4 ms·

Everything is give and take. Having self-driving cars would be an absolute godsend for the environmental impact of automobiles. Automated driving can easily be

Everything is give and take. Having self-driving cars would be an absolute godsend for the environmental impact of automobiles. Automated driving can easily be made more fuel-efficient than human drivers (especially when/if most cars are automated and can coordinate traffic plans), and would also open up much more radical car-sharing plans to minimize the number of cars on the road. Transportation might not be the biggest producer of CO2 emissions, but it's a pretty hearty slice, and unless we do something quickly our planet's pretty screwed. I agree that creating technology for the sake of technology isn't necessarily good, especially if there are negative ramifications in other areas (like job security). But this isn't a black and white "automated cars give us no tangible benefit other than cool factor" situation.

joering2 15y ago
Sure, let me send check to Al Gore. So far I haven't seen any reliable research that would reasonably explain how self driving cars navigated by central algorithms would help to unclog the streets. I for once would be upset if my car decided to put additional 15 miles to get me to work or wherever I want to go just "for a sake of common good of all other automobiles and streets being unclogged". Henry Ford II: Walter, how are you going to get those robots to pay your union dues? Walter Reuther: Henry, how are you going to get them to buy your cars?
- mryan 15y ago
  Many of the factors affecting traffic jams are a result of human drivers. This animation demonstrates two causes of traffic jams - 'waves' and zip merging: http://trafficwaves.org/tanim.html http://trafficwaves.org/tanim.html I think it is rather easy to see how a central algorithm would be much better at a) maintaining a safe distance from the car in front, and b) merging quickly and fairly. Route selection is just one of the things an algorithm could do. Traffic, by Tom Vanderbilt, has many interesting things to say on the subject: http://www.amazon.com/Traffic-Drive-What-Says-About/dp/0307264785 http://www.amazon.com/Traffic-Drive-What-Says-About/dp/03072...
  - joering2 15y ago
    should be a good read, thanks!
  - Someone 15y ago
    Also, the 'safe distance from the car in front' could be dramatically decreased. Reaction times of computers are way lower than those of humans, and also less variable. A car could signal that it is going to brake to a car following it with a meter of space between them, and brake a ms later, trusting the following car to do follow its speed decrease. Lower inter-car distance on highways will also mean lower air resistance, and hence, higher mpg. I doubt that the net effect on fuel consumption will be substantial, though. Chances are that people will want their driverless cars to drive faster, so that they can live farther away from work. IMO, the only thing that will help there is pricing (fuel, road use, etc)
michaelbuckbee 15y ago
I'm in agreement with you, but I still worry about second order effects. To me, the big one is that if driving becomes significantly less of a hassle will there be a corresponding increase in commuting / long car rides for vacations, etc.

joering2 15y ago

Sure, let me send check to Al Gore. So far I haven't seen any reliable research that would reasonably explain how self driving cars navigated by central algorithms would help to unclog the streets. I for once would be upset if my car decided to put additional 15 miles to get me to work or wherever I want to go just "for a sake of common good of all other automobiles and streets being unclogged". Henry Ford II: Walter, how are you going to get those robots to pay your union dues? Walter Reuther: Henry, how are you going to get them to buy your cars?

mryan 15y ago
Many of the factors affecting traffic jams are a result of human drivers. This animation demonstrates two causes of traffic jams - 'waves' and zip merging: http://trafficwaves.org/tanim.html http://trafficwaves.org/tanim.html I think it is rather easy to see how a central algorithm would be much better at a) maintaining a safe distance from the car in front, and b) merging quickly and fairly. Route selection is just one of the things an algorithm could do. Traffic, by Tom Vanderbilt, has many interesting things to say on the subject: http://www.amazon.com/Traffic-Drive-What-Says-About/dp/0307264785 http://www.amazon.com/Traffic-Drive-What-Says-About/dp/03072...
- joering2 15y ago
  should be a good read, thanks!
- Someone 15y ago
  Also, the 'safe distance from the car in front' could be dramatically decreased. Reaction times of computers are way lower than those of humans, and also less variable. A car could signal that it is going to brake to a car following it with a meter of space between them, and brake a ms later, trusting the following car to do follow its speed decrease. Lower inter-car distance on highways will also mean lower air resistance, and hence, higher mpg. I doubt that the net effect on fuel consumption will be substantial, though. Chances are that people will want their driverless cars to drive faster, so that they can live farther away from work. IMO, the only thing that will help there is pricing (fuel, road use, etc)

michaelbuckbee 15y ago

I'm in agreement with you, but I still worry about second order effects. To me, the big one is that if driving becomes significantly less of a hassle will there be a corresponding increase in commuting / long car rides for vacations, etc.