User:DirkvdM/Cars

Cars kill and cause climate change

Cars are by far the most prominent cause of deaths under the age of 40, so in terms of 'lost years', cars are more lethal than anything else. They are also responsible for about 10 to 20% of the emission of greenhouse gases, which may in the long run turn out to be even more lethal. So what to do? I've been doing a lot of thinking about this over the last decade and here are some of my findings. Please let me know what you think on the talk page.

Public transport isn't the solution ...

Public transport has some inherent problems that can't be solved.

• Lack of privacy (although that can also be seen as an advantage), but more importantly:
• Having to switch between modes of transport or between lines
• Not being picked up and dropped off 'at the front door'

A taxi would solve all these problems, but would be too expensive for everyday use. Well-organised taxi-sharing could be a good compromise - if this would be used extensively then there would be enough taxis to have one arrive at your front door (or wherever) within a minute. But you would still miss the privacy and have to make detours to pick up and drop off people.

... it's simpler

A better idea would be to quickly develop driverless car techniques like automated highway system and autonomous cruise control system. Of which VaMP is an extreme example. But that will take some time to develop and we need a solution now.

Basic idea

The basic idea is to have cars drive bumper-to-bumper. That itself would already reduce fuel consumption by about 40%. But because it uses roads more efficiently, it will also largely solve the traffic congestion problem, which is another major cause of fuel waste. And it is something that people will be more than willing to cooperate with. This is relatively simple to do on highways, where the congestion problem is often biggest. It would take some more time to make it safe to use in built-up areas, but once that works too, we've got the best of both worlds. Also, cars driving in blocks will make it easier to implement phased traffic lights. A promising initiative is the DARPA Grand Challenge, which was won by my favourite, Stanley, in 2005. The next challenge will be to have the car drive though an urban environment. This will be held on 3 November 2007.

Details

In case you doubt that 40% savings figure - ever ridden a bike in the slipstream of a car? It makes a huge difference. You actually even get 'sucked in' a bit, it seems. Move back a few metres and the effect is gone. Cars on a busy highway usually leave about 10 m space between them, so they each get the full 'blast' of the wind.

And as for safety, that distance is about the most dangerous one one can have. Best is some 20 metres or more (or thereabouts - it also depends on the speed and road conditions). Move closer and when the car in front of you suddenly brakes, the chance increases that you will hit the brakes too late and hit it. Move closer still and you'll hit it harder. Up to a point. If you're bumper to bumper, you'll hit the car before it has started to decelerate much, so you won't hit it very hard. So that is actually the second best distance. Somewhere between 0 (or 1) and 20 metes lies the most dangerous distance and I once heard that that is a very popular distance on busy highways. Of course, until you start braking, the car in front will have to brake for two cars, making the distance covered until it stops longer, which means a danger for whatever it is hitting the breaks for greater. But on a highway, accidents are usually not the result of something else on the road. There are just cars. So if they all 'play at this game', there should not be a problem. The cars form little 'trains', resulting in more space, so the 'trains' can keep a sufficiently safe distance between them. When they can't anymore, the two trains unite into one train.

This is the simplest form - no added technology. Of course, one could let the cars speak to each other, so that when one car hits the breaks, all cars will do so and they all come a standstill at the same time. Well, of course not as perfect as that, because different cars will react differently and synchronising them perfectly won't be easy. But even without this added, it will be safer than the distances cars keep now on busy roads. So it should only be allowed on busy roads, where the effect on safety and energy savings will be greatest. One could also limit it to the left lane (or right lane in Britain), so people whom this makes nervous will not have to participate. People do this a lot already, and if no-one gets nervous and everyone wants to participate then people will be quite able to do this safely. It's the agitation (such as over bumper to bumper driving) and different driving styles that cause a lot of accidents. In this system, people are forced (voluntarily) to drive at a constant speed. And I suspect that therein lies another energy saver. More constant driving will improve both safety and efficiency.

So even if there is any added technology, the car doesn't need to take over completely, just a little bit, and that is very easy to do. But it can be simpler still. Just let the cars lock in physically, possibly with something like what real trains use. That should be easy to develop, although it would be more expensive to implement. Several hundred euro per car at least, but the fuel savings will pay that back very quickly.

This is pretty much like public transport - the cars form trains with private compartments. But those compartments can split off and go their own way whenever they want to, like ordinary cars. That's what I meant with 'best of both worlds'. Much of the safety and energy savings of a train plus the freedom of a private car. This should be very easy to 'sell' to voters, especially when it's just an option for left lane drivers.

Car redesign

The effect would be even better if cars had a shape that makes them aerodynamically 'lock in', so to say. And of course, making them more fuel-efficient (lighter engine) would reduce emissions even more. Outside the train, the car will need enough power to move itself (although most cars have way too powerful engines for that), but when part of a train, very little power is needed. High speed and low power means a very high gear, so cars could be fitted with that. This will make the car react more 'sluggishly', so to say, but that is actually an advantage. Sudden changes (braking and accelerating) cause a lot of accidents and fuel waste. I am wondering if something similar could be done for the brakes and steering, but I'll have to think on that a little more. The car could then be put in a 'train mode'. People who are used to cruise control and automatic gears should feel comfortable with this - it just takes those ideas a little further. When a car is in train mode, a specific light could start burning, to notify the cars in a train that it is going to 'lock in'.

Problems

One problem I see is the changing into and out of the train. Cars or trains on the same lane can easily join or split, but what if a car (in the left bumper-to-bumper lane) wants to leave the train (move into the right lane)? Normally, a car adapts its speed to the cars in the lane it changes into, but here that is not possible. So either the cars in the other lane have to adapt or the train has to completely split in two to give the leaving car room to manoeuvre. The former sounds accident-prone and the latter will partly undo the positive effects, which becomes a problem when it happens a lot. Of course, a similar problem could occur if a car wants to join a long train. If it's to far from the front or end, the train may have to split. So is the former a solution for the latter? This is hard to assess through reasoning and would require statistical info, real life testing or modeling or something. Of course, one could also say that cars can't always join a train. You can't always get what you want. And that goes for much of the above. Not ideal, but better than what we have now.

Any thoughts?

If you wish to comment on the above, please do so at the talk page. Positive is nice, negative but well argumented is even better.