Google’s latest technology experiment isn’t about helping you navigate the Web—it’s about teaching cars to drive themselves. Exhibiting its usual boundless faith in technology, the company says it hopes its work leads to cars that are safer and more efficient. The company has logged more than 140,000 miles in seven modified Toyota Priuses, along with one modified Audi TT, in self-driving experiments, primarily around the San Francisco Bay area.
The company won’t say how much it’s spending on the project, but it’s clear that the equipment is far from everyday: the cars are loaded with sensors that feed data to an onboard computer that, in turn, controls the vehicle.
A. LIDAR Laser
Mounted on top of the car is a light detection and ranging (LIDAR) laser, a technology commonly used in police speed-detection guns. Like any other LIDAR system, it determines the distance and position of surrounding objects by sending out laser pulses and measuring how long it takes for the light to bounce back. On the Google cars, the device makes about 10 rotations per second, gathering data about objects within a circle of roughly 70 meters.
B. Radar
Three radar sensors in the front part of the car, and one behind, also help detect nearby objects and determine how close they are.
C. Position Sensors
A sensor on the left rear wheel detects small movements. This data, when combined with information from a GPS system and internal gyroscopes, helps the self-driving car calculate which direction it is facing.