![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
The cars of the future?
Aug. 20th, 2005 07:47 amPopular Science has a very good story on new technology being developed by automakers to increase the safety of vehicles. Nothing new there. But this is rather different. If you were to combine the technologies discussed in the article, you would have an ad hoc wireless data network flowing everywhere of vehicles sending information of road and traffic conditions to each other, vehicles would know when you're drifting out of your lane and activate a vibrator in the car seat on the left or right side (depending on the direction of drift), and cars could sense impending collisions and pre-tension the seat belts and reduce the speed of the car.
One interesting example was if I'm entering a highway, going up an on-ramp, and my traction control system kicks in. My car broadcasts an alert that includes GPS (and presumably highway) information. Anyone else turning to go up that on-ramp gets a slippery road alert.
I find that quite interesting. One very good thing is that the technology is available off the shelf right now and it seems to be developing along the lines of an interoperable standard, so it won't be a case of GM cars not talking to Fords. Each car may have different ways of dealing with the information, but they'll all transmit and receive the same data.
I see a couple of problems. In the example of the slippery on-ramp, how long does that information persist? Does each car that receives that info repeat it, if so, what is the time to live of the factoid/message? Does the info only get repeated if subsequent cars slip? How long does it last, if it's ice, it'll melt (presumably), if it's a gravel spill, it'll eventually be swept off. They're not talking about smart highways, just smarter cars.
But the main problem that I see is infoglut. I'm in downtown LA. I'm surrounded by literally a hundred thousand cars, all sending out information updates? First, the system is creating an RF horror. So much energy is being transmitted that you're going to have to look at interference with other radio services. When you have that much glut, you're going to have interference with each other, resulting in garbled signals. Easy enough to reject, but that takes processing power and time. And that's the biggie: how much processing power is going to be required for ultradense traffic areas like NY or LA or Chicago? All of these theoretical systems seem fine and dandy: another car reports or is involved in a collision ahead and your navigation system reroutes you. But when dozens of cars are reporting a slowdown, I could see some chaos creeping in.
I won't even begin to touch on a couple of issues regarding privacy and security, I'm just sticking to the technology side. The article says that car identifying information won't be sent, but you know that won't last as it would be so convenient to integrate automatic toll payment into such a system.
Anyway, pretty cool stuff. I don't know how much of it I'll see in my lifetime, but cool stuff.
One interesting example was if I'm entering a highway, going up an on-ramp, and my traction control system kicks in. My car broadcasts an alert that includes GPS (and presumably highway) information. Anyone else turning to go up that on-ramp gets a slippery road alert.
I find that quite interesting. One very good thing is that the technology is available off the shelf right now and it seems to be developing along the lines of an interoperable standard, so it won't be a case of GM cars not talking to Fords. Each car may have different ways of dealing with the information, but they'll all transmit and receive the same data.
I see a couple of problems. In the example of the slippery on-ramp, how long does that information persist? Does each car that receives that info repeat it, if so, what is the time to live of the factoid/message? Does the info only get repeated if subsequent cars slip? How long does it last, if it's ice, it'll melt (presumably), if it's a gravel spill, it'll eventually be swept off. They're not talking about smart highways, just smarter cars.
But the main problem that I see is infoglut. I'm in downtown LA. I'm surrounded by literally a hundred thousand cars, all sending out information updates? First, the system is creating an RF horror. So much energy is being transmitted that you're going to have to look at interference with other radio services. When you have that much glut, you're going to have interference with each other, resulting in garbled signals. Easy enough to reject, but that takes processing power and time. And that's the biggie: how much processing power is going to be required for ultradense traffic areas like NY or LA or Chicago? All of these theoretical systems seem fine and dandy: another car reports or is involved in a collision ahead and your navigation system reroutes you. But when dozens of cars are reporting a slowdown, I could see some chaos creeping in.
I won't even begin to touch on a couple of issues regarding privacy and security, I'm just sticking to the technology side. The article says that car identifying information won't be sent, but you know that won't last as it would be so convenient to integrate automatic toll payment into such a system.
Anyway, pretty cool stuff. I don't know how much of it I'll see in my lifetime, but cool stuff.
