Driverless car crash damage could be minimised with intelligent system
Engineers at the University of Waterloo have developed a system to help driverless cars minimise injuries and damage in the event of an unavoidable crash.
After recognising that a collision of some kind is inevitable, the system works by analysing all available options and choosing the course of action with the least serious outcome.
The first rule for the autonomous vehicle crash-mitigation technology is avoiding collisions with pedestrians.
From there, it weighs factors such as relative speeds, angles of collision and differences in mass and vehicle type to determine the best possible manoeuvre, such as braking or steering in one direction or another.
“We consider the whole traffic environment perceived by the autonomous vehicle, including all the other vehicles and obstacles around it,” said mechanical engineering professor Dongpu Cao.
Amir Khajepour, who led the project, said that while vehicle safety should improve dramatically with autonomous vehicles, there are just too many uncertainties for self-driving vehicles to handle them all without some mishaps.
“There are hundreds, thousands, of variables we have no control over,” he said. “We are driving and all of a sudden there is black ice, for instance, or a boulder rolls down a mountain onto the road.”
AVs are capable of limiting damage when a crash is unavoidable because they always know what is happening around them via sensors, cameras and other sources and routinely make tens, even hundreds, of decisions per second based on that information.
The new system decides how an AV should respond in emergency situations based primarily on pre-defined mathematical calculations considering the severity of crash injuries and damage.
Researchers didn’t attempt to factor in extremely complex ethical questions, such as whether an AV should put the safety of its own occupants first, or weigh the well-being of all people in a crash equally.
When carmakers and regulators eventually hammer out the ethical rules for self-driving vehicles, Khajepour said, the system framework is designed to integrate them.
Earlier this year, ride-hailing service Uber was found to be not criminally liable in what is arguably the most infamous fatal self-driving car crash to date, when one of Uber's self-driving cars failed to identify a pedestrian crossing the road at night in Tempe, Arizona, in March 2018. Elaine Herzberg was killed when the SUV struck her as she wheeled her bicycle across the road. The autonomously controlled vehicle did not brake in time, despite there also being a human Uber representative behind the wheel at the time.
Ms. Herzberg's relatives reached an out-of-court private settlement with Uber soon after the crash, as the Silicon Valley company sought to quickly resolve the matter and forestall a potentially lengthy and damaging legal battle over the first-known fatality involving an autonomous vehicle.
Sign up to the E&T News e-mail to get great stories like this delivered to your inbox every day.