Uber has been forced to halt autonomous vehicle tests after an accident involving one of its autonomous cars claimed a woman’s life in Tempe, Arizona. The accident could force the company and other autonomous vehicle makers to review existing concepts and to invest more in safety in the coming days.
Britain may have to wait longer for autonomous vehicles
Back in November last year, Chancellor Philip Hammond indicated at BBC’s Andrew Marr show that the government wanted autonomous vehicles to debut on Britain’s roads in the next four years. He added that the move was necessary as it would enable the UK to ‘lead the next industrial revolution’.
He added that he would announce new regulations which will pave the way for leading car manufacturers to invest on driverless technologies and to test their concepts on the UK’s roads. Jaguar Land Rover has already begun testing and may soon be followed by other manufacturers in the coming months.
Even though the government is strongly in favour of introducing autonomous vehicles in the UK, the concept received a major jolt ahead of its introduction in the United States when one of its autonomous vehicles struck and killed a woman in Tempe, Arizona, forcing Uber CEO Dara Khusrowshahi to release a statement on Twitter.
“Some incredibly sad news out of Arizona. We’re thinking of the victim’s family as we work with local law enforcement to understand what happened. Our hearts go out to the victim’s family. We’re fully cooperating with @TempePolice and local authorities as they investigate this incident,” he tweeted.
The incident highlights that despite the recent advances made in autonomous vehicle technology, safety issues are yet to be plugged and the introduction of such cars on roads need not be rushed until their manufacturers demonstrate that they have technologies in place to prevent road accidents and loss of life.
Commenting on the challenges faced by autonomous vehicle manufacturers, Evgeny Chereshnev, CEO at Biolink.Tech said: “At the moment, the autonomous vehicles we have already – such as airport trains and the Docklands Light Railway in the UK – thankfully work hazard free because the journeys are simple and predictable. Statistically, there is very little risk of anything going wrong because they’re programmed to only go from A to B. Even with other forms of partly automated transport, such as planes and ships, there is still always a human there to monitor them.”
“However, on roads, autonomous cars have the potential to be compromised almost 100% of the time. There are many factors that bring significant risk to the use of autonomous cars, all of which are from external sources such as other human drivers, pedestrians, cyclists and even sudden extreme weather changes.
“Unless all cars on the road are autonomous, sadly there is the potential for more people to be injured or worse. That being said, I think it’s important not to kill the initiative of autonomous cars. Any risk of road accident by self-driving cars will be significantly less than the current number of road accidents – which is reported to be a staggering 1.25 million per year
“Unfortunately, unless we have fully controlled environments on the roads, accidents can and will continue to happen,” he added.
Cyber threats faced by autonomous vehicles
Aside from grappling with safety issues, manufacturers of autonomous vehicles also need to keep in mind the various cyber threats that nation-states and hackers could pose to such vehicles.
“If there was a war or escalation with a country with strong cyber capability, I would be very afraid of hacking of vehicles. Many of our enemies are nuclear powers but any nation with the ability to launch a cyberstrike could kill millions of civilians by hacking cars,’ warned Justin Cappos, a computer scientist at New York University, in an interview goven to The Times.
“It’s daunting. They can send messages to the brakes and shut off the power steering and lock people in the car. and do other things that you wouldn’t want to happen. Once you are in the network you are able to communicate with any device so you could send a message to engage the brakes,’ he added.
Cappos also said that at the moment, car components do not have the ability to understand where messages come from and whether they are authentic, and are thus succeptible to hacking attacks.