Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
Anyone who has flown in a small plane along the interstate knows that except for wrecks, most of the highway is empty space punctuated every few hundred yards by clusters of cars. and, it is in these clusters that most accidents occur. That is why the first step will be to give control of the interstate traffic to the roadway electronics. Our highways could handle two to three times the number of vehicles but for the insistence of drivers in clustering around lower moving vehicles.
Nah. No need. You just set up the behavior algorithm in the cars. And the cars cooperate with each other. That way you need little on the highway other than perhaps a communication capability to allow messaging of what is going on up ahead.
I remain of the opinion that over the road trucks are the first vehicle that will really get implemented. Simple rational. Best payoff. And they can afford immense cost per vehicle. Also potentially suitable for initial flock games where you convoy a whole mess of trucks with as single live director. Long term not needed but helpful in the starting phase.
Coming fast. At least some substantial examples prior to 2020.
Nah. No need. You just set up the behavior algorithm in the cars. And the cars cooperate with each other. That way you need little on the highway other than perhaps a communication capability to allow messaging of what is going on up ahead.
I remain of the opinion that over the road trucks are the first vehicle that will really get implemented. Simple rational. Best payoff. And they can afford immense cost per vehicle. Also potentially suitable for initial flock games where you convoy a whole mess of trucks with as single live director. Long term not needed but helpful in the starting phase.
Coming fast. At least some substantial examples prior to 2020.
Trucks need to look forward to lane exclusivity. Only truck driver entertainment justifies trucks traveling in auto lanes.
I remain of the opinion that over the road trucks are the first vehicle that will really get implemented. Simple rational. Best payoff. And they can afford immense cost per vehicle. Also potentially suitable for initial flock games where you convoy a whole mess of trucks with as single live director. Long term not needed but helpful in the starting phase.
Coming fast. At least some substantial examples prior to 2020.
A maturing economy is characterized by more autonomy in its transportation system -- for both personal vehicles and shipments of higher-priced, more time-sensitive goods. Our earliest freight transportation systems, canals and horses and wagons, moved at slow speeds though they allowed for a great deal of freedom of movement.
The railroad turned this balance upside down; the low coefficient of friction between iron wheel and rail allowed vehicles to move with much less effort, but limited grades to about 4 per cent, or a rise of one foot in 25. but since only movements in the same direction could occupy a given stretch of track at any given time, a method for keeping them apart had to be devised.
The earliest railroaders relied on strict schedules, and if either of two opposing moves were late at a scheduled meeting point, little could be done until the scheme went according to plan. After about twenty years, the invention of the telegraph facilitated communication at points along the route, and the first system of dispatching emerged.
After all-weather highways emerged post-World War I. truckers began to take high-value, time-sensitive freight away from the railroads precisely because they offered autonomy. The railroads were able to counter this trend after 1985 only because their rebuilt physical plant could offer reliability, rather than autonomy, but at much lower cost.
Progress in the development of "self-driving" vehicles would permit the re-imposition of some form of centralized control, but I can assure you that the development will be a very long process, and will begin in a limited number of venues to which it's best suited.
You might be able to read the news or catch a little more shut-eye on your daily commute one day, but only once you reach the on-ramp to the freeway; the people who think that something to take Gramps (who lives ten miles out of town on a highway that began as a footpath and was never re-engineered to a stage where gradients and curvature are a matter of written record) to the Doctor's office, are in for a letdown.
Last edited by 2nd trick op; 01-30-2015 at 11:14 PM..
You might (and I stress might) have trouble with irregular roads in a place like Boston where you have irregular roads combined with lots of cars on the road. A country road where you're unlikely to meet more than one or two cars should be well within the capability of a vehicle designed to handle a congested highway.
More than that though once you've freed up the driver to do something that is productive for him instead of driving the car could crawl along most of the time for all he cares. So if you're car has to severely ratchet down its speed to handle irregular roads, it won't be a problem most of people care about.
For what it is worth though I don't think Google is really shooting for driverless cars. After their purchase of the company that designed the Big Dogs for DARPA I have always suspected that they were aiming for something like the chevaline a from THe Diamond Age. I hadn't even read the book and a mechanical self driving horse was pretty much my first thought after seeing the BIg Dogs footage. Wish I had the money to build it because kissing the roads goodbye in their current state would be almost as good as no longer having to drive.
You might (and I stress might) have trouble with irregular roads in a place like Boston where you have irregular roads combined with lots of cars on the road. A country road where you're unlikely to meet more than one or two cars should be well within the capability of a vehicle designed to handle a congested highway.
More than that though once you've freed up the driver to do something that is productive for him instead of driving the car could crawl along most of the time for all he cares. So if you're car has to severely ratchet down its speed to handle irregular roads, it won't be a problem most of people care about.
I hope you're correct in this, but I have my doubts; rural roads which were never built to standards set by engineers have any number of quirks -- what our British friends call "deceptive bends". For example a curve may be either "banked" (lower on the inside, as on a racetrack), flat, or "crowned" (lower on the outside, making it easier to lose control).
And there's one on a rural road not far from my home with a strange combination of subtle changes in both the sharpness of the curve, and the gradient, with a very short straight stretch -- perhaps 50 yards -- thrown in for good measure. most locals know and respect it, but neophyte drivers and people from outside the area often lose control. Fortunately, the speed limits are tight enough that injuries are rare, but I fail to see how issues like that one can be addressed without a comprehensive database for such roads as part of the "self-driving" system
I hope you're correct in this, but I have my doubts; rural roads which were never built to standards set by engineers have any number of quirks -- what our British friends call "deceptive bends". For example a curve may be either "banked" (lower on the inside, as on a racetrack), flat, or "crowned" (lower on the outside, making it easier to lose control).
And there's one on a rural road not far from my home with a strange combination of subtle changes in both the sharpness of the curve, and the gradient, with a very short straight stretch -- perhaps 50 yards -- thrown in for good measure. most locals know and respect it, but neophyte drivers and people from outside the area often lose control. Fortunately, the speed limits are tight enough that injuries are rare, but I fail to see how issues like that one can be addressed without a comprehensive database for such roads as part of the "self-driving" system
You are missing the point. These things descend from machines that ran many miles over open desert with no charted roads. They actually made it. Though there was a lot of backing up and turning around.
The close in problem is really not significant. The vehicile maps everything close. It could not care what the road spec says...it looks and counts on its own mapping. The only one that is difficult is the pothole. If it does not have a record of it caution wil be required and it will approach the hole at a slow speed.
When in doubt it slows down. Doubt gets bad enough it stops. The basse algorithm is unbeatable...don't go where you are not sure.
The system will also use feedback. When a car transits a section of road it reports any significant decision from what it was expecting. After a reasonably short period of time the route is well reported. Then when something change the change appears almost immediately on the first exposure to it.
Snow and rain will have the standard problem it does with humans but the automatons will be less bothered. They have a well recorded base line. they will also be much better at driving on a slick or snow covered road. Human skill varies widely...the machines will be better than the best of the humans.
This will all have to unfold from increasingly sophisticated implementations. But it is actually simple art and will surpass the human operator almost from the get go.
Getting back to the point of the article, another issue that will affect congestion is that some people who currently leave earlier or later to work to avoid the traffic will now feel more comfortable commuting during the "proper" rush hour no matter how long it takes, further increasing the level of congestion.
In the end, reducing the spacing between vehicles is similar to adding a lane and will be subject to the well-known induced demand conundrum, in which increasing capacity increases usage, but does not decrease congestion. Plus the unanswered question of where we're going to park all of these automated cars.
So in the end, the best way to have "productive commuting" (and to improve commuting in general) is to improve mass transit coverage. If not automated rail, then automated busses with GPS tracking in dedicated lanes.
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.
Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
