By Colin Bird on October 10, 2010
We’ve been hearing that the age of the robotic car is just upon us ever since the car was invented, it seems. After all that hot air, a company announced it has tested a driverless car over 140,000 miles on highway and urban traffic in California, clandestinely, with little error.
That company isn’t an automaker, either. It’s Google.
Google’s autonomous car experiment uses video cameras, radar sensors and a laser range finder to navigate its cars driverlessly. Most of this technology is already used on some adaptive cruise control and emergency mitigation systems in luxury cars, but Google’s system goes a few expensive steps further. Google’s system uses detailed maps (which, from the way the company describes it, sounds like a high-def version of Google’s street view) to navigate the road ahead.
Google points out that its cars were never unmanned in the experiment. A trained safety driver was always available to take control of the vehicle if necessary, and local police were briefed on the cars’ routes beforehand. Only one accident occurred during the test, according to the New York Times, and that one Google car was rear-ended at a stoplight.
Google isn’t the first to make an autonomous car. Challenges by the Defense Advanced Research Projects Agency — better known as DARPA — have experimented with the technology for years, but its urban terrain test was simulated, not on real roads like Google’s. In fact, many people behind Google’s driverless car were contestants in the DARPA challenges.
Google says that its aim is to make cars safer. More than 1.2 million lives are lost every year in road traffic accidents, according to the World Health Organization. Google believes this number can be halved with its technology in future vehicles. The technology could also drastically reduce traffic jams, reduce wasted fuel consumption and potentially reinvent the relationship between the driver and car.