Google may seem an unlikely company to take on the challenge of autonomous vehicles. But when you look at the team the company has put together, it's clear that Google is very much at the forefront of research in the field.
"To develop this technology, we gathered some of the very best engineers from the DARPA Challenges, a series of autonomous vehicle races organized by the U.S. Government," Sebastian Thrun, who leads the project at Google, wrote.
"Chris Urmson was the technical team leader of the CMU team that won the 2007 Urban Challenge. Mike Montemerlo was the software lead for the Stanford team that won the 2005 Grand Challenge," he explained.
"Also on the team is Anthony Levandowski, who built the world’s first autonomous motorcycle that participated in a DARPA Grand Challenge, and who also built a modified Prius that delivered pizza without a person inside," he added.
Thrun himself is a household name in the autonomous vehicle field, robotics and machine learning. He led the Standford team that won the 2005 DARPA challenge. His team took second place in the 2007 challenge. He is also a Principal Engineer at Google where he co-developed the Street View project.
More recently at Google he has been working on the autonomous vehicle project. The project stems directly from work done previously by the Standford team and the other researchers.
The seven cars have been hitting the streets for at least a year now, there have been sightings as early as November, 2009. The cars are fitted with pretty much the same technology
as the prototypes built for the DARPA Challenge, but there have been some adaptations for driving on public roads.
For obvious safety reasons, the cars aren't left wandering the streets by themselves, there is always someone at the wheel. While the operator doesn't need to intervene on a regular basis, he can take over at a moment's notice by either pressing the break pedal, turning the wheel or pushing a big red button.
Google says that police authorities have been notified about the tests. But, because, a human operator is always able to take control, it believes that the cars operate within the law, which, understandably, is trailing the technology advances.
There are some caveats, a lot of them actually, like the fact that a car has to be driven through an area first before it can do it autonomously. This is because it needs to first gather all the mapping and terrain data along with data on other obstacles such street crossings.
The cars should be capable to drive through unknown areas, but, for safety reasons, Google doesn't do that on public roads. The DARPA Challenge required the cars to drive through an unknown desert area completely autonomously.