Researchers have created an algorithm that could work alongside an extremely sensitive laser technology that reflects off nearby objects to help self-driving cars see around corners.
Imagine that a driverless car is making its way through a winding neighborhood street, about to make a sharp turn onto a road where a child’s ball is rolling across the street. Although no person in the car can see that ball, the car stops to avoid it.
By: David Danks, Carnegie Mellon University
In 2016, self-driving cars went mainstream. Uber’s autonomous vehicles became ubiquitous in neighborhoods where I live in Pittsburgh, and briefly in San Francisco. The U.S. Department of Transportation issued new regulatory guidance for them. Countless papers and columns discussed how self-driving cars should solve ethical quandaries when things go wrong. And, unfortunately, 2016 also saw the first fatality involving an autonomous vehicle.
Autonomous technologies are rapidly spreading beyond the transportation sector, into health care, advanced cyberdefense and even autonomous weapons. In 2017, we’ll have to decide whether we can trust these technologies. That’s going to be much harder than we might expect.
Trust is complex and varied, but also a key part of our lives. We often trust technology based on predictability: I trust something if I know what it will do in a particular situation, even if I don’t know why. For example, I trust my computer because I know how it will function, including when it will break down. I stop trusting if it starts to behave differently or surprisingly.
In contrast, my trust in my wife is based on understanding her beliefs, values and personality. More generally, interpersonal trust does not involve knowing exactly what the other person will do – my wife certainly surprises me sometimes! – but rather why they act as they do. And of course, we can trust someone (or something) in both ways, if we know both what they will do and why.
I have been exploring possible bases for our trust in self-driving cars and other autonomous technology from both ethical and psychological perspectives. These are devices, so predictability might seem like the key. Because of their autonomy, however, we need to consider the importance and value – and the challenge – of learning to trust them in the way we trust other human beings.
We want our technologies, including self-driving cars, to behave in ways we can predict and expect. Of course, these systems can be quite sensitive to the context, including other vehicles, pedestrians, weather conditions and so forth. In general, though, we might expect that a self-driving car that is repeatedly placed in the same environment should presumably behave similarly each time. But in what sense would these highly predictable cars be autonomous, rather than merely automatic?
The death of a person earlier this year while driving with Autopilot in a Tesla sedan, along with news of more crashes involving Teslas operating in Autopilot, has triggered a torrent of concerns about the safety of self-driving cars.
But there is a way to improve safety across a rapidly evolving range of advanced mobility technologies and vehicles – from semi-autonomous driver assist features like Tesla’s Autopilot to a fully autonomous self-driving car like Google’s.
The answer is connectivity: wireless communication that connects vehicles to each other, to the surrounding infrastructure, even to bicyclists and pedestrians. While connectivity and automation each provide benefits on their own, combining them promises to transform the movement of people and goods more than either could alone, and to do so safely. The U.S. Department of Transportation may propose requiring all new cars to have vehicle-to-vehicle communication, known as V2V, as early as this fall.
Tesla blamed the fatal crash on the failure of both its Autopilot technology and the driver to see the white tractor-trailer against a bright sky. But the crash – and the death – might have been avoided entirely if the Tesla and the tractor-trailer it hit had been able to talk to each other.
Autonomous vehicles that aren’t connected to each other is a bit like gathering together the smartest people in the world but not letting them talk to each other. Connectivity enables smart decisions by individual drivers, by self-driving vehicles and at every level of automation in between.
Despite all the safety advances in recent decades, there are still more than 30,000 traffic deaths every year in the United States, and the number may be on the rise. After years of steady declines, fatalities rose 7.2 percent in 2015 to 35,092, up from 32,744 in 2014, representing the largest percentage increase in nearly 50 years, according to the U.S. DOT.
