Jump to content

Self driving cars.


Recommended Posts

http://www.theatlantic.com/technology/archive/2014/08/if-a-self-driving-car-gets-in-an-accident-who-is-legally-liable/375569/

Self driving cars are a great addition to the driving industry, but I still have my doubts.

Do you guys trust self driving cars?

Edited by Chester199

"I'm a marked man, so I'm getting out of here"

 

Ray Machowski

Link to comment
Share on other sites

I dont care if a car drives itself, if it has good mileage, if its enviormentally friendly, if its compact, if its solar powered, im not driving an ugly piece of shit like that. Thats my thought on this wave of smart cars and other things like that. But as for me trusting it, no. Is america this lazy? They need a car to drive itself? I thought automatic transmissions were lazy enough..

Link to comment
Share on other sites

I guess should have intelligent traffic on highways or roads that traffic is heavy. For example when the driver on the freeway, the system automatically runs (pilot automatic) if the driver activates the autopilot. Now when he goes to regular transit route (eg residential) he has to drive. Like the pilot but now is for car.

Edited by Hayran
Link to comment
Share on other sites

It would have to have manual drive capabilities though, because likely it autodrives on a gps, right? So what if the lane merges on the highway, but that data isnt on a gps. You would have to drive yourself, also not sure how it drives without crashing into other cars - But im assuming sensors?

Link to comment
Share on other sites

It would have to have manual drive capabilities though, because likely it autodrives on a gps, right? So what if the lane merges on the highway, but that data isnt on a gps. You would have to drive yourself, also not sure how it drives without crashing into other cars - But im assuming sensors?

Self-driving cars do not use stored maps for small scale tasks (like "is there a merge lane"); they do that using cameras that detect road markings, and detecting what other cars are doing.

Link to comment
Share on other sites

We all know that eventually every computer breaks. So in about 20 years, you'll be driving along and suddenly one of the "Self Driving" death machines in front of you slams the breaks for no reason causing a massive pile up on the freeway (motorway for our British friends) that kills 300 people. Or maybe it mistakes a sidewalk for a road and runs over children walking home from school. Horrible idea.

-Proud Texan-

Link to comment
Share on other sites

We all know that eventually every computer breaks. So in about 20 years, you'll be driving along and suddenly one of the "Self Driving" death machines in front of you slams the breaks for no reason causing a massive pile up on the freeway (motorway for our British friends) that kills 300 people. Or maybe it mistakes a sidewalk for a road and runs over children walking home from school.

So true.

Link to comment
Share on other sites

It could be potentially safer. But will mostly shine only when 99% of vehicles on the road are automated, something like in iRobot. Each vehicle communicates with each other, basically there wouldn't be much need for traffic lights & junctions.

 

But I still would prefer a mix of human & automation. I know there's a shitton of dumb drivers out there, so as long as the automation kicks in only to prevent accidents, I'm ok with it.

Plus I love driving & can't wait for my license :3

 

Also, like Steele said, if it ever breaks or becomes faulty, then it's going to turn a routing lounge (since you aren't driving, you can expect people to be on the phone/laptop/magazine) into a sudden roller-coaster of terror & possibly death.

EDIT: Added a space where I forgot to add :P

Edited by Rocking_Star101

If my post made you laugh (or giggle) in any way, smash that blue grey "Like this" button & like it :smile:

If my answer to your Support Thread fixed your problem, please Vote it up by clicking the ^ (up) arrow to the left of my name.

Link to comment
Share on other sites

Wouldn't happen, or would be implemented as a hybrid option even more so than it already is. One, a lot of people enjoy driving. Also, I would imagine there's WAY too many variables to take into consideration. Weather, construction, sudden car movements that are directly in your way, destinations being known to the computer on board, should it take the shortest route or the quickest route and choosing on the fly plus detours, etc. You would have to convert an entire area over at once. Trying to have even 100 self driving cars on the road with 1000 human drivers would be disastrous, not to mention all the variables I mentioned. Only way I could see it happening, is if, say, LA or NY converts, at once and entirely, to "Self driving cars ONLY". Then maybe it would work. Although I don't see the Gov't shelling out billions for a science experiment that may not catch on to the general public. 

 

Either one of two ways would be more plausible. If they implement the technology in existing cars now, even more so now that "accident avoidance". That could work although how cost effective would it be.

 

The best way to me is converting all city transit to self driving. Like 1 or 2 big buses maybe. 

 

I hope I'm not alive whenever it happens. I don't want to live in the world of iRobot.

 

Although how funny would it be to see a self-driving cop pull you over and spit out a computer generated ticket?

detroidmpt.jpg

Edited by Pavelow
Link to comment
Share on other sites

We all know that eventually every computer breaks. So in about 20 years, you'll be driving along and suddenly one of the "Self Driving" death machines in front of you slams the breaks for no reason causing a massive pile up on the freeway (motorway for our British friends) that kills 300 people. Or maybe it mistakes a sidewalk for a road and runs over children walking home from school. Horrible idea.

Because people never ever do that, and the traffic fatality rate is notably low right now? If the car in front of you slams on the brakes and you hit them, it's your fault anyways - if you're keeping proper distance, you can stop before you hit them. In this case, either a person or a self-driving car can *notice* the car slamming on its brakes, slam on their own brakes in response, and so on down the line. With sidewalk vs road, it's not as though people never drive onto sidewalks, or as if sidewalks are in any way similar to roads visually, or as though these cars don't already have sensors to detect pedestrians and avoid hitting them anyway (they do *not* assume "this is a road, I can drive on the road and not worry about anything"; they actively look for obstructions, can detect pedestrians in the middle of the road [far more likely to happen than a car going on the sidewalk], and are able to stop before hitting them). You seem to be under the impression that one issue will lead to catastrophe, which is simply not true - if it were true, we'd be seeing 300-car pileups right now, and swarms of pedestrians getting run down right now.

Yes, there will be bugs, and some people will die due to software issues with self-driving cars. That will happen. But it's intellectually dishonest to say "X will kill some people, so Y is better" without even looking at the outcome of Y, which in this case is many, many deaths. If one self-driving car has an issue and kills 10 children walking home from school, but large-scale adoption of them reduces traffic fatalities by one percent (by 2012 numbers, over 300 people in the US), then self-driving cars win by a huge margin. The 10 people killed in one incident hits the news, because you can look at the pictures of the victims, while the 300 people who don't die don't show up in any newspaper anywhere (even noticing it requires actually looking at traffic statistics), but that's why you simply cannot validly decide which of two alternatives is safer without taking a broader view and looking at overall statistics -- if you rely on your gut feeling, you'll undercount common incidents killing only one or two people, and overcount dramatic and rare incidents.

Link to comment
Share on other sites

Wouldn't happen, or would be implemented as a hybrid option even more so than it already is. One, a lot of people enjoy driving. Also, I would imagine there's WAY too many variables to take into consideration. Weather, construction, sudden car movements that are directly in your way, destinations being known to the computer on board, should it take the shortest route or the quickest route and choosing on the fly plus detours, etc. You would have to convert an entire area over at once. Trying to have even 100 self driving cars on the road with 1000 human drivers would be disastrous, not to mention all the variables I mentioned. Only way I could see it happening, is if, say, LA or NY converts, at once and entirely, to "Self driving cars ONLY". Then maybe it would work. Although I don't see the Gov't shelling out billions for a science experiment that may not catch on to the general public.

You realize that Google is already sending self-driving cars into city traffic? And that it's legal to operate autonomous cars in Florida, Nevada, and California? It's not just a proposal, it's actually happening. It is hard to keep track of things like sudden car movements, detours, etc., but they are getting pretty good at it (they started on highways, which are much easier to handle, but have now advanced to handle city streets). While many of the things proponents say about them would only work if all cars were self-driving, Google's prototypes are made for an environment surrounded by normal cars and pedestrians. It's not even remotely disastrous, and has only been associated with two collisions: in one, the car was being driven manually (so far, autonomous cars have to have manual controls as well; Google's working on a version with just an e-stop, but I don't think it's being used yet), and in the other it was rear-ended while stopped at a light, which is generally considered the fault of the driver in the rear car.

The best way to me is converting all city transit to self driving. Like 1 or 2 big buses maybe.

This is probably the best application - one of the current main uses of autonomous vehicles is rapid transit systems (which are wholly grade-separated, so it's rather simple to be automatically controlled). Labor is one of the main costs of public transit, and eliminating it lets frequency stay high 24 hours a day.

I hope I'm not alive whenever it happens. I don't want to live in the world of iRobot.

Too late; it's already being done as a prototype on real city streets.
Link to comment
Share on other sites

Because people never ever do that, and the traffic fatality rate is notably low right now? If the car in front of you slams on the brakes and you hit them, it's your fault anyways - if you're keeping proper distance, you can stop before you hit them. In this case, either a person or a self-driving car can *notice* the car slamming on its brakes, slam on their own brakes in response, and so on down the line. With sidewalk vs road, it's not as though people never drive onto sidewalks, or as if sidewalks are in any way similar to roads visually, or as though these cars don't already have sensors to detect pedestrians and avoid hitting them anyway (they do *not* assume "this is a road, I can drive on the road and not worry about anything"; they actively look for obstructions, can detect pedestrians in the middle of the road [far more likely to happen than a car going on the sidewalk], and are able to stop before hitting them). You seem to be under the impression that one issue will lead to catastrophe, which is simply not true - if it were true, we'd be seeing 300-car pileups right now, and swarms of pedestrians getting run down right now.

Yes, there will be bugs, and some people will die due to software issues with self-driving cars. That will happen. But it's intellectually dishonest to say "X will kill some people, so Y is better" without even looking at the outcome of Y, which in this case is many, many deaths. If one self-driving car has an issue and kills 10 children walking home from school, but large-scale adoption of them reduces traffic fatalities by one percent (by 2012 numbers, over 300 people in the US), then self-driving cars win by a huge margin. The 10 people killed in one incident hits the news, because you can look at the pictures of the victims, while the 300 people who don't die don't show up in any newspaper anywhere (even noticing it requires actually looking at traffic statistics), but that's why you simply cannot validly decide which of two alternatives is safer without taking a broader view and looking at overall statistics -- if you rely on your gut feeling, you'll undercount common incidents killing only one or two people, and overcount dramatic and rare incidents.

 

Well then here's my next problem. The United States government will most likely start taxing people who don't have these automated things because they'll say we're somehow more of a road hazard (never had an accident). It will turn into a huge money scheme. Of course, we could argue that this would happen with any new technology.

 

But the difference between myself and an automated machine is I have a moral conscience. If I'm about to plow into a brick wall at let's say 45 mph, and to my left and right are people walking on the street, I can make the decision (and would) to apply my brakes as best I could but avoid turning so I don't kill other people even if this means my death. The machine on the other hand will simply avoid the wall and possibly kill everyone else because it thought it was the right thing to do. I don't mind something being automated when it's on a track. But if it has free will to move around at high speeds, I just don't think it would be a good idea. Maybe for the first few years, but when the machines break as all machines do, it will be chaos.

 

Also, think of this. Suppose you happen to be in a very dangerous neighborhood when all of the sudden someone points a weapon at you and is threatening to kill you. In a normal vehicle you can quickly change direction and get out of there (though still dangerous) but in the automated vehicle, you'll just sit at that red light while he takes everything you have. Can you imagine how easy it would be to start kidnapping people? Especially at night. Just get them at a red light, the car won't know the difference. Then you could say it has auto locks. Well what if it catches on fire? Now you're locked in a burning vehicle. There's just no way in Hell I'd get in one of those things.

 

-Proud Texan-

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.




×
×
  • Create New...