It’s Time for the Car of the Future

One idea that I’ve seen floated about - but not by anyone who is going to actually build it- is a hybrid of the “private car on rails” and “Personal Rapid Transit” ideas- you have a overhead or underground (grade separated) rail system that has two kinds of cars on it. The first kind of car is public. You enter a station, and tell the station where you want to go. The station will pull the next available empty car into the station, and you enter the car. The car accelerates up to speed, and merges into the traffic on the rail - the traffic on the main rail continues past the station freely. (During heavy traffic periods, the traffic on the main rail might be slowed down or sped up to “make a hole” for the oncoming car.) You get to the destination, the car moves to the siding, and slows down and stops, you get out, and the car is either used by someone who needs it, or is sent to someplace where it is needed.

The other kind of car is a private car, which can be used on and off the rails. The driver of the vehicle enters it in his driveway, and drives it to an on ramp. Control of the vehicle transitions to the central system while it is on the rail system. The driver programs his destination into his vehicle, and sits back and enjoys the ride. When the destination is reached, the vehicle merges to an off ramp, and control passes back to the driver, who continues to his ultimate destination. Of course, there are safety concerns - the owner/renter of the private car would have to maintain certification of his vehicle before it could be used on the rails, which I would presume would require a combination of electronic monitoring of the vehicle’s health, periodical inspections, and consent to being kicked off the network at the first hint of a problem. Of course, the off-ramps would also need to be designed so that if the driver did not take control of the vehicle, it would not cause an accident.

You could design the system to allow for different types of vehicles to operate on it- including unattended cargo pods (can’t leave the rails, cargo stations?), attended cargo pods (replacement for delivery trucks), variety of personal pods.

It could be used for both inter-urban transit and intra-urban transit, depending on the network you design. Of course, you would need to have higher rail speeds on intra-urban transit, but you could design a system that would take care of both. Of course, for intra-urban, you would probably end up with private vehicles and private for-rent vehicles and not public vehicles. (I’d imagine that you could either rent a pod like you rent a car now, or rent a seat on a pod that traveled between two points, like an airline.)

For safety, you’d probably want both a central Pod Control system and an in-pod safety system (checks track ahead for obstacles, ability to follow a different pod and automatically brake to avoid or minimize collisions, - stuff that we can do with current technology).

How are the economics of running frequent trains down every two-lane road? I’d think they’d be prohibitive.

Of course. But you can take a lot of traffic off the highways without running trains down every two-lane road.

Or you could compromise with this vehicle.

Or this one. Plus ca change . . .

We oughtta have a thread on models of cars that didn’t last long or never made the expected market inroads in the first place, like the Tucker and the Dymaxion and Amphicar and the Aerocar – and why they failed – but what forum would that belong in?

Interestingly, last week’s Economist has an article on the subject of innovation in self-guidance for cars. There’s apparently a lot happening in the field. But at the same time, you can give the concept a prodigious boost through tackling the infrastructure as well – making it more compatible with the idea of cars with advanced guidance or advanced engines. And it’s there that I find the public will to be sorely lacking. It’s all about cash, and alas, we’d rather spend our public cash on other unworthy things.

I’m not convinced that any changes to infrastructure are necessary; a viable self-driving car is going to have to be able to react properly to something jumping in front of it, be it child or deer. By the time your AI/visual processing is good enough to handle that, I would think that turn lanes would be a breeze for it.

Part of the reason for my suggesting an elevated track was to simplify this very problem. By limiting the scope of travel, and by keeping the deer and the kids out of the way of cars, you limit the amount of artificial intelligence needed. In fact, I’d say we have the requisite amount today.

Plus infrastructure means more than just the roadway, and might include things like refueling or recharging stations, depending.

Well, as noted, an elevated track is expensive to make; impractically so for most rural travel, I would think. It certainly would be possible to do though, assuming that what you’re referring to is basically an elevated highway. (If the cars were to run beneath the track I doubt you gain any benefits over a standard road, and you would certainly buy into a slew of other problems.) I think that access problems would kill the idea of raised tracks for intracity travel, beyond their use in public transportation a la trains.

Conceded on the infrastructure =/= roads angle; in fact I’ll go one further and say that, what with oil being finite and possibly over half gone now, I think it’s a certainty that we’re going to see some new feuling infrasture popping up in the next dozen or two years; pretty much as soon as we can convince a major automaker to abandon oil.

Seems like the Brits are taking making EVs a practical urban vehicle option seriously.

I was thinking about the idea of a ‘carbrain’ on the way to work this morning - something that would participate in something like an ad-hoc roving local network - communicating with other carbrains in the vicinity (this is on the assumption that every car would be so equipped) to prevent collision and to keep traffic flowing optimally through junctions.

It seems like a good idea for in-car computers to be able to collectively negotiate the movement of traffic, however, the concept has a rather obvious potential flaw - it could be hacked - not by people with malicious intent to control other cars, but by people who want to leverage an unfair advantage in their journey; by hacking their own vehicle - perhaps spoofing it to underestimate or understate the manoeuvrability or acceleration/braking properties, the process could be biased so as to favour their passage.

Is that networking strictly necessary? Our brains manage to do that (most of the time) with no input but that from our eyes. A carbrain designed to do the same would work whether the other cars nearby are carbrain-equipped or not.

Networking wouldn’t be necessary if visual processing could be adequately implemented, but it would definitely be an advantage; an ad-hoc network of cars over, say, a 500 yard radius would enable them to interact far more efficiently than if they just had to look for each other visually.

It’s far easier to make two or more fast moving objects pass through the same space without colliding if they simply know each others’ locations, vectors and intentions (which could be done with GPS and networking). Visual processing would still be required for avoidance of unexpected obstacles such as stray pedestrians, debris or animals.

Don’t forget the legacy non-carbrained cars on the road, either. Any carbrain that exclusively relies on networked information is going to be in for a surprise when Grandpa runs a red light in front of it.

And this would increase loading time at every stop stop by a factor of at least one for every person getting on the train.

Fine, you have invented Holland. Let’s get back to the vastness that is the current US town.

Greyhound gets by. Poorly, but it does, and without rails.

“Carbrains” with limitted acceptance could be as simple as being allowed into an ad hoc highway caravan which runs cars more closely packed than otherwise possible at a uniform speed (which allows not only for greater car density but also conserves energy akin to bicycle peletons.)

Also huge benefit could be garnered just by having all cars equipped with a cheap device that reports their position to a central registry that other cars that are “carbrain” equipped can use. Privacy concerns emerge, however.

Which is why I said

Yeah, okay, so I didn’t incorporate the post two posts up in replying; my bad.

In my (weak) defense, I should note that if the carbrain cars are run on the same roads we currently use, the chance of avoiding a ‘transition’ period during which they share the road with legacy cars is zero. There’s always gonna be some guy who wants to drive the tractor into town like he always has.

(Though, if you installed deactivated carbrain systems into all new cars and then activated them all at once some twenty-forty years later, you could avoid most of the overlap, assuming everything still worked.)

DSeid, aside from aiding in police work and of course various Big Brother applications like tracking traffic density and road usage, what possible benefit would there be to equipping car with a centrally-tracked transponder? Especially in comparison to the direct on-location car-to-car communication already being discussed.

Another way to smooth the transition would be for new cars to be equipped with full carbrains that control the car completely, but for inexpensive passive ‘beacon’ carbrains to be issued (probably under subsidy) for retrofitting into all other existing road vehicles, so the full carbrains would at least be able to get information about the position and vector of the older vehicles, even though they would not be able to negotiate on a full two-way basis with them

The context was in a situation of limited acceptance. The transponder is cheap enough to require placement in a legacy vehicles and allows all other fully equipped carbrains to benefit from knowing what amounts to very detailed and timely traffic reports.