Google’s self-driving cars have spent the past year cruising the roads of Mountain View, California. Google engineers use the data from those drives to make the cars’ onboard computers more sophisticated. But how good are the cars’ driving records? Pretty flawless, according to the local cops. "We have not cited any Google self-driving cars," Sergeant Saul Jaeger, the press information officer at the Mountain View Police Department, told The Atlantic.
He also claimed that no special understanding about the cars exists between his police force and Google, suggesting that an officer who saw one of the vehicles driving in an unsafe manner would either stop or ticket it. (Mountain View Police also wouldn’t have an accounting of any unreported fender-benders or similar mishaps.) Click here for vehicle-related jobs.
While California law states that responsibility for a car’s actions belongs to the human being in the driver’s seat, Google thinks traffic tickets for self-driving vehicles should go to the company responsible for building the software. "What we've been saying to the folks in the DMV, even in public session, for unmanned vehicles, we think the ticket should go to the company. Because the decisions are not being made by the individual," a company spokesperson explained to The Atlantic. Google claims that onboard sensors and cameras directed by its autonomous-driving software can recognize hundreds of objects typically encountered in the process of navigating a city street, such as people, stop signs, cyclists, buses, and train crossings; by this point, the vehicle can even react to subtler signals, such as a cyclist gesturing a lane change, although much work remains to be done: Navigating the chaotic streets of a major metropolis, for example, might still prove too much for Google’s systems. Tesla Motors, Nissan, Audi, Mercedes, and other companies are also exploring self-driving vehicles; like Google, all have expressed a need for driving engineers and software developers
skilled in tasks such as designing and validating algorithms.