Thrun's team also won the 2007 DARPA challenge with an autonomous car called the "Junior." (Brown) This all led to the development of the Google driverless car, which began in 2010. Since then, the DARPA challenged has tested and proved that the Google autonomous vehicle can traverse an abandoned military base, obey traffic lights and stop signs, yield at intersections, and merge with traffic. (Brown) .
How is Google's autonomous vehicle able to do this? There are three components that will make the Google's driverless car go: Google's software and mapping system called "Google chauffeur," and built in sensors. "The sensors are configured much like the ones used by teams in the DARPA Urban Challenge, but are neatly tucked away in the car's body rather than mounted, laboratory-style, on a roof rack."(Brown) The only thing that is clearly visible is the rotating sensor that is mounted on top of the car. The sensor is a "Velodyne high-density LIDAR-light detection," that contains "64 pulsed lasers."(Brown) The system is able to rotate 10 times per second and can capture 1.3 million points of the cars surroundings in three dimensions. This will allow for the sensors to detect 165 feet ahead and objects within 400 feet. There is also a high-resolution camera in the car that will detect pedestrians, bicyclist, and other moving obstacles. The car is able to track its position with a GPS system that is integrated with Google's Chauffeur." The mapping database for Google's driverless not only maps the roads, but also is also able to create a "street-level" imagine of them as well; thus creating a more realistic view and perception for the car. (Brown) Before each test drive, car is able to map out the road conditions, says Sebastian Thrun. When the car is ready to travel, it will refer back to the database and update as the route continues.
