Some sort of half-local/half-global optimization algorithm running in the "smart grid" would be a good idea.
I'm a big fan of the "green for a few seconds then flashing yellow" arrow.
When I lived in Tucson AZ almost every light was set up for lagging left turns. Where you can go when it's safe, but in the event of heavy traffic you'd be able to count on having a green arrow to empty the left turn lane - worked great!
You can't demand and reward excellence with car driving, because by assuming that people are all highly competent, you'll end up with lots of fatal crashes because of all the people who aren't competent. You can only assume car drivers are competent in societies where there's plenty of options for people who don't or can't drive, and America made the decision to not be such a society, so everyone has to be able to drive, even if they're an idiot.
Sorry, but that's completely impossible. You simply cannot have a spread-out, car-based society with "reasonable alternatives" for people who can't drive, unless perhaps you're going to heavily subsidize Uber and Lyft. Buses and trains don't work well in suburbs; trains only make sense with enough density to have enough ridership, and buses only make sense to fill in where train lines don't exist. In the suburbs, people stuck taking the bus get penalized with ridiculously long transit times, because the bus takes a long, circuitous route, only comes once or twice an hour, etc. That isn't a "reasonable alternative" when people with a car get get there in 10 minutes or less.
So you have two choices: a society like Germany's where at least in urban areas it's easy to get around without a car, and then you can have a very high competency requirement for car drivers; or a society like America's, where the whole society is designed for cars (except for a small handful of urban cores) and you have to have a car to achieve a middle-class existence, so you cannot have a high competency requirement because otherwise you are restricting peoples' right to movement and travel.
The way it is now (with the sprawl that we currently have), yes, due to physics. You can't change physics. There just isn't any way to make a bus a viable transit alternative in a sprawling suburban area; everything is just too far apart.
>Why can't we begin to restructure our society to be more like, in your example; Germany, so that it is possible?
Well I wasn't really talking about long-term, I was talking more short-term, with the assumption that the sprawl we have will remain or be extended. I simply don't see any attempts in this country to increase density, except perhaps the "urban growth boundary" around Portland.
Sure, I'd love to see this country decide it wants to be more like Germany or Japan and stop all new sprawl, and instead require higher density development near city cores, so that public transit is more feasible, and also provide more funding for transit construction (since that usually drives density: people frequently want to live close to the subway stations). But there's a lot of fanciful things that I'd like to have; that doesn't mean very many people agree with me, and I don't see much agreement with me on this issue in America today, at least not from the voters.
I find that an odd perspective on successful public safety rules and policies with the aim of reducing serious injury and death during a dangerous activity.
Specifically related to driving: I believe that the licensing requirements should be slightly higher and that more should be done to guard against bad driving and not just illegal driving. E.g. I've never seen anyone get a ticket for getting on the highway going a third of the speed of traffic, but that's certainly dangerous. Instead, people get tickets for exceeding a limit set by the capabilities of the lowest common denominator by a small percentage regardless of conditions (often not particularly dangerous - not an excuse for reckless driving, of course).
I simply think that there are better solutions. For one, public transportation.
In other states, like Washington, the red arrow is equivalent to the red circle. We even have left-turn-on-red, so (unless you’re stuck behind an out-of-towner) this tends to be much less of an issue.
(For instance, Kentucky is no left-turn-on-red [either red arrow or full red; mostly the state doesn't use red arrows because they'd be redundant in that case], but allows for left-turn-may-yield on flashing yellow/amber.)
https://devblogs.microsoft.com/oldnewthing/20181101-01/?p=10...
https://app.leg.wa.gov/rcw/default.aspx?cite=46.61.055
I've never done it myself and don't intend to start since I think it's a bit obscure and I don't want to argue with a cop about it if I get pulled over.
It took a while to get used to because in the past there simply just won't be any signal, to imply you can go if you want.
It's a separate light to make the flashing-yellow to solid-yellow transition clearer.
I always thought this was normal. If the right turn arrow is green you have right of way, if it's red you must stop, and if it's not illuminated you give way to oncoming traffic and turn when safe.
At that scale trying to do anything clever will cost billions. Most red lights don't even detect motorcycles standing right on top of their sensor, we're decades away from smart lights detecting your car speed/distance and computing the most optimised yellow light timing.
Also, in most cases you should be able to roll off the throttle and coast to the light without having to aggressively brake. If it's not the case the yellow light is not timed well and you can contact the company handling the roadwork in your city.
I.e. if you at or very close to the line you prepare to stop but if your very close you proceed carefully and make the turn.
(Anyway, in most places the fines from the automatic cameras do not go to the same administration that decides the speed limits and places the traffic lights, so in the end the common accusation of "they only want to make money from it" doesn't make that much sense)
He meant jurisdictions in America. Local governments here are infamous for shortening yellow lights. The companies that make automated cameras even have it in their contracts that the cities must do this.
Citation?
https://www.digitaltrends.com/cars/red-light-camera-controve... "The contracts in place with red-light camera providers often specify maximum yellow light times, and impose financial penalties if the city lengthens the yellow period."
More: https://slate.com/news-and-politics/2014/10/chicago-shortens... https://www.chicagotribune.com/news/ct-red-light-camera-yell... https://www.motorists.org/blog/6-cities-that-were-caught-sho...
There's a lot of room for weaseling the way that's written. The contracts could specify maximum times to cover incorrect configurations like 2 minutes - that would be sensible. Technically the article is correct but it leads the reader to conclude malicious intent where there may be none.
Only the DigitalTrends article talked about contract imposed maximums. In the Chicago article it looks like the city that was acting in bad faith. There is no established link between provider imposed contracts violating best practices.
Also, checking behind you has minimal value for drivers. Rather than any pattern it should be your absolutely last priority as it very rarely provides useful information and may be dangerous. That said, you should maintain basic awareness behind you.
...unless you are entering a situation where you may need to decide if stopping is preferable to some other option.
In one case, I’d say checking my rearview was worth thousands of dollars and maybe three lives. When approaching a crosswalk and reducing my speed to stop for pedestrians I saw clearly the guy behind me wasn’t going to stop (he maintained his speed and was looking down, probably texting). I honked and flashed to warn the pedestrians who stayed out of the crosswalk. Then I took my foot off the brake as the other car hit me. The damage was minimal because I just let him push me. I pulled into a nearby parking lot. The other guy took off without stopping. It was a new car so no rear plate.
If I hadn’t looked I’d have tried to stop, been pushed into the pedestrians anyway unless they noticed on their own, and my car would likely have been totaled.
Also, talking your foot off the break increased the risk in that accident. If the driver had swerved at the last second they could have sent your car into the pedestrians you where trying to protect. In a rear end collision momentum pulls your foot from the peddle. So, you want to be hard breaking at the moment of impact to minimize the distance you move post collision.
Honking may have been useful, however unless you gave then ~2+ seconds of warning it was probably futile. Walking speed may be 4.5 feet per second, but acceleration and reaction times are not instantaneous.
No one suggested doing anything of the kind.
Regarding my collision, you're just making stuff up.
That was the OP situation.
> Oh, and I just completed my regular mirror check, and so I don't know how long the yellow has been up. Do I stop or continue on?
As to breaking, or saving the car:
https://www.cartalk.com/content/bracing-impact-it-better-bra...
On it’s own that’s not going to make a huge difference, but I have seen recommendations to apply the parking break for similar reasons.
> Oh, and I just completed my regular mirror check, and so I don't know how long the yellow has been up. Do I stop or continue on?
These statements are not describing the same activity.
> As to breaking, or saving the car:
Your link agrees with me. Read Tom's last paragraph. There was no traffic in front of me. The pedestrians were safely off the road because I honked and flashed. In the case that the pedestrians had kept going, I'd have done things differently. But it doesn't matter; because the point isn't whether braking or not braking was the correct decision.
The point is that by looking in the rear view mirror I became aware that a decision needed to be made.
Yes they are. The window when the length of a yellow light matters is short. If check your rear view mirror right before the intersection then even if the light turns yellow that’s irrelevant you’re still going to cross. Alternatively, if you’re very far from the intersection, then not noticing a yellow light for ~2 seconds is irrelevant, you need to stop regardless of how long it’s been yellow at that point. The transition between those states is very rapid.
> The pedestrians where safely off the road.
There was still pedestrians around you. A rear end collision that’s directly bumper to bumper pushes you forward. But, it’s possible for someone to try and evade at the last second which means they and you go in some random direction, possibly onto a sidewalk or other objects near the road.
Examples of off angle impacts with solid barriers: https://www.cnbc.com/2019/03/21/iihs-crash-tests-slam-some-p.... Though if these where cars the car would be violently pushed in the opposite direction.
Various rear end collisions: https://m.youtube.com/watch?v=HD6QGx9d8ow. Which show a wide range of possibilities.
So, unless they where well behind you, picking your foot off the break was a poor choice for their safety.
No, they aren't. You misrepresented his point, and now you're changing the argument.
> So, unless they where well behind you, picking your foot off the break was a poor choice for their safety.
You're still making things up. Nothing you have linked here supports your claims. When a car collides with another car, things can happen with variable degrees of predictability. Yet, you are arguing with certainty that despite a successful outcome I made the wrong decision in a situation for which you don't have all the facts. I'm not sure how you hope any of this will convince me I did the wrong thing.
Perhaps ironically, even by arguing that there was a right thing to do, you implicitly acknowledge that your previous statement about rear view mirrors which I objected to was wrong.
No, I said his situation was avoidable and said how to avoid it, thus invalidating his point. Talking about how to safely drive drunk ignores the possibility to not get into that situation.
As to your situation, you where not responding to a traffic light. In my original post I specifically said maintaining awareness was import, just a low priority. Checking your rear view window while breaking is perfectly reasonable as long as your not maneuvering to avoid a collision. As described not checking the rear view minor would have likely resulted in a near identical outcome, honking may have have helped or been pointless we don’t know.
As to making things up, my statement is in response to the correct action on a rear end collision based on the information you had at the time. Saying it was a good idea using information unavailable at the time is like lottery winners saying they made the correct choice. However, as people can’t actually see the future it’s possible to make a poor gamble that pays off.
Similarly, you where lucky that the oncoming car did not swerve at the last second which you had not way of predicting or responding to as your limited by human reaction times. The first videos I showed demonstrated my point but I don’t think you understand the physics involved. Watch the first shot here. https://m.youtube.com/watch?v=JszLYsU1Duo On a side impact ball 1 moves to the left, and ball 2 moves to the right. Now, considering that rewatch the first video with that idea. If you where rear ended at an angle you could have been shoved into the direction of the sidewalk as the other car was shoved away. The second video I posed starts with a collision like with a car being shoved to the left, except it’s hitting a stationary school buss who’s breaks and higher mass avoid significant movement.
If you still don’t understand, I can only suggest learning some basic physics from online videos or whatnot.
PS: And really your argument about not breaking is balancing people’s safety vs an slightly lower repair bill. That’s really hard to justify.
I think I'll keep tracking who's behind me, how close, whether anyone new has appeared, whether anyone looks likely to pass me soon, and so on, because it's sure as hell been important in practice. Most notably when something goes even slightly wrong in front of me.
http://www.legislation.gov.uk/uksi/2016/362/schedule/14/made says: "(9) An amber signal, when shown alone, conveys the same prohibition as red, except that, as respects any vehicle which is so close to the stop line that it cannot safely be stopped without proceeding beyond the stop line, it conveys the same indication as the green signal which was shown immediately before it."
If you interpret this to the letter, it means that if there's no car behind you, you should indeed come to a screeching halt at an amber light. If it's "safe" to stop, then amber means stop.
Not saying I agree with this, but that's what the law says.
Maybe we're using "screeching" differently. If I need to jam on the brakes such that there is a chance the car my skid, or someone with a seatbelt might be injured, that is not "safe," and the law says I should keep going. Absent those things, I'm not screeching to stop but just stopping short.
Is there a definition of "safe to stop"? I would imagine a screeching halt wouldn't qualify.
Traffic simulations are generally quite complex. In your proposal, if your approach is not detected correctly, you either stop traffic for a car which isn’t there, or you never stop traffic for a car which is there.
It’s hard to imagine, given how bad traffic is, that our streets were planned by sophisticated computer simulations. There’s one place near me where it can take nearly an hour to go 3 blocks during rush hour, due to people blocking the box. Do simulations take that into account?
I’d be very curious to see how a blinking green system plays out in tests.
Another thought: if we manage traffic like regulating fluid flow and pressure, we could avoid situations where a single car exiting a parking lot shuts down the highway instantly. That single car should have to wait maybe 5-10 seconds before triggering the lights, and the blinking green would let highway drivers adjust. Forcing 20 cars to rapidly stop for an already stopped car makes no sense.
As the single car waits, pressure “builds” so to speak, and eventually the light decides that the fluid flow should change.
This bugs me. $50 of hardware can accomplish this, today. Raspi running vehicle detection code. If something is coming and the other way is clear, why not flip the intersection before it arrives?
$50 * 300 000 (quick google search about the number of traffic light in the US), That's not taking into account installation and regular maintenance.
I'm all for optimising things but sometimes it's just not worth the hassle. A well timed orange light is good enough.
That's only true if you want hardware that sometimes detects some kind of vehicles, in the right weather and from the right angles.
Truly reliably detecting this is much further away. The consequences of not getting it right are not as catastrophic as getting it wrong in a self-driving car, but they could still mess traffic up pretty badly.
Your reaction might be that it has been $50 for like 5 or 10 years now, and if so my response is that in this area, that is recent. It's a pretty slow-moving, late adopter industry. Most of the equipment getting installed probably hasn't been redesigned in years or decades. The people approving stuff want to buy from known vendors. The vendors are making money steadily now and don't want to go out on a limb with unproven tech.
Which doesn't mean it's impossible. I interpret it the opposite way: the fact that it hasn't happened yet doesn't mean it won't or can't. I would expect it to take this long or longer.
Also, I think the angle for getting this adopted isn't really efficiency, it's cost saving. Right now, some lights are on sensors (instead of just timing) but they do it with induction loops (https://en.wikipedia.org/wiki/Induction_loop). If you could match or surpass the reliability of those at a lower cost, you might get people to switch.
But once you've crossed that hurdle, you can improve efficiency too, of course.
Of course, you have to keep in mind that the camera has to be mounted in the proper position. So it's probably not really $50 of hardware unless you can (in most cases, at least) piggyback on something that already exists. Although that seems possible because the signal light does need to be visible by cars approaching the intersection.
One of these is really stupid, and one isn't.
That sounds like something a dashcam would be able to easily show.