Select Page

The world may not have the flying cars people expected when The Jetsons was our vision of the future, but we’re awfully close to a technology that in many ways seems more difficult to achieve: the self-driving car.

If you’re in the San Francisco Bay Area, in fact, autonomous vehicles are already a part of life, said Sean Smith, a computer science professor at Dartmouth College.

“A friend of mine works out there. He finds it very hard when he’s at an intersection and the other car is a Google car. He’s never sure: What is this car going to do to me on a bicycle?” Smith said.

Come to think of it, however, bicyclists feel the same uncertainty when there’s a person behind the steering wheel, so perhaps that’s not a good example of autonomous cars changing life.

But change life they will, which is why Science Cafe Concord will ponder them tonight.

Smith will be one of three panelists available to answer questions about self-driving cars at the monthly Science Cafe Concord, upstairs at The Draft Sports Bar, 67 S. Main St., starting at 6 p.m. Show up early if you want a table; last month’s event was standing-room only.

I’ll personally buy your dinner if you arrive in a self-driving car. As long as I get to ride in it, that is.

Also scheduled as panelists are Andrew Kun, UNH professor of electric engineering, and Joe Cunningham, professor of the Robotics and Automation Engineering Technology program at NHTI in Concord.

Smith’s expertise focuses on the ancillary effects of the technology, as you’d expect from his job as director of Institute for Security, Technology and Society at Dartmouth. Last summer, for example, he ran a class on the risks inherent in the Internet of Things, which includes self-driving connected cars.

“As computers go into lots of previously non-computerized things in our daily life, like autonomous vehicle, what are the impacts of society?” he asked.

That’s a really interesting question, and it isn’t going to be answered tonight, but I bet we’re going to gnaw on it a lot.

Speaking as somebody who read science fiction in the 1960s, I think the most interesting thing in the development of autonomous cars is how different they are from the way they were envisioned by Robert Heilein and his peers.

Back then, the assumption was that self-driving vehicles would be controlled by sensors and computers and signals in the highway itself, designed and run by central authorities. What has happened instead, as we know, is that they are being controlled by sensors and computers within the vehicle itself, without any change in the roads.

This is a bottom-up design, instead of a top-down one, that echoes the way much of technology is proceeding, the internet being the prime example. As a rule, technology doesn’t wait for complete blueprints with the bugs worked out, it just plunges ahead and fixes things as it goes.

Is that a good idea when you’re talking about 2-ton pieces of metal moving at 60 mph? Maybe not, which is another thing we may want to discuss.

“Technology has rolled out before it’s ready in relation to cars,” Smith said, telling the tale of a friend who rented a car with anti-theft provisions under which it could only be started when there was a cell phone connection to a database that confirmed the identity of the driver.

Alas, Smith said, the friend stopped along Interstate 89 between Concord and Hanover and found “he was not able to start the car because the car had no cell phone coverage.” He had to be towed to a place where a cell phone would work.

That’s pretty funny (although I bet the friend didn’t laugh much), but unintended consequence in a self-driving car might be less amusing.

Pin It on Pinterest