Thursday, January 3, 2013

Cars That Drive Themselves - Google Has Big Plans

Google self-driving car on 680 with me today
I've been very interested in the stories I've read lately about Google's self-driving cars. I've thought for a long time that traffic problems could be solved by computers better than by building roads. What causes traffic isn't that there are too many cars, it's that the cars aren't operated in a synchronous way. You could have a mile long line of cars, and if the first car and the last car and every car in between started and stopped at exactly the same moment, there would be no "jam". What happens now is that each car waits for the car in front of it, there is slack in the chain, and it takes a long time for the message to get back to the last car.

Well, we're nowhere near having computers in every car but Google is well on its way to developing cars that could do exactly what I described: coordinate their driving so as to lessen or even eliminate traffic jams. So, impressed as I am with this project, I was surprised today when I read an article today at about Google. In it, I read this:
"Self-driving cars are still a small project within Google that many outsiders, including investors, think of as little more than a flashy, fanciful science experiment."
Really? How could an investor not see it as a serious product? Google's self-driving car is not yet a commercial product, but it's also not a laboratory experiment. Self-driving cars are navigating California roads today.

Google CEO Larry Page has an interesting way of describing an attribute an idea must have to be worthy of Google's attention. He says it must pass the toothbrush test. It must be something we use at least twice a day. That's a clever way of saying a product concept must solve a large enough problem.

Well, tens of thousands of traffic deaths and the time wasted in traffic jams is a pretty significant problem! As I said, laying concrete isn't going to solve the problem. Think about it, if more concrete was the solution to traffic problems, Los Angeles would be traffic free by now! But that's not doing anything to solve the  problem. It's some kind of law: Cars proliferate so as to fill the available pavement.

City of freeways, traffic, and smog

Is there another company that's as far along as Google is in developing self-driving cars? Not that I'm aware of. If I were a betting man, I'd bet on self-driving cars on the road easily within ten years.