Car2Go + DriverlessCar + Uber - Better than mass transit someday? (Austin: sales, Home Depot)

[austin-steve]

As I jogged pre-dawn past a parked Car2Go downtown the other morning which had it's parking lights on, and I wondered if that would kill the battery, it got me to thinking. It dawned on me that a combination of Car2Go, Driverless Car technology and Uber might someday represent a cheaper, more effective, environmentally friendly, and efficient way of moving the scattered masses.

I haven't thought it completely through but it's been stewing.

What if I didn't own a car but could use an Uber-like app to summon a driverless car to pick me up and drop me off?

This solves the following problems:
1) No routes or schedules needed. It's 24/7/365.
2) Solves the "last mile" problem as it gets me to any Point B from any Point A, door to door.
3) Traffic congestion will be immensely reduced by driverless cars which don't rubberneck, have emotions or get into wrecks. They communicate with each other in a Borg-like way, increasing road capacity through advanced technology, much as the bits and bytes of internet traffic are managed.

This could be subsidized using the money not wasted on Austin rail, and by ditching Cap Metro entirely and using the 1 cent sales tax to build and maintain the fleet of driverless cars.

Much as computerized inventory intelligence lets Walmart, Home Depot, Costco and other distribution centers learn and predict demand and need, so would the traffic system learn the travel patterns of Austin and be able to pre-stage cars close to areas where demand peaks are about to occur.

We might still have a Family Car for vacations, beach trips, etc., but if getting around was as easy as some taps on an app followed by a short wait and a stress free ride to a destination while I work on my tablet or talk by phone, why would I want to drive myself anywhere?

I think this is the future, and has probably already been thought of, but if that's the case, why are we building a rail line that will be made obsolete in 10-20 years by the system described above, or something similar?

Steve

[Trainwreck20]

Well, to play Alex Jones for a second....what if the police wanted to talk to you..and oops! They make the car take you to the police station instead of to the store. Just a paranoid thing, but I suspect not entirely out of the realm of worry - when you are not driving your car, who else can?

[sojourner77]

Go home Steve, you're drunk! - kidding. In all seriousness, this is still 'Murica, and people won't stand for it, especially in the suburbs where they hang on to their guns and bibles with greasy KFC fingers.

[Novacek]

Originally Posted by austin-steve

why are we building a rail line that will be made obsolete in 10-20 years by the system described above, or something similar?

Short answer, it won't.

Longer answer. It's going to take significantly longer than 10 years for driverless car systems to be fully up and operational, and operating completely autonomously and empty (constrasted with automated, but with the driver there to take over in an emergency).
In fact, the driverless technologies are _more_ likely to show up first in, wait for it, rail applications (it's just significantly simpler, a more constrained problem).

Even once you get to that point, rail (or other large vehicle systems) will be significantly more efficient in moving large numbers of people through capacity constrained areas. Consider: even if you had these autonomous car2go vehicles _today_ they'd just be getting stuck in I35 bottleneck traffic.

[scm53]

I was reading something interesting recently (Jarrett Walker?) and the writer was saying we should focus on transit outcomes, and not transit methods. What you are talking about, Steve, is an outcome based solution that is guaranteed to outrage the method focused folks.

When I think of all the things in my life that have gone from ubiquitous to non-existent, this seems eminently do-able.

[gpurcell]

Originally Posted by Novacek

Short answer, it won't.

Longer answer. It's going to take significantly longer than 10 years for driverless car systems to be fully up and operational, and operating completely autonomously and empty (constrasted with automated, but with the driver there to take over in an emergency).
In fact, the driverless technologies are _more_ likely to show up first in, wait for it, rail applications (it's just significantly simpler, a more constrained problem).

Even once you get to that point, rail (or other large vehicle systems) will be significantly more efficient in moving large numbers of people through capacity constrained areas. Consider: even if you had these autonomous car2go vehicles _today_ they'd just be getting stuck in I35 bottleneck traffic.

I predict you will end up being 100 percent wrong in about 20 years.

[gpurcell]

Originally Posted by scm53

I was reading something interesting recently (Jarrett Walker?) and the writer was saying we should focus on transit outcomes, and not transit methods. What you are talking about, Steve, is an outcome based solution that is guaranteed to outrage the method focused folks.

When I think of all the things in my life that have gone from ubiquitous to non-existent, this seems eminently do-able.

Yep. Also scalable and can develop in an incremental fashion.

[austin-steve]

Originally Posted by gpurcell

Yep. Also scalable and can develop in an incremental fashion.

That too.

Think of a train with many personal mini "cars", traveling like a convoy, yet not connected or restricted to fixed tracks (just roads) and the ability to split off or join up the main flow as needed.

We already have "driver assisted" cars, mostly safety features such as rear end avoidance, and auto parking for the metro-sexual dads who can't parallel park old school, but also auto-cruise-control that keeps the space between you and the car in front without you having to toggle the steering wheel +/- all the time. I think "lane drift" warnings are in some cars now, as well as blind spot warning. Also, smart-GPS technology that knows realtime traffic and can recommend alternate routes.

So the "manually driven" cars are serving right now as a test bed for many of the driverless car functionalities.

Steve

[jb9152]

Originally Posted by gpurcell

Yep. Also scalable and can develop in an incremental fashion.

Driverless autos will actually reduce either speed or capacity. You can't have both. The only reason that highways are able to achieve the throughput that they achieve today is that we allow human drivers to follow at unsafe stopping distances. Police don't enforce a minimum separation. A driverless car manufacturer, on the other hand, if they're not already buried under a pile of safety regulations, will not take on the liability of building in unsafe operation. As speeds increase, safe stopping distance increases exponentially. Driverless car manufacturers will enforce safe stopping distance, and likely will also include some additional margin to account for less than optimal brake performance, road conditions, etc. So, you have to either reduce speeds, or reduce capacity. Which one do you think people will accept - 70-75 mph on I-35 with long enforced safe stopping, or crawling along at 30 mph with shorter enforced safe stopping but higher capacity?

If we operate these vehicles on highways that have the same capacity constraints that we have today (and are likely to have for a long, long time), then not only will they not replace the need for traditional mass transit, they will actually *reduce* capacity on the system as a whole.

Driverless cars are not a panacea. Yeah, they're cool. But they're not high-capacity, even in huge numbers. Since the 800-pound gorilla in Austin transportation is capacity, driverless cars will either have a negligible or negative effect on the single biggest problem we face.

[jb9152]

Originally Posted by austin-steve

That too.

Think of a train with many personal mini "cars", traveling like a convoy, yet not connected or restricted to fixed tracks (just roads) and the ability to split off or join up the main flow as needed.

Unless you find a way to attach the cars together and operate them as a single unit (like an actual train), you will never get that "train" effect at higher speeds due to the need to maintain safe stopping distance.

Originally Posted by austin-steve

We already have "driver assisted" cars, mostly safety features such as rear end avoidance, and auto parking for the metro-sexual dads who can't parallel park old school, but also auto-cruise-control that keeps the space between you and the car in front without you having to toggle the steering wheel +/- all the time. I think "lane drift" warnings are in some cars now, as well as blind spot warning. Also, smart-GPS technology that knows realtime traffic and can recommend alternate routes.

None of which do a darn thing for capacity, which is the primary problem with the Austin transportation system. In fact, driverless cars will reduce capacity on the whole.