May 02, 2010
Bob Lutz Foresees Automated Cars
Retiring (again) 78 year old GM executive Bob Lutz says cars in the future will be totally automated pods that form into chains. The romance of handling the thundering horsepower under the hood is going to come to an end.
Lutz has a vision for the future of transportation, and it's not necessarily a car guy's dream.
I say this without a great deal of joy, OK, but it's going to be individual transportation pods that are charging in your garage. You hit one of your normal programs, like your "go to work" program. And the thing goes out into the street--GPS-guided or wire-guided--blends into traffic, consolidates, goes to some sort of mass-transit station, converges onto a train and parks itself along with all the other little modules. At some station, your module gets off. There will be no driver involvement.
The automobile is a transient stage in the evolution of mankind's ability to transport oneself to any other place rapidly. And, ultimately, as we evolve, I think physical travel will become less and less important. Because if you can bring 95 percent of the experience in virtually--in, let's say, a 360-degree, full-screen, three-dimensional environment--you get 95 percent of the experience instantaneously. The need to travel goes down drastically. And ultimately I don't think humanity is going to be able to afford a situation where every citizen of the Earth has a car and is zooming around all over the place on concrete highways. It's just not going to work, and at some point it's going to stop.
The interview provides some insights into what went wrong with bureaucratic General Motors. Worth reading in full.
I'm sure Bob Lutz is well informed about how things are moving in the R&D department of the big car companies. Cool to hear it from him.
Cars are a transient stage, but not ephemeral. I think we'll still be using cars for at least another 30-40 years, although they're likely to go driverless within the next 10 years. I think you're right about tactile life-like VR and how it'll reduce the need for physical presence, but driverless cars aren't that far away, and the 95% sensory experiences could take a lot longer to go main stream than the pods.
I have this vision of a chain of pods cruising the beach at Corpus Christi or Ft Lauderdale...and the pick-up pod whirring its way across a snow-swept pasture to deliver hay to hungry cattle.
low-tech magazine has an article about a system of underground freight pods similar to this: the track moves all the time, and each podcar just enters at sendoff and comes off at delivery point; the things go very slowly, but without stopping, which beats trucks and stoplights on the surface of whatever Dutch city it is.
I would instead of Lutz's vision rather see a far simpler system: cheap electric vehicles or NEV's (powered by lead-acid batteries; in essence covered golf carts unable to go more than 35mph, limited range but good for getting around town) for local, with electric rail for everything else.
The big problem with radical restructuring efforts is building the infrastructure to handle them. These auto-pods would need light rail everywhere, but traditional electric light rail would only need inter-city lines, not too much more than the old rail lines we have now.
Most people these days would rather be checking Facebook on their iPods than looking at the road. The Lutz vision is the future, though maybe rather far future.
1) The technology to guide the vehicles is essentially here to pull it off.
2) The technology to do other things while commuting is here.
3) Safety is an ever-improving priority.
4) Automated driving would improve fuel efficiency, particularly if it relieves traffic congestion, and the slinky-like jams.
5) Lower insurance costs.
The problem it seems is the chicken-egg question of how you migrate to this solution in a piecemeal fashion. China will probably leapfrog us on this.
The problem it seems is the chicken-egg question of how you migrate to this solution in a piecemeal fashion.
You incrementally design in navigation that doesn't depend on an infrastructure, or other cooperating cars.
For instance, you develop cars that park themselves; proximity sensing sensors; cruise control, etc, etc. Gradually more and more functions are taken over by relatively simple forms of A/I, until one day you notice that you don't have to pay much attention to the road anymore.
Infrastructure starts to develop where it's cheapest, like on limited access freeways and dense urban area; as a critical mass of cars develops, they start communicating and cooperating. Gradually manually-driven cars begin to be seen as primitive and dangerous, and they begin to be excluded from certain zones, and then the zones expand.
Nationwide traffic network would be a prime target for hackers.
Why wouldn't existing traffic signals be just as vulnerable?
If it's not safe to use complex, networked systems to run our society, we're already done for...
One factor which seems to be ignored: when cars can drive themselves, we won't need nearly as many parking lots, because the cars can either go home, park at a satellite lot far away from anywhere anyone would want to go, or (better) go find someone else nearby who needs a ride.
For that matter, self-driving cars will probably see "the end of" personal car ownership. (I put quotes around "the end of" because obviously some people will still want to own them and this will still be possible, but not owning a car will no longer be the barrier to personal freedom that it is now in most of the U.S.)
It may also mean the end of city buses, since who would take a bus (full of all kinds of people you might not feel like sitting with) that only stops at your location every half-hour when you can summon a car that will be there in five minutes, and take only you (and anyone else you're with) directly to where you want to go?
What would be really cool is if cars could be made dual-mode so they could ride on train-tracks as well (and set up an interface between the car's self-guidance system and the railroad's scheduling system; you'd pay for access to the rail like you might pay for a toll road now, only it would all be done wirelessly and electronically)... that would make long trips much more efficient, without sacrificing the conveniences of a vehicle that can go on regular roads.
You are missing one big wrench here. Liability. Who gets sued when your car crashes? The buyer, the guy who designed the system? the car manufacturer? The lawyers will have a field day. Getting over the liability issue is a large hurdle in deploying any new technology. The obvious example is motercycles would never be approved today. To dangerous.
Who gets sued when your car crashes?
The firm which deployed pod-cars to that city, which will most likely have liability insurance. Also, motorbikes are not dangerous IF you wear whole-body protective equipment.