Google Sees Self Driving Cars Soon, Federal Regulators Doubtful

After months of successful testing, Google believes that their self-driving vehicle technology could be ready for use by the public within the next three to five years, though regulators and the insurance industry are much more skeptical. Even as public testing in Washington, California, and Nevada has delivered sterling results, there are still major issues to resolve, such as who or what would hold liability in the event of an accident, or how states will license these driverless automobiles.

Nevertheless, Google is confident of the success of their system, and believes that the issues in the way of mass implementation can be solved swiftly. “The improvement can be such that we can make cars that drive safer than people do,” says Anthony Levandowski, product manager for Google’s self-driving car technology, in a meeting with the Society of Automotive Engineers in Washington. “I can’t tell you you’ll be able to have a Google car in your garage next year. We expect to release the technology in the next five years. In what form it gets released is still to be determined.”

The largest roadblock, according to the technologies designers, is ensuring that the software is reliable for the long term in any driving situation. If something were to go wrong with the system, a vehicle would then suddenly have nothing piloting it. “We’re really focusing on building in the reliability so we can trust and understand the system will perform safely in all conditions,” Levandowski adds. “How can you trust the system? How do you know how it can perform? How do you design it with proper processes in order to understand and minimize failure? How do you bake into a car redundant braking?”

In normal driving conditions, a self-driving system can perform expertly, but in a unique and dangerous situation, only a human driver is able to make in the moment decisions to avoid and accident. An autonomous system would need to have responses to any situation preprogramed to know how to react, though some situations are impossible to prepare for. Human drivers also possess a sense of judgment behind the wheel, able to interpret what is happening around them and infer possible danger. If a ball were to bounce into the road for example, a person would be able to make the assumption that a small child may soon dart into the street after it, despite not seeing anything.

And yet despite these potential issues, the accident avoidance potential of autonomous vehicles is exciting even the federal government. Each year, more than 30,000 people die in car crashes, according to the National Highway Traffic Safety Administration, and by removing the kinds of drivers that create a majority of accidents today, that number would plummet significantly. Before the system can be made available however, the federal agency will need to establish clear safety regulations for these vehicles, and develop matter of fact means of testing them.

“It gets to be a massive challenge to figure out how the government will come up with a performance standard that is objective and testable for so many different scenarios where failure could possibly occur,” says Dan Smith, associate administrator for vehicle safety for the NHTSA. “Part of that has to do with if we should be looking at the underlying electronics.”

Liability issues could delay public use the longest, as the insurance industry is still working to determine who would be at fault in a crash: would it be the company who built the vehicle, or the software itself? The solution to the questions could take significant time, some industry insiders say, meaning the technology could actually be between 10 to 20 years away from public use. If and when these issues are finally resolved, autonomous vehicle technology may quickly become the norm on public roads. What was once believed to be a dream of science fiction will soon be made a reality, and transportation as we know will change forever.