Humans need to pass driving tests, so why shouldn’t robots?
That’s the question being asked by researchers at the U-M Transportation Research Institute. Michael Sivak and Brandon Schoettle are proposing that self-driving cars be subject to tests that measure their vision, knowledge of traffic laws and other skills.
“Current prototypes of self-driving vehicles are not perfect,” they write. “For example, some of them occasionally cross the centerline on curves even in good weather conditions.” Rain, snow, and darkness make it more difficult for self-driving cars to recognize potential hazards, the researchers write, so the cars should be tested under a variety of weather conditions.
The researchers propose a graduated license system to fix the weather problem. For example, a certain model could be licensed to drive in good weather, but not in the snow. And another model, which has a problem seeing at night, could be licensed to only drive during the day.
Now the researchers don’t think every single self-driving car needs to be tested, but the new models that come out should be put to the test. They also say it should be easy for the self-driving cars to remember the rules of the road.
“Programming all driving and traffic laws and regulations into an onboard computer should be relatively easy.. In principle, all that needs to be done is to program the complete set of laws and regulations that are contained in any state’s booklet for prospective drivers.”
Google, which hopes to have a commercially-available self-driving car by 2020, recently said it won’t offer self-driving cars in areas where it snows, at least not initially.
And judging by the recent fail videos of Tesla Autopilot, autonomous driving technologies still face many caveats before hitting the roads. Tesla’s Level 2 autonomous driving system is still in the beta-testing phase and people have been advised to never fully take their hands off the steering wheel, but the hair-raising videos point to the need for self-driving car standards to be a priority in the near future.