Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Can a single car break a traffic jam?

bbc.com

Post reply on HN
Print view View original

61–70 of 108 posts

Re: Can a single car break a traffic jam?

#61

Post by 0xfeba »

serge2k wrote: post #37
logn wrote: post #15

I often try the technique of leaving lots of space and going at a slower, steady pace. One problem is that everyone re-routes around you, and you induce road rage and traffic weaving. The other problem is that unless it's a straight stretch of road, you have no idea what speed is slow enough.

> One problem is that everyone re-routes around you, and you induce road rage and traffic weaving Probably a good indicator you are going too slow. Which is okay, just please, please, PLEASE stay in the right lane.

> Probably a good indicator you are going too slow. Which is okay, just please, please, PLEASE stay in the right lane.

Not necessarily. It dense, slow traffic, short-sighted drivers see open spaces; they rush for it, then slam on their brakes. It's amusing the frustration that this generates--for no real reason other than "I'm mad and going to tailgate you because you aren't tailgating the person in front of you".

If people swoop in front of me, Oh well. It's a pittance on my overall commute time, if effecting it at all. Generally it's just limited two a few very aggressive drivers, and I'd rather they tailgate the person in front of me rather than me.

Re: Can a single car break a traffic jam?

#62

Post by alphapapa »

bagacrap wrote: post #30
logn wrote: post #15

I often try the technique of leaving lots of space and going at a slower, steady pace. One problem is that everyone re-routes around you, and you induce road rage and traffic weaving. The other problem is that unless it's a straight stretch of road, you have no idea what speed is slow enough.

The other problem is that this isn't scalable. Imagine a traffic jam that stretches for miles. Now imagine that same jam where everyone has left an extra three car lengths in front of them. You're suddenly using a lot more road per car. Tailgating actually seems like the most efficient use of space to me, provided you tailgate effectively (don't yo-yo).

Not only is tailgating unsafe, but it's not possible to do it effectively. Even if you are a super-skilled racing driver, the driver ahead of you isn't, and he will slam on his brakes at the last second, forcing you to do the same, forcing the car behind you to do the same...rinse and repeat. Or you will hit him, or the car behind will hit you, etc, causing another accident and another traffic jam.

That's how the traffic jam happens in the first place. An accident happens, closing a lane, but the impatient people in that lane don't merge until the last second, forcing the people in the next lane to slow down quickly. Some people in the next lane don't leave room for cars to merge, forcing them to merge later, forcing them to do so at a slower speed, slowing down all the cars behind them in a cascading wave of brake lights.

What you're imagining might be possible with cooperative, AI-driven convoys of cars...until a black hat breaks in to their software over the Internet and makes them crash.[1]

What helps is to provide a buffer in your lane by moving at a consistent speed, avoiding braking altogether. Of course idiots are going to get mad and zoom around you once in a while, and that's fine, because they will clear space behind the wave, allowing traffic behind to move more smoothly.

The bottom line is that you can't control what other drivers do, and many drivers are stupid. You can control what you do, leaving more space, maintaining safe following distance, leaving room for cars to merge easily ahead of you, increasing your gas mileage by not braking, etc.

1: http://www.wired.com/2015/07/hackers-remotely-kill-jeep-high...

Re: Can a single car break a traffic jam?

#63

Post by brobinson »

arprocter wrote: post #46
noonshine wrote: post #12

I have adaptive cruise control in my Subaru Legacy and I have often wondered what will happen when more people have ACC. Or even if I were following another car exactly like mine - would the lag time grow exponentially with more intense stop/go or would it even out? Or does it depend on how well the car implements it?

There was an Acura ad recently which said the car would change your route if it knew there was a jam ahead. I thought 'what happens if the ad is so successful everyone buys that Acura?'

Then people complain about rising traffic through their neighborhoods: http://www.cnbc.com/2014/12/11/la-residents-complain-about-w...

Re: Can a single car break a traffic jam?

#64

Post by seanp2k2 »

0xfeba wrote: post #60
notlefthanded wrote: post #52

Earlier quoted context omitted.

Actually it's easier to manage speed and spacing in congestion with a gently curving stretch of road - cars further up are visible so you can anticipate better and slow down earlier/more gradually, reducing jerk. This way too, you won't seem slow to the people behind you. It's not so much going super slow as it is being able to correctly guess the average speed of the road and acting as a damper. This is basically wh…

> I also like to try and impose a rule of not touching the brakes unless absolutely necessary, and using 20% throttle input at most, that helps with the whole dampening thing. I do the same. It's funny the frustration this seems to cause other people. Frustration from nothing, if they'd think about it.

"But I want to go fasterrrr! Why doesn't he goooooooo?!?!?" One time, a man and a lady in a Harley Davidson edition F150 got so mad at my for doing this that they were both screaming at me through my open sunroof, threatening to get out and physically assault me etc. They kept yelling at me to "learn how to drive". It as grid-lock traffic and we'd have minutes at a time of complete stoppage. That was a pretty awkward hour of my life.

Re: Can a single car break a traffic jam?

#65

Post by seanp2k2 »

arprocter wrote: post #51
logn wrote: post #15

I often try the technique of leaving lots of space and going at a slower, steady pace. One problem is that everyone re-routes around you, and you induce road rage and traffic weaving. The other problem is that unless it's a straight stretch of road, you have no idea what speed is slow enough.

A game I used to play with myself on the highway was to simply try and use my brakes as little as possible - when you see brake lights ahead it's instinctual to also brake yourself, but often if you have a long enough follow distance you can just lift off the gas. Brake lights being a binary representation of an analogue pedal is another problem, although iirc on some newer cars the lights are brighter/more lights go…

IMO if you have to brake on the freeway, someone made an error.

Cool about brake lights; I've wondered for years why they don't do this. I knew about some newer cars flashing the hazard lights in a very hard stop, but didn't know any were doing brake light meters. Any idea which specific models? I'd love if they standardized the messaging across brands and countries; imagine a new mustang where 1 vertical bar on each side means light braking, 2 means moderate, and 3 + rapid flashing means very hard braking. Not sure what standard would make the most sense for this, but the sooner, the better, so new cars can adopt it. It's way less useful / less clear when each car does it differently.

Re: Can a single car break a traffic jam?

#66

Post by mjevans »

No, at least not for I-405 / I-5 near Seattle.

These roads, due to a complete lack of effective urban planning and development, are several times over capacity.

I like to think that this heuristic would be effective for such cases.

* Aim for a hard speed limit (maybe the actual speed limit).

* Actually obey the law of this state: Keep Right Except to Pass.

* ALWAYS allow merges (from either side) with higher priority.

Edit:

After reading the github link from one of the other posts I want to expand why I disagree and suggest always allowing merges.

It is to allow traffic to leave the freeway (merge right to exit) as well as to enter the freeway (merge left, mostly to enter at all, but also in case they're going a long distance or need to get to special use lanes on the left).

Re: Can a single car break a traffic jam?

#67

Post by rev_bird »

labster wrote: post #50
bagacrap wrote: post #30

Earlier quoted context omitted.

The other problem is that this isn't scalable. Imagine a traffic jam that stretches for miles. Now imagine that same jam where everyone has left an extra three car lengths in front of them. You're suddenly using a lot more road per car. Tailgating actually seems like the most efficient use of space to me, provided you tailgate effectively (don't yo-yo).

You're proposing changing the traffic pattern to look like a liquid instead of a gas. Guess which state has a higher viscosity? If we all computerized cars that introduced 0 stochastic motion, this might work. But if one deer gets a dumb idea you're going to have a multi-kilometer pileup.

>liquid instead of a gas

I'd never have thought of it this way, but it's so simple. Thanks.

Re: Can a single car break a traffic jam?

#68

Post by mjevans »

I want to propose a second more radical suggestion.

I think pacer cars might work very well for phantom traffic jams, but I very much disagree about /how/ they should be used. Instead of encouraging an over-capacity jam, I believe that the pacer cars should expressly communicate to other traffic something along the lines of.

"Temporary" / "Speed Limit" / "Follow at XX"

On a rear message board.

The pacer car would then draw out the stuck traffic in to the space /ahead/ of the jam and encourage the compression wave to expand to the front instead.

Re: Can a single car break a traffic jam?

#69

Post by gerbilly »

arprocter wrote: post #49

I wonder how much of this is mitigated by good lane discipline? My SO actually got pulled over the other day (or at least that's what the cop used as an excuse) for driving in a passing lane when not passing anything. I said I was glad because it's a terrible habit - every lane ending up going the same speed as the slowest driver

>every lane ending up going the same speed as the slowest driver

And if two drivers drive abreast, then it's like having two roads going to the same place, where no one can pass anyone.

Re: Can a single car break a traffic jam?

#70

Post by SwellJoe »

j_s wrote: post #24
SwellJoe wrote: post #20

Earlier quoted context omitted.

Not only that, AI will eliminate huge amounts of wasted road even during peak times and with traffic that can't be scheduled for off hours (like people going to work/school). A networked self-driving car can know where the traffic is and route around it. It can provide tools to everyone (from city planners to individuals) to maximize the availability of resources and minimize the time wasted on waiting. Red lights ca…

Yes, all (but one?) lane going in the "to/from work" direction based on traffic needs will make a huge difference!

Theoretically, one doesn't even need directional lanes, anymore. You just have a road, and cars go where they most efficiently fit; lanes won't have any meaning anymore, just like most road signs won't have any purpose. If every car is communicating with every other car, and every car knows where it needs to go and how to get there in the shortest time, it can look like utter chaos to the human eye, and still be entirely safe (assuming we go to great lengths to insure reliable software and hardware).

Of course, I expect it to take decades for us to reach that point, which is really unfortunate. That's a lot of senseless auto collision deaths due to human error while we wait for the future to arrive. If governments were capable of looking forward, at all, they would recognize the massive savings an all AI road represents, both monetary and health, and would be pushing hard for it (both enabling private industry to implement it and guiding public funded research).

Post reply on HN