AUTONOMOUS CARS ARE COMING, BUT WE DON’T KNOW WHEN
Chances are, your dealership personnel are already taking questions from customers about autonomous vehicles. Those who have paid any attention to recent mass media reports are probably convinced such vehicles will be on sale by 2020, if not sooner. Some may believe they’re already here.
They can be forgiven for such beliefs for, in many cases, they are driven by the hyperbole of automakers, or in some cases ‘wannabe’ automakers, themselves.
Several manufacturers have publicly set a 2020 target for introducing autonomous technology, fueling collective expectations. Ditto for Google’s ongoing tease about building such cars itself. The use of misleading terms, like “Autopilot” for the latest iteration of Tesla’s Advanced Driver Assistance System (ADAS), confuses the issue further.
Autonomous cars really are coming. On that point virtually everyone in the industry agrees. But there is little agreement in terms of when they will arrive and exactly what form they will take when they do.
What is not being said, at least not loudly, is that there are multiple levels of autonomy, at least in the current lexicon, and what is being touted for 2020 or before is a long way from the fully driverless, go-anywhere in-any-weather vehicle that the term evokes in many minds.
In a presentation at the 2016 Consumer Electronics Show (CES) in Las Vegas, Gill Pratt, CEO of the recently-established Toyota Research Institute, summed up the situation concisely: “These systems (at their current stage of development) can only handle certain speed ranges, certain weather conditions, certain street complexity, or certain traffic.”
He added: “Most of what has been collectively accomplished has been relatively easy because most driving is easy. Where we need autonomy to help us is when the driving is difficult.”
Expanding on that point, the U.S. National Highway Traffic Safety Administration (NHTSA) has developed a hierarchy of vehicle automation, where ‘Level 0’ is no autonomy at all and ‘Level 4’ — the highest level — is no driver input at all. On a larger scale, the International Organization of Motor Vehicle Manufacturers (OICA), along with Society of Automotive Engineers (SAE) International, has established a similar ladder of automation levels ranked from 0 to 5.
In either case, vehicles at the highest level will need destination or navigation input but will not require that a driver be ready to take over control at any time during the drive.
Some of the challenges to achieving full autonomy are technical — things like consistently being able to identify and analyze what’s going on with and around the vehicle in all circumstances and conditions
That initial input might even be provided remotely, so such vehicles could operate without anyone onboard, as robotic taxis, so to speak. They’ll also be capable of operating on any road in any weather conditions, at least to the same extent that a human driver could.
But we’re not there yet. Not even close, beyond some narrowly defined locales and circumstances — primarily in warm, dry climes, on controlled-access roads with clear lane markings.
Some of the challenges to achieving full autonomy are technical — things like consistently being able to identify and analyze what’s going on with and around the vehicle in all circumstances and conditions.
The future of autonomous driving can’t rely on ideal conditions. “It’s one thing for a car to drive itself in perfect weather,” said Jim McBride, technical leader for autonomous vehicles at Ford. “It’s quite another to do so when the car’s sensors can’t see the road because it’s covered in snow.”
That’s why Ford has announced it is now testing its Fusion autonomous development vehicles in real winter conditions, including on snow-covered roads. Much of that work is focused on developing mapping technology sufficient to locate the vehicles precisely — within centimetres, not metres — without relying on visual cues, particularly in bad weather.
Other challenges yet to be overcome are regulatory, including a lack of international protocols and standards that allow vehicles to communicate with each other, wherever they are. There are also social, ethical and purely legal matters to consider, some of which may be in conflict with each other.
With today’s most sophisticated Advanced Driver Assistance Systems, we are well along the path to semi-autonomous driving, and near fully autonomous in many highway driving scenarios. In many respects we’re almost there. But that last step is a big one.
Those closest to the challenges are, perhaps, the most cautious in predicting when they’ll all be overcome.
Axel Gern, head of the autonomous vehicle department at Mercedes-Benz Research and Development North America, is optimistic about the near-term potential for autonomous operation in prescribed highway driving conditions. But when asked about full autonomy in dense city environments, he suggested that 10 years would be
“a very aggressive timeline!”
