This has always seemed a bit obvious to me: the problem you are trying to solve is to take something which weighs several hundred pounds (passengers) and put them on something which is traveling at 100mph and weighs several hundred tons (a train). From an energy efficiency perspective, stopping the train is just silly.
Similarly, when you drive from A to B, most of the energy expended is used on the side-effect of moving their car, not on moving them.
However there's the problem of accelerating and decelerating your fragile body to and from 100 mph. This gets pretty nasty with trains that run off of overhead wire (which is more likely the higher the density of stops), you'd have to get the shuttle in motion and then move it sideways once the engine cars are out of the way (to get on top once the connections to the wires are past you).
There's also the problem of "what if more people want to get on or off this stop than the shuttle has capacity?".
Somehow I'm reminded of (mostly) bad combining giant robot anime....
Embarking and disembarking is still a problem. What if 100 people in a row want to get off at the same stop? If the queue is too long for the low-speed section, it will back up onto the high-speed section, bringing the whole system to a halt. What if 100 people in a row suddenly want to get on at the same stop? Where will the cars come from?
Maybe you would need loops for overflow? If you look around on the site, somewhere they talk about being able to call empty cars to a station. I don't think the system is unusable because of these issues; solutions just need to be worked out, like in any new technology.
Edit: Additionally, it's unlikely that 100 people will want to get off at the same time, which is one of the advantages: it's on demand, rather than based on a schedule like a train. Taxi lines have the same issue, but they work out.
If passengers were encouraged (perhaps with a reduced fare, say) to register "transit plans" (like flight plans) before/while boarding, the system would always know roughly how many people were going to get on/off at each stop. Presumably it would use SMS, so you could text a train/bus/whatever to stop at your stop much like calling a taxi to your house.
fyi, since the blog asked for some translation help, the man in the orange explaining with the model is Chen Jianjun (陈建军). The show is called "Win In China" http://en.wikipedia.org/wiki/Win_in_China
Instead of running the car on top of the train, why not run it along side? You load up all the passengers getting on the train into a car, the car drives alongside the train at high speed. Some passengers hop on, some passengers hop off, red lights flash, and the car drives away.
Less dangerous: only 2 things to keep track of. because of the parallel tracks, nothing can crash.
More dangerous: Limited time to do the transfer, until the "access road" runs out.
The problem is the length of the parallel track.
A high speed train does 200mph, if you allow 15mins for the people to transfer on/off the catch-up car then you need them to be parallel for 50miles!
And that's only with current trains, this system is only really necessary when trains get a lot faster.
The fatal problem: the train is arriving in N minutes whether or not the car is ready. Just as car loading is to be finished (no more people allowed), the last person trips and falls, twisting an ankle. Do you have time to help them on board? How do you get them out of the way?
If the car is not ready for whatever reason there is no fallback... that car has to attach or else there will be a collision. It seems far too rigid an approach to accept regular humans as a moving part.
Here is another approach. The passengers get on a helicopter which lands on a helipad on the train, and quickly get off. The disembarking passengers board the helicopter on the train and fly back to the station.
Benefits: Does not require complex and dangerous docking/undocking mechanism. Does not require perfect synchronization (unless there is a tunnel a mile from the station -- this can be avoided by planning stations more intelligently).
Once the train is in operation I'd like to see the insurance rates on the infrastructure.
16 comments
[ 2.8 ms ] story [ 37.9 ms ] threadSimilarly, when you drive from A to B, most of the energy expended is used on the side-effect of moving their car, not on moving them.
There's also the problem of "what if more people want to get on or off this stop than the shuttle has capacity?".
Somehow I'm reminded of (mostly) bad combining giant robot anime....
Edit: Additionally, it's unlikely that 100 people will want to get off at the same time, which is one of the advantages: it's on demand, rather than based on a schedule like a train. Taxi lines have the same issue, but they work out.
I submitted this to HN a few days ago: http://news.ycombinator.com/item?id=1275848
Or you could, in those circumstances, just... stop the train. ;)
Instead of running the car on top of the train, why not run it along side? You load up all the passengers getting on the train into a car, the car drives alongside the train at high speed. Some passengers hop on, some passengers hop off, red lights flash, and the car drives away.
Less dangerous: only 2 things to keep track of. because of the parallel tracks, nothing can crash.
More dangerous: Limited time to do the transfer, until the "access road" runs out.
And that's only with current trains, this system is only really necessary when trains get a lot faster.
If the car is not ready for whatever reason there is no fallback... that car has to attach or else there will be a collision. It seems far too rigid an approach to accept regular humans as a moving part.
Benefits: Does not require complex and dangerous docking/undocking mechanism. Does not require perfect synchronization (unless there is a tunnel a mile from the station -- this can be avoided by planning stations more intelligently).
Once the train is in operation I'd like to see the insurance rates on the infrastructure.