No matter how convenient, the carriage is not your space, nor can you leave emergency supplies or a gym bag in it to use on your way home. And God help the poor person that develops an emergency need for a toilet en route. At least on the bus you can ring for a stop at the a gas station.

PRT=Great Hanging=Stupid (expensive, complex and dangerous) Solar=Super Stupid (super expensive, Complex and Dangerous.
Must of been designed by a PRT hater

I really like the base concept, very close to what I had dreamed up.
1) 30mph is waaay too slow! Find a way to safely go 200mph for real cross country travel. Will need modern cars for this, not ones that look like they belong on a Disney Land ride.
2) No reason you could not talk to or touch the panel en route and ask for a bathroom break, and just have the pod stop at the next one. With high speed and enough stopping points, the bathroom break is really not a problem.
3) Since they are elevated, can share space with roads, railways, rail trails (leaving rail bed empty the possibility of heavy freight).

Wow, can't believe my strong reaction to this article. I know this is just a concept. But NOOOOO!!! My reaction is based on living in Southern California.
1. A oil shortage will come ($4/gal already), but nothing will be done until too late. Human nature. (over population another great example, we build malls on our best farm land, gonna bite us in end, but hey money now!) 2. We love, in California, to invest in new transport systems that never get completed or work properly. Super speed train for example or mag-lev. There are proven systems in place around the world, why not use them. Like efficient light rail. 3. In Southern Cal. we are spread wide not high or dense, which means you'd need rail everywhere or other connections like bus, or people won't use it.
Why not a electrical assisted velomobile? Its pedal powdered and it can be designed to run on rails using the third rail to power it. So at beginning and end of destination you pedal (with power assist from battery) allowing flexibility, and while on rails you use third rail power. (use some kinda of power counter for fee) They could easily go 40mph are enclosed, so all weather, can use solid tires because of low speed, so no punctures. Plus bike paths are a lot cheaper to build for outlying areas.
Right now a 2012 suzuki DR200e motorbike does over 100mpg, one guy says at lower speeds he gets 170 mpg. http://www.cleanmpg.com/forums/showthread.php?t=41832 That's a bike with aerodynamics of a brick and isn't even fuel injected. Direct injection 2-strokes are even more efficient and have low emissions, but they only make a few because motor companies don't make as much money on them and people still think 2 strokes are "bad". 200 mpg can right now be done, but we don't. about 1/10th fuel usage that most people use right now. (Of course that would cause gas price to fall, which means everyone jumps back into a SUV, cause its cheap again!)
There is a ENORMOUS amount of things that could be implemented/done right now to save ridiculous amounts of fuel/energy. But we don't do them because we don't like change (especially in our lifestyles), feel we can't afford it, may affect our quality of living, it's inconvenient, there are laws in the way, there are potential law suits, or a whole host of other stupid reasons. Adding another idea like this article just muddies the water instead of making our choices clearer.
Sorry just my 2 cents, I'm probably wrong. And apologies for cynical take on human nature, even if I feel its true.
(dumb question, who's the guy who gets to fix the solar panels on the freeway, that would be a hairy job!)

I love the Overhead solar panels! Overhead Rail requires some extra costs but can also overcome space issues. The overheads can also have wind generators, or thermal generators.
It could be a good solution for specific locations. Eg. where roads & trains already criss cross the ways underneath.
There is a need to shape pods for maximum efficiency, instead of this love for rounded cubicles. The shape needs to be aerofoil like so that lift is generated, minimising load on bearings, etc.

Why don't they pay engineers instead of Starbucks hippies to design stuff like this? PRT (personal rapid transit) is the right answer but @Michael Mantion was right in that being upside down outweighs the benefit of being able to staple solar panels to the bottom of it.
We need to instead look to something more like bumper cars and highway lanes for the design. Pods can travel a decent pace when on the main line. When they get to a entry/exit station they simply proceed slower in case of accident while some of the pods exit and enter on a sort of switch track.
Pods on the on ramp can speed up to the correct speed before entering traffic the same as with cars. The pods can be equipped with a mostly simple location aware technology to prevent them merging into each other but there are multiple simple safety mechanisms that are possible like having transit pods occupy the left lane when pods are merging like with highways today, giving onboard humans access to an emergency "oh crap" button etc.
As long as you model the flow like a highway with on and off ramps even if cars bump/rub into each other they should traveling to closely the same direction and speed so impact should be pretty minimal like a bumper car.
Because the cars are independently location aware (using less technology than a cell phone), and the track is aware of the cars as they pass by sensors, there are 2 completely redundant systems so if the technology completely fails inside a pod the other pods and the track based system will detect that failure.
Both systems can be monitored by a central control center for things like congestion and track obstruction. If there is a traffic jam ahead it is trivial for the system to reroute the pods destined to travel through that location.
This is where it gets interesting. Because the system treats cars the way packets are treated on the Internet, the system can monitor total travel times (latency) of all the paths based on congestion and "load balance" traffic to longer but potentially less congested routes automatically or through some manual traffic engineering of a technician at the control center.
Another concept that can be used is if there are 10 lanes (5 in each direction) depending on commute traffic (early or late) they can be shifted around so morning could be 3 going out of the city and 7 going in and evening commute could be 7/3. This is used in some places (like bridges) now but it would be easier to do with the pod system. The main problem would be the need to pass closely with other vehicles in the opposite direction at speed would scare people but subways do this now and you can always move a divider wall back and fourth.
All in all the system allows much higher density, greater safety, and more situation awareness than normal road systems and better commute times and flexibility over large stop and go vessels like subways and buses. I would even add that you could easily bring a bicycle or segway in the capsule for the last couple blocks of your commute but people would get mad about it on a crowded bus or subway.
Since there isn't a set size, route, and/or amenities of the capsules you can cater to different user preferences. Things like toll roads, different numbers of seats, on board TV screens etc. are all possible.

How is this system any different than the Skytran system that I've been reading about for the last 10 or 15 years?
www.skytran.us

I don't get how this would work in any practical fashion. - The pod station would be a huge queue of people waiting for their pod. - The chance of more than one person to a pod seems remote, much like cars. I mean I wouldn't want to get into a confined space with some stranger who is as likely to smell bad as be drunk / high / looking to mug someone or listen to loud music. - A touch screen to pick your destination? Do you know how many kids would hit every destination and then jump out? Or pick the wrong destination? How does the pod swap tracks without everything turning into a traffic jam? Trains work so well because they go along a predetermined route at a regular time. - Speed. 48km/hr? Might be useful to cover a couple of blocks but I'd hate to try my commute to work (80km). The train I currently get takes 45 mins for a total door to door travel time of 1-1.5hr. This thing would take closer to 2-3hr. - A little pod for just a couple of people seems wasteful (even if filled to capacity) when you could just have a train that carries hundreds of people at once since most people want to go to and from the city.

While I am not a fan of mass-transit systems that have to be subsidized by the people who don't use it. Once you are using a subway or elevated track system I fail to see the problem with hanging the cars under the tracks. The only real difference operationally is the shape of the rails. Plus with a dampened hinges the cars will passively lean for the curves and if you end up stopped on a curve the car will hang level and there is a lower likelihood of there being debris on the track.
Keeping the cars reasonable full is simply a matter of charging the same rate for one passenger or a full load. If the door closes or the fare is payed before the destination is accepted you won't have a problem with a-holes sending the cars on their way empty.
Of course if you have cars capable of limited autonomous operation (road trains and self navigating a parking lot), and building codes that require adequate parking in the building for the building most of the problems of private cars goes away.
I also doubt that the pods will run through the drive-through at the Burger Barn.

Thanks for the article and comments. In answer to some of the concerns.
Speed in cities: Buses average 8-12 mph, trains about 18 mph, cars (counting free ways, track with your GPS) about 24 mph. Being able to get within an urban network at a reliable 30 mph is a pretty dramatic improvement over current modes of urban transportation. I am always amazed when people who waste 40 to 75 hours a year stuck in traffic jams find 30 mph without any congestion a bad idea. Speed increases are possible, but at least on initial systems such as between a hotel, airport terminal and car rental, 30 mph is adaquate.
At 30 mph the big causes of inefficiency are vehicle parasitic mass and repeated applications of power (start-stops). Similar to ski lifts and cargo nets below helicopters, suspending the vehicle below the rail supports radical reduction in parasitic mass. The computer network removes the repetitive start-stops.
Start small, the Internet started at 300 baud. If your community would like to have a small system contact me. We will see where one can be privately financed.