Having taken MUNI (SF's light rail) underground in SF, I have some doubt about the efficiency of such a system. In my station (Church), in the morning, I can see trains coming in every 3-4 minutes or so; and about 40 people get on at once. With a PRT system, you'd have to first line up at the turnstile to punch in your destination. Assume 15 seconds each. Then, when a car comes in, you'd have line up to get in. A person gets in, swipes her card, punches in the destination, etc. Minimum time: 1 minute, if you're being fast. Suddenly, the bandwidth is just not there; the platform becomes the bottleneck. Same would happen at an elevated pathway too, I imagine.
So, a system like this would be infeasible in a dense city like SF. Above ground is another story altogether: the NIMBYs would block all construction.
What could work is a system of autonomous vehicles (taxis) the size of a Google self-driving car, running above ground, summoned via a smartphone app or a kiosk at every corner. So, instead of dedicated pathways and elevated tracks, use autonomous cars.
Also: very importantly, each car must have odor detectors / biohazard detectors in case someone has an 'accident' in the car.
You're thinking about it backwards. It's not a "better MUNI". There's no turnstiles or gates, just bays. Not 4 or 6 like a train station but 40 to 60. Boarding a few people or more likely 1 takes less than 30s. That's 2/second. Plenty of bandwidth. And these stations are smaller and more dense, not every 15 blocks in the city but every few blocks, maybe in a large building's lobby. There's no swiping, it's on your smartphone and automated ( Open app, click "go home", app says "bay 21, approx, 2 min, pod #123, 1 other passenger"). Doors close when all the passengers are present. Pod has "pad requires service" button on console if there is a cleanliness issue.
This is attractive to the NIMBY crowd, fewer cars, less non-resident parking, less noise ( electric motors ) less local pollution. Who wouldn't want local rapid transit? They're not putting in Walmarts.
> There's no swiping, it's on your smartphone and automated ( Open app, click "go home", app says "bay 21, approx, 2 min, pod #123, 1 other passenger").
And in this hypothetical world, absolutely everyone who uses public transit has a smartphone?
Let's imagine the system engineers are smarter than that: if you have a smart phone, you can use the app. If you don't, there'd be a terminal in the station you could use. I'd imagine the engineers for this could have thought of that.
Easy enough to just assume people have smart phones and only have it in cities where mostly people will have them even the poorer people. A cheap smart phone isn't that expensive any more.
It's not just the cost, really: I know I can have a second hand smartphone for peanuts. There'll probably always be some people who don't use smartphones, for whatevre reason.
For getting an intuitive sense of the kind of throughput here, I think it'd be useful to picture a rollercoaster. If anything, rollercoasters are even faster boarding than one-person pods would be, since the whole rollercoaster train shows up at once.
Now picture the line at Space Mountain in the afternoon, except in a subway station.
> you'd have to first line up at the turnstile to punch in your destination. Assume 15 seconds each. Then, when a car comes in, you'd have line up to get in
Why can't you load 15 people into 15 pods at once?
> very importantly, each car must have odor detectors / biohazard detectors in case someone has an 'accident' in the car.
The subway doesn't have this, and it works okay. People just use their built in odor detector, and if there's a problem, they avoid sitting in it and notify a conductor.
Subways dont offer much privacy. How many pods are you going to have to report for smelling like sex before you can find a clean one on Monday morning?
They'll likely have security cameras to prevent vandalism on the inside of the vehicle. I doubt most people want to be recorded having sex and have it watched by strangers when an "issue" is reported.
What if you required everyone to punch in their destination on a companion app, Uber-style, before even approaching the platform? Then all you have to do is wait until the app gives you the green light to approach the platform (your car is ready), get in, and maybe tap your phone to an NFC reader mounted in the car or something to confirm you're the passenger it thinks you are, and you're off. I imagine that the entire process from green-lighted to doors-closed-and-moving could be 15-30 seconds on average.
There could still be a phoneless kiosk/swipe system for people who don't have/don't want to use the app version, but with cheap Android phones available for $20-$50 (and that's today -- imagine where we'll be in 5 years) I figure 95% of throughput could use the streamlined flow.
---------
All that said, I think I agree with the article's sentiment that this system's time may have come and gone. It's hard to imagine what transport will look like 15 years from now, but I suspect that electric robot taxis as part of a multi-modal transport infrastructure with existing trains/subways could deliver most of the benefit of a personal pod system without the huge infrastructure investments of building a completely new set of tracks all over the city.
In the 4 minute / 40 people scenario, you're only talking about 10 people every minute. 10 pods stationed at that stop solves it.
Now, everyone doesn't wait 3-4 minutes. They just get on the next available pod. Remember, people coming into a typical station don't come 40 at a time. They come in clumps of 1-4.
All of which transforms a batch system (subway/light rail), to a continuous system.
There is an assumption that stations are off the main rail. Which, given the size of a typical pod, is feasible. It is certainly much smaller than your typical light rail system
===
btw. wrt to speed.
There's nothing stopping you pre-ordering a destination at a machine (or smartphone) before you get to a pod. Swipe your credit card, see your name on the pod you should go to. Walk in. Sit. And go.
And in the case of your daily commute, even easier. Have it pre-set. Jump in to a pod, face-recognition (or nfc/ble) and if it's your 'typical' time, you're off. No pressing/no swipes. Just go.
===
NIMBy's are always going to be an issue and off-grade (eg. above/below ground) is typically going to be required.
Self-driving cars.. are definitely -the- alternative. But only if there's actually enough road. Eg. it's not going to go any faster during rush hour. It just means you're not going to be driving.
All excellent points. This is why we should invest more heavily in proven mass transit systems rather than divert limited public funds into these inefficient pods or hyperloop systems.
Corollary: Is it really that bad to have to ride a bus or train with other people on-board?
Busses are the least-worst alternative to walking. People are noisy, they smell, they crowd you when you're holding groceries, they make a trip with friends or God forbid a date unpleasant, they're late when you're early and early when you're late, and chances are you'll still have to walk a few blocks to get where you want to go.
Obviously not every case, but I've taken a fair amount of Bay Area busses in my time and there was always at least one of the above complaints.
Agreed that it's not as pleasant as having a car to yourself (I ride the NYC buses and subways everyday), but I think we could solve a lot of the people issues with proper policing rather than outright isolation.
Chances are if you make more "quality of life" laws you'll either have some schmuck causing trouble when enforcement is nowhere to be found, or they'll be used as excuses for cops to pad their stats by arresting first and asking questions later, with most of the punishment hitting minorities, as tends to happen in the U.S.
The problem is cultural. Another poster here mentioned Japan, where busses are pleasant. After all, that is a place where "the nail that sticks up gets hammered down." If we really want public life on par with Japan or Europe, change has to start with each and every one of us. No government program or well-funded startup could ever do it alone.
I have ridden SF public transit of all kinds. MUNI buses are the worst, and it's not because of things that could be solved with policing. The worst part is the super high variance in door-to-door times, even if you try to plan your life around bus schedules it's bad.
If you have to take multiple forms of public transit to get where you are going, good luck. A 40 minute drive can easily be 2.5 hours on public transit if everything goes perfectly.
Also the average speed of SF MUNI buses is 8 mph, probably among the lowest in the nation. If I'm going any distance less than a 30 minute walk, it's probably going to be a better idea to walk than take the bus.
The problem with mass transit mostly isn't other people, it's the fact that it rarely goes from exactly where you start to exactly where you want to go. I take mass transit to and from work every day, but in order to get from my house to the commuter rail station, I have to drive a mile to a bus stop where there's parking (the commuter rail lot is small and fills up quickly) and take a bus 2 miles to the train station. Then on the other side of the train ride, I'm still 3 miles from my office, so I hop on a bike share bike and take that to the station nearest my work. Then I walk 2 blocks to my office. And then all of that back again in the afternoon. Lots of people tell me that they drive because they couldn't deal with my crazy routine.
A PRT system would possibly have a station much closer to my house, so there wouldn't need to be a single commuter rail station for a town of 20K people, and therefore no dealing with crowded parking lots. It would likely have a stop at or near my office, as it's a very busy business district (Kendall Sq, Cambridge.)
If you are in Japan, then no, trains are clean and people are mostly sane. In the US here, esp Portland, they are filled with interesting individuals, smells and acts. I have rode the Max train for years, but now drive since it is more pleasant and shortens my overall trip considerably.
A hybrid pod could act as an autonomous car, picking you up at your house, then merge onto and "mount" a shared electrified rail for the speedy middle part of the journey—the inside-a-city equivalent of a freeway system. Effectively, the flow would look similar to bicycle commuters biking to the subway and then getting on, but without the need for the subway train to encapsulate the pod.
Here in Ithaca, NY, we have three major communities that are relatively close, but just too far away from a combination of distance, elevation from hills, and miserably cold temperatures: downtown, Ithaca College, and (Cornell's) collegetown. A simple pod connection between collegetown and downtown, and then another route to IC, would do wonders for our transit system.
See, the nice and terrible thing about Ithaca is that it's a small valley hemmed in by hills on three sites and a lake to the north, so most things tend to be bunched together, which can be almost walkable, depending on distance, temperature, and whether you're carrying groceries. The downside is that there are like 5 arterial roads to get anywhere meaningful in the area, and if just one of them is closed for construction, traffic slows to a halt.
We don't need long or complex routes. I've been on the WVU pods, and while they're nice, they go FAR. I'm talking about a quarter or less of the distance they cover, and only in two directions (and maybe a third because you have a better chance of getting into Willy Wonka's chocolate factory than finding parking at the farmers market). So please, by all means, if you're looking to trial travel pods somewhere, start here.
That's a good application for Intamin's Mountain People Mover.[1] Intamin builds both roller coasters and monorails, and for a ski resort, they combined the technologies. It's a personal rapid transit system which runs on roller coaster track. Steep hills are not a problem. Nor is snow.
1. They should convince company with $$$ such as Google, FB, Apple to build this for building between their campus buildings.
Those company has $ and regulation issues might be might a lot smaller compare to convincing a city like SF to do it.
Might be easier to sell because of the "cool factor".
Those companies can easily convince City of Mountain View, Sunnyvale, Cupertino to let them do it.
2. If #1 works, those companies can build routes between the public transit hub, large shopping malls to company.
Help ease the parking/commute issues.
3. Convince Disney Land, World to do this for the "cool factor".
4. All Major convention centers, sport stadium to build routes to major shopping mall/large parking structures nearby.
A tech campus deployment makes a lot of sense, since it's always a lot of people going different places— eg, difficult to service effectively with buses or conventional rail.
I think the only option with real value is #4. The WVU pods are really nice for ferrying students allll the way from their dorms and apartments to the enormous football stadium. A place like Morgantown, WV, is pretty much using every square foot of land it can, so parking is out of the question. But let's face it, getting to ANY stadium on game day sucks when you're sharing the experience with tens of thousands of people.
Only problem is that it's a very specific use case and is likely only to have seasonal ridership. And any example that can be used to show why something can be built is one backfire away from being a reason not to do so.
The Disneyworld one is still there. Funny that I rode it as a kid, then I was able to take my own children on it just a few years ago.
Disneyworld also has a railroad (and the monorail) that serves a purpose of serving the whole park. I went at peak season and the railroad seemed to keep up with demand, although I think that's because it serves the "outside" of the park, and if you're wandering the inner parts of the park it's just easier to walk to where you need to go.
I've been thinking about this a lot and think there's a few things to think about.
First, the reason we haven't seen these is because until very recently the tech needed to make them cheap wasn't there.
Second, the technology seems to be caught a classic "worse is better" situation. The actual PRT tech is clearly superior to buses and LRTs, but those are well understood and there's a simple evolutionary path (car, bus, bus lane, lrt, lrt lane, subway). PRTs require a leap of thinking from people and governments don't have any incentives or mandate to take such risks.
Third, PRTs tracks may be hard to connect to a true network where pods seamlessly switch between tracks.
Finally, the self driving car could very well kill this in its tracks, but no form of transport can deliver speed or consistency without a dedicated right of way, be that suspended track, tunnel, marked lane etc.
Self-driving cars will likely kill PRT ambitions by being a better band-aid for long enough to not have to think about the problem for a while, like how successive attempts at solar power were buried for more than 100 years by more efficient coal extraction, then oil, then nuclear, then oil, again.
Self-driving cars aren't a band-aid. In the long term, where every vehicle on the road is self-driving, then they're the best solution possible (for any travel distance that doesn't warrant something like Hyperloop / planes / trains). I say this because self-driving cars can go everywhere (that cars can go). Pods have a fixed track, they can only go where there's track to go. The only way for Pods to be better than self-driving cars is if we rip up all the roads and replace the entire road network with a pod track network, but that obviously will never happen (and even then it's not necessarily better, because a pod track network could only be used for pods, but roads can be used by a lot more than just cars).
Self-driving cars would still be less workable than rail when there's a certain level of density. I can't see New Yorkers all loading up in self driving cars; they just don't scale as well as trains do. Even assuming quadruple capacity from the roads, you would still need a large number of fairly beefy bridges and tunnels to get everyone coming from outside of Manhattan into Manhattan.
Oh sure, there's still a need for high-capacity mass transit options. But mass transit cannot go everywhere (or even most places). My argument here is strictly about personal transport (which is what Pods are).
In what way do they not scale as well as trains do? I think self driving cars will probably introduce a huge pile of problems, but that particular claim seems odd to me.
It comes down to space. Cars occupy way more space per passenger than trains do. A fully loaded subway train can easily carry almost thousand people. The 7000 series in the DC Metro carries ~175 people per car maximum, and the trains are 8 cars long, for 1400 people per train. Normally, 26 trains per hour are scheduled during rush hour, allowing up to 36,000 people per hour if all trains were 8 cars long (they currently aren't due to power constraints). Other systems that have longer trains, better headways, and better seat configurations have way higher capacities. New York for instance could achieve much greater capacity given its longer trains and a modern signalling system.
By contrast, a road lane (which uses the same space) currently typically carries around 1000-2000 cars per hour. In ideal circumstances it can be almost 5000 cars per hour. Even assuming quadruple capacity from self driving cars (not realistic), 20,000 cars per hour in ideal circumstances would compete with trains at a far higher environmental cost.
But in more realistic scenarios with quadruple capacity, 4000-8000 cars per hour would fall way short of a train. Given that we have to account for pedestrians, cyclists, etc., I can easily see traffic lights continuing to exist, making self driving cars not achieve the miracles people expect.
This certainly would not solve traffic congestion issues. Other modes of traffic address traffic much better than more cars/same number of cars does. If a car can't make it across town in an hour at rush time, then "everywhere cars an go" isn't that much of a selling point since cars can't really go that far then in a good amount of time.
Once everything is self-driving, they can coordinate much better than humans and fill roads to their capacity (as well as routing intelligently since we'll be able to have completely accurate traffic predictions, although that demands a level of widespread coordination that won't necessarily come at the same time as 100% self-driving vehicles). You can also throw vans/busses into the mix if the roads still can't keep up, or automated carpooling services.
100% self-driving vehicles will be able to fill roads to capacity for 2 reasons:
1. It's all automated, so most of the common causes for traffic jams simply won't happen (most traffic jams happen spontaneously once the road reaches a high percentage of capacity, rather than for any concrete reason that demands a jam, and self-driving cars will be able to avoid the turbulence that causes the jams).
2. Because it's automated, cars can also drive faster while remaining safe, and drive closer together as well. By the time we're at 100% self-driving cars the driving software will be pretty well-tested, so the only significant risk* will be a mechanical failure in the car, and even if that happens, because everything's automated the surrounding cars will be able to adapt much better.
*I'm talking here about accidental risk. Any kind of intentional issues (hacking the car, throwing heavy stuff at the car from another vehicle, etc) are not particularly relevant here.
I guess you (and danmaz74) have a point here. In both my supposition and yours, we are assuming computers are smart enough to coordinate the vehicles well, only for cars, the infrastructure already exists. I guess my only reply here is that, yes, there is the cost of the infrastructure itself; however, let's say that this PRT thing takes off, the fact that their infrastructure is physically separate from the road, it would initially alleviate traffic issues much more than self-driving cars would. Having to deal with irrational, nondeterministic human drivers would probably make traffic alleviation not as apparent at first compared to a completely parallel infrastructure.
Still, this does ignore adoption rates. In a more real world, I think you'd argue that adoption of self-driving cars would be quicker and easier than adoption of PRT, that might hamper my model. So, perhaps the conclusion of the article is right after all.
What is the capacity of a road with 100% self-driving cars, and how does that compare to the capacity we are utilizing today? And how could traffic jams not happen? What happens when you exceed the capacity of a road? If you just wait in your autocar to get on the road, that seems similar to a traffic jam.
Traffic jams spontaneously occur before roads reach theoretical capacity. You should look up information yourself on that if you're curious (because offhand I'm not sure what to search for).
Right, so 40 years later we have 100% autonomous cars and use the entire road capacity of 20% more than we currently use. Also more people are driving so we are back where we started with traffic jams or waiting in line to get on the road.
Why are more people driving? Unless you just mean because of population growth, but we've always needed to build new infrastructure to handle that, and the next 40 years won't be any different.
If anything, self-driving cars will reduce the (relative) number of cars on the road. Several reasons:
1. Since they will probably finish their journey faster than a human driver, they won't be on the road for as long.
2. Once we hit critical mass of self-driving cars, we can get rid of taxis (and Uber and all that), in favor of calling a self-driving car to come pick you up. These cars won't need to be out on the road trolling for passengers.
3. By the time we have 100% self-driving cars I guarantee you there will be multiple companies that are in the business of automatically arranging carpooling, which will mean more people per car (and therefore fewer cars).
4. We'll still have the option to use busses, or mass transit, like we do today. Self-driving cars might even make those options more appealing; take mass transit to the nearest stop, have a self-driving car take you the rest of the way (as opposed to driving the whole distance yourself).
Also, once we reach critical mass of self-driving cars, we can start exploring other infrastructure options for moving self-driving vehicles around besides surface roads. Remember crazy-looking highways for self-driving cars in I, Robot? I don't think we'll end up with something that looks like that, but that's the general idea.
In the long (very long) run it could solve this problem too - if all cars were fully automated and very safe, they could travel much faster even inside cities, thus augmenting the bandwidth of roads.
I'd argue that they can't go too much faster in cities. For starters, people live in cities and tend to walk or bike places. On a city street with no people walking around, sure, cars could go faster and be safe.
But in a city with people, cars will need to go slower and stop for people (or some combination). Even with instantaneous reaction, a car can only stop so fast.
Of course self-driving cars are the necessary future, but it will drastically increase ridership as it becomes so much more convenient than the alternatives (including being sober/awake enough to drive). Over time I suspect that even the most efficient car system will result in traffic sucking just as much today, except with the option to watch tv instead of the road.
This is just rail with person-sized vehicles. Maybe you can dub it "modular rail" to make it easier for people to grok, but it's not that weird. Also, you could have them run on the existing subway system and just build off-line stations. Though before you do that, you should take into consideration designing the pods so I can take my bicycle on it, or my dog, etc.
It's just too far ahead of its time. You need to be able to do this without building special infrastructure for the pods, which is an enormous risky up-front investment for something that is, frankly, almost certain to fail. Times will change and the infrastructure can't change with it. And you can't roll the system out a little bit and get a little bit of the benefits; it's useless until you fully cover a significant set of use cases.
By contrast, if the pods were self-driving cars on the roads that already exist, the economics change completely. No special infrastructure. Can easily roll out on a small scale for small benefits. Can easily ramp up. Massively smaller initial outlay. System can grow and pay for itself as it goes.
The problem with self-driving cars as a cure-all is the Jevon's Paradox: if the costs of auto transport are knocked down, then everyone will use them at the slightest whim, especially when tired or drunk. People can live 2+ hours away from work and it won't be a problem when they can work or sleep in their car.
I'm not saying this will happen but I do fear that mobile homes will make a comeback as people live permanently on the road in cars they don't have to drive or pay property taxes for, all while taking up 2-3x the space of a regular car.
In terms of getting a system like this set up, you seem to be complaining that it might succeed too much. Perhaps so, but that's a problem the inventor here would prefer to have rather than the one he has now.
Also, if that did happen, the RV might take up 2-3x the space of a car while moving, but then they end up not taking up the space of a house while stationary, which they usually are, so, you know, probably a net win for density. On the off chance this took off bigtime the economy would further adjust (rentable spots, etc.).
Let's put it this way: we had te technology for basic solar energy in the 19th century. If we had forbidden fossil fuels and exclusively used solar and other renewables like hydro, we would have taken a much longer time to get where we are today, but the environment would have benefited and we wouldn't be in such trouble with climate change. The path we choose is just that: a choice. And cheap choices are easier.
I'm all for innovation, but one of the most laughable human qualities is the idea that there will ever be a solution to a problem that makes the problem go away. Ex: We may have eliminated smallpox but if it ever came back we'd be in serious trouble.
The RV idea assumes minuscule fuel costs and I doubt it would be anything more than a niche appeal, like bikers, but like bikers, it could be annoying and disruptive (god I hate how the tech industry made that word only mean good things) for the people around them.
People could sleep in their RVs as they drive around at night. Again, assuming low fuel costs.
This type of system would seem to have the same failure modes that complex baggage handling systems have. [1]
They are both moving baggage (passengers) on demand between two points. I seem to remember seeing an article indicating that the failures Denver was having (all the empty pods end up at one spot) was visible in experiment, but I can't find it now.
What I sometimes wonder about is whether or not we'll some day seen these kind of personal pods configured for use on existing railway stock. Many times while waiting for a train, I've sat there pondering - "what if I could build some sort of portable segway-like thing that I could just clamp onto one of those rails on the track, step onto it, and propel myself down the line" .. and I honestly can't think of any major reasons why this wouldn't be viable in this day and age. Pods could clamp onto the existing rail, shuttle individuals around, communicate with each other to avoid collisions, and so on .. maybe I'm just not thinking it through well enough, but couldn't we just build better rail-transport devices at the personal level which utilize the existing infrastructure, instead of having to rebuild it all from scratch with incompatible systems?
+1. Having something that can use the existing rail infrastructure in off periods would be great as a transitional adoption method. That way you have time to address the institutional jitters while still providing/proving real service.
For example, during weekends / night when the load is much lower (or service doesn't run anyway).
One big obstacle is that the existing trains with more passengers should & need to always have the right of way. For example, should the non-pod people be stuck at the platform if you have trouble clamping your device onto the track?
This sort of thing was tried and apparent failed in the planned "ecotopia" of Masdar City.
"The initial design banned automobiles, as travel will be accomplished via public mass transit and personal rapid transit (PRT) systems, with existing road and railways connecting to other locations outside the city..."
But "Under a revised design, public transport within the city will rely on methods other than the PRTs."
I don't think you can argue that self-driving cars killed PRT, when we are still building brand new subways and light rail.
I think anything new/innovative loses in the sound byte war. I remember ~2 years ago, the election for mayor of Honolulu hinged on the candidates' stances on building a railroad vs. expanding the existing bus system (rail won). I think if either of them were arguing for PRT, you couldn't explain it to voters in the sound bytes summaries the news gives us.
I would think tracked PRTs would need less road space than driven cars. In medium density cities roads are a quarter of square footage. You could probably take out 1/2 or 2/3rds existing roads then and convert them to yards or parks.
So, a system like this would be infeasible in a dense city like SF. Above ground is another story altogether: the NIMBYs would block all construction.
What could work is a system of autonomous vehicles (taxis) the size of a Google self-driving car, running above ground, summoned via a smartphone app or a kiosk at every corner. So, instead of dedicated pathways and elevated tracks, use autonomous cars.
Also: very importantly, each car must have odor detectors / biohazard detectors in case someone has an 'accident' in the car.