Driverless cars may not be able to stop some types of accidents despite registering hazards faster than humans. Researchers have revealed driverless cars may not be able to prevent basic prediction errors such as judging how fast another vehicle is travelling.
Missy Cummings, a robotics professor at Duke University has revealed even laser, radar and camera detectors could fail under some circumstances.
She has warned obstacles can still be missed and warned the technology will not eradicate all accidents on the roads.
She said: “There is a probability that even when all three sensor systems come to bear, that obstacles can be missed. No driverless car company has been able to do that reliably. They know that too.
“Researchers and people in the autonomous vehicle business never thought the technology would be capable of preventing all crashes now caused by humans.”
Richard Cuerden, from the UK’s Transport Research Laboratory says experts need to understand what will happen in an emergency and motorists need to get control of the car quickly.
However he still feels autonomous cars will be a safety revolution with big changes expected.
He predicts the religion in harm on motorists could be on the sma escale or even more than when seat belts became mandatory.
Electric car champions Tesla have claimed vehicles which have their AutoPilot technology engaged have lower accident rates than those which don’t use the tool.
Autopilot does not offer full autonomous control but offers a range of extra technology such as parking assist features and lane change tools.
Tesla says: “Our data shows that, when used properly by an attentive driver who is prepared to take control at all times, drivers supported by Autopilot are safer than those operating without assistance.”
However, CEO Elon us has previously admitted it would require a “massive effort” to get the technology 99.99 percent safe.
He highlighted intersections with complex traffic lights and supermarket car parks as two areas with difficult software challenges.
Published at Fri, 05 Jun 2020 10:29:00 +0000