Hacker News Reader

Comment by lukas | Hacker News Reader

I would love to see the subways in SF improve but maybe self driving cars changes the calculus? Could this become a tealistic alternative to public transportation? I'm sure some people here have thought about this a lot more than I have - want to weigh in?

View on HN

Self driving cars still require a significant amount of surface area that could better be utilized for: housing, bike transit, pedestrian areas, parks.

And though coordination by autonomous vehicles could cut down on highway congestion, speeds within the city are unlikely to match what a subway system could offer. You don't want cars rushing through your neighborhood at 60 mph, no matter if it's a machine or a human behind the wheel.

If gridlock is a major issue now, self-driving vehicles won't help in a major way, since you still need X square feet to move Y people (up to a maximum M, as defined by the size of a vehicle) Z miles. Pushing more individual cars onto buses would help in this situation, but not as much as a dedicated subway line that is less impacted by vehicular traffic flows.

Self-driving cars would help by removing some drivers not paying attention and causing additional congestion (eg: not moving on green lights or causing accidents).

In theory, intelligent self-driving buses could make a difference, though. Still less than subways though, of course.

Here's a nice image that makes your point:

http://www.wearetraffic.org/sites/default/files/images/Bike_...

Self driving cars would allow minibus routes that just go between a neighborhood to a local grocery/retail center. This would make carless living more possible.

Would self riding bicycles help the situation perhaps?

The problem there is density. If it's still mostly 1 or 2 people per vehicle you haven't really solved anything.

At first, sure. Eventually, though, self-driving cars really blur the line between public and private transportation.

You could have a large network of self-driving cars that you simply hail on your phone and quickly pick you up. It won't look like a 4-door car, but more likely a small bus like you find in many parts of the world. People going in similar directions travel together, only picking up and dropping off with minimal interruption. In lower periods of demand, these are stored in lots with no wasted space as only a LIFO 'stack' is needed.

Furthermore, self-driving cars could be programmed to move in caravans with minimal following distance and aerodynamic efficiency to boot. Not terribly helpful in an urban situation (probably some minor traffic density increase), but wonderful for commutes which could dramatically decrease density.

So, you'd not only increase occupancy (easily 8 people in a car-sized footprint, presuming electric cars), you could also increase traffic density. All told, an order of magnitude capacity increase seems within reason, and in a reasonable amount of time.

That said, I think subways are still a good investment for the time being.

The thing is, all that stuff is a long ways off. Something to work towards for sure, but we still need mid-term solutions. Self driving cars that are empty and come pick you up will, if anything, lead to more traffic on the roads initially since now you've got more vehicles than people.

Yes, pools and lines are rapidly changing the options here and allow you to pay a premium for convenience and rapidity - to potentially reduce the number of cars on the road.

These might be helping to encourage last mile usage, the questions are whether it's A) available B) successfully reducing car ridership.

Reply on news.ycombinator.com