in Traffic
Google Cars Drive Themselves in Traffic originally appeared on Switched on Mon, 11 Oct 2010 10:58:00 EST. Please see our terms for use of feeds.