Geometry is Killing us. Literally at one intersection.
 

For you guys out there, ever find yourself cycling along and notice an attractive woman walking toward you on the sidewalk with a tree between you and her? The tree is so close to the woman that as you cycle at say 15 mph and she walks at 2 mph, the tree will STAY between you and her foiling any chance you might have for a better look. Unless of course you, or she, alter speed.

There is a scientific phenomenon where this occurs with a cyclist (the woman), a car (the man), and the front pillar post of the vehicle (the tree). Only instead of resulting in frustration it often ends in tragedy.

If you like geometry, and want to learn how to avoid crushing or getting crushed at an intersection with clear lines of sight, then this is for you. Enjoy

Collision Course: Why This Type Of Road Junction Will Keep Killing Cyclists


» Collision Course: Why This Type Of Road Junction Will Keep Killing Cyclists

Is the tree transparent, or am I using some sort of towed-array hottie spotter?

There are good looking women all over the place. I'm more interested in this apparently unique tree that I can sort of see through.

Funny. Tree is solid. Woman is spied far up the road briefly, then disappears behind the tree forever. Even though both people keep moving at a steady rate of speed the woman and the tree remain in "eclipse".

Simple solution:

http://www.michigan.gov/images/mdot/...2_269926_7.jpg

This is a problem for pedestrians as well - I see it all the time.


That said, I am now firmly anti-geometry.





I am also against molecules.

There are many decent roadway design guidelines available.

All of them specify having intersections meeting at 90°, flat, level w/ unobstructed view. As any of those deviate from that ideal, then other changes are required to maintain safety, such as decreasing speed or adding stop signs.

The linked, unimplemented modification is okay, but a better solution would be to have modified both minor roads so that both met the major roadway at 90°, with the added benefit of being offset, to ensure the vehicles on the minor road come to a stop.

Too much math but a good looking woman is hard to beat

This is an important and worthwhile article for anyone wanting to improve his own safety on the road. I strongly recommend that people read the original in it's entirety.

While the car's pillar is a factor, these collisions also happen at sea where there's no visual obstruction.

The reality is that humans are not good at calculating the vectors involved and predicting the future positions of themselves and crossing objects. When the speeds and angles are right a crossing object will appear to stay in the same place (angle in the field of vision) and so it's very easy to underestimate it's speed (it's not moving at all, right?) and not be aware if the risk of collision.

When you and a car are approaching an intersection at right angles, odds are that you'll under estimate it's speed, but much more important is that the driver will greatly under estimate yours, so neither of you won't make the adjustment to avoid collision.

There's not much anyone can do about this, but awareness will ensure that at least one party (you) will make the necessary adjustment, or at least be prepared to if the other party doens't do so first.

What about gravity? Not only is it a good idea, it's the law!



Kidding aside, it was a very interesting article. I would surmise the strange angle is somewhat of a fluke which made this particular phenomenon particularly prevalent there. Seems it would be a simple and relatively inexpensive fix.

Not at all. The problem is somewhat independent of angle, and at sea manifests at just about all crossing angles.

As an example, I was sailing on a large schooner, and we were slowly passing another sailboat that seemed to be on a parallel course. Turns out it wasn't and both crews were watching a collision in slow motion until it became obvious and we altered course at the last minute.

I detest non-90º intersections. Nearby Riverside is chock full of them.

Simpler solution; when approaching an intersection where you don't have right of way, slow down. Stop if necessary. If there are obstacles blocking your view of cross traffic, move your head.

Oh, I can imagine the return of the bubble shaped Pacer. :)

https://upload.wikimedia.org/wikiped...009_D42119.jpg

Perhaps a little more work to reduce the pillars. There are quite a few cars with frameless windows on the doors, but still have the main pillars.

That is one thing I dislike about Mom's Prius, really bad rear visibility :(

Not quite so simple. What about when you have the right of way.

Intersection accidents are usually the result of 3 incorrect assumptions.

1- I've correctly estimated the speed and position of the approaching vehicle.
2- the approaching driver has correctly estimated my course and speed
3- I have the right of way and the crossing driver will recognize that and yield

We can blame intersection design, but the fault is human, not technical. I'm sure that I'm not the only one who's come to an intersection where the crossing driver is dutifully stopped at a stop sign. Then at the last, and most inconvenient moment has decided it was a good time to proceed.

Stay alive by being ready to react not only to what drivers should do, but what they might do.

The collision occurred because the drivers blew the stop.

It's a shame that the article focused so much on the main pillar. Has it occurred to anyone here that bicycles don't have these pillars?

While drivers were at fault here, the bicyclists contributed by relying on false assumptions about what those drivers were going to do.

Assumption no.3 in my post immediately above yours.

I should point out that the crash in the scenario Joey is using as an example is different; it wll lead to a single vehicle crash. Cyclist will veer toward the right as he focuses all his attention to the right hoping to see that woman. This rightward veer will take him into the tree. Woman walks past unharmed.

Ben

True,

A basic rule of driving is to assume all the other drivers are going to do something entirely stupid. So, for example, when sitting at a stop sign, you see a car signalling a right turn. One should assume they're not going to turn until it is proven otherwise.

So, there are two cases...

1. Car never stopped, and blew the stop sign. Cyclist had right-of-way, not stop. The cyclist SHOULD have seen the car approaching and not stopping, and should have simply stopped and yielded to the car.
   However, that does require a little extra attention on the part of the cyclist. Perhaps an accident can be avoided 99 out of 100 times, but there is always that 1/100 time. :eek:
2. Car did a quick stop & go stop. These are the least predictable. Car appears to be stopped, but when one gets right in front of them, they start moving again.
   Fortunately these stop and go stops are the least lethal as speeds are often reduced, and both parties often note their error and take evasive actions before an accident occurs. This may also be the reason why the first cyclist in the article was spared serious injury.

I'm not sure the staggered intersection favored by the author is the answer. I find them to be quite annoying. Yes, they should slow traffic, and that might be worthwhile considering. Although, for a cyclist, it often means merging onto a busier street for a block or so.

I hit this one a few times a year.

Google Maps Views

IMO, the problem with the proposed solution is that it trades one problem for another.

The new design means that all through traffic in either direction will need to make a left turn. Left turns are their own hazard and already account for a fair number of collisions, including head-ons, which generally have the highest fatality rates. That's a key reason roundabouts are becoming so popular with traffic engineers.

In Mexico, an intersection like this would be left alone, except for inclusion of speed bumps or humps in advance of or at the intersection, so blowing through is no longer an option. Slowing for the bumps means more time to recheck cross traffic before proceeding.

The article is set in the UK, so driving on the left, and a left turn is equivalent to a right in the USA.

I don't think it's an assumption.

My one sentence makes the issue much more clear.

The whole closing-angle stuff is a red-herring. And based on a lot of speculation. The closing-angle argument assumes the cars didn't really stop.

Even if there wasn't a pillar, the drivers might not have noticed the cyclists. Heck, the might not even been able to notice a car.

The point of requiring a stop is to eliminate the issue of things like closing angles.

What other accidents have occurred there?

Now of this means cyclists should not drive defensively.

One hopes that will be the case, however, the difference between being missed entirely and being at least partly sucked under the front wheels is a couple of feet at best. Once the bike starts laying over with the weight of the car on top of it and your legs somewhere in that mess, serious, debilitating injury is very likely. Fortunately in mine, I coupled swerving and braking with standing up as high as I could, so I went on top of the hood. That wouldn't have been possible with a lot of trucks that have a taller front end.

We agree.

I mentioned assumption No.3 responding to your suggestion of adding a stop sign. The assumption I had in mind wasn't yours, but was the cyclist's expectation that the driver would in fact stop.

Good design and traffic laws can offer a certain measure of protection. But one needs to accept that with a grain of salt and not assume that everyone is working from the same play book.

The best protection is what we do for ourselves by being ready for anything.

I never was very good at geometry and all that stuff, but yes, it was a good article, even if reading it did give me a headache. :D




:)