Hacker News Reader

Comment by alooPotato | Hacker News Reader

To have the city planning work correctly you kind of need participation from the cars. I.e. if the traffic light infra structure knew the destination of every car it could do things more globally optimal.

So in that sense, driverless cars will actually help because of kind of nuanced reason. My guess is that there will only be a small number of driverless car companies (tesla, waymo, cruise, apple, etc) and it'll be much easier to add regulations that those companies need to send their destination data in realtime to the traffic light infra. I think that would basically be a non-starter for peoples personal cars.

View on HN

> To have the city planning work correctly you kind of need participation from the cars

Perhaps for the optimum solution, but you can get a _lot_ better (than things are at the moment) without that. There are plenty of cities that use sensors and cameras to adjust traffic light timing throughout the day, and have done for some time (In the 90s I studied the system used in Sydney, which had already been in existence for some time then). But these things cost money.

Also, you don't need driverless cars for destination data - anyone using Google/Apple Maps/Waze, or even any recent car's inbuilt nav system can provide this.

And I reckon that even with the optimal solution in this regard, you'll still end up with traffic jams, as individual cars take a lot of space. Start thinking in terms of mass transit, cycling, and city planning to allow walking to the shops.

You're right the sensors help a lot too - but if a city doesn't have them then thats physical infra that needs to be bought and installed. Unfortunately, this isn't that easy any more in the US at least.

Maps/Waze data is a good point and I saw an article recently that Google is providing anonymized data to cities to help them optimize their traffic lights.

The one thing the driverless cars get you is the ability to traffic shape. You could imagine cities granting permits to waymo/cruise/etc only if they support dynamic pricing (or dynamic taxes, whatever) and dynamic routing to help distribute vehicles more evenly in space and time.

On the public transit thing... mainly because shared driverless EVs are basically a form of mass transit and I think a really good one. You can see some of my other comments on why...

Maps/Waze can also shape traffic by suggesting different routes.

One interesting half-way-there solution (and this is a technical solution involving cars, not reducing them as I'd much prefer) would be to have nav apps be able to read the traffic light timings for the intended route, and give you a suggested speed to travel at to maximise green lights and/or traffic throughput. It would piss off cars that want to go exactly at the speed limit, but would probably end up getting them to their destination faster.

In both cases, cities could offer incentive schemes (e.g. points -> $ or discounts off licence/registration fees) to encourage people to 'follow the computer'. Would probably be cheaper than building infrastructure.

The streets are otherwise empty - they don't need any data other than there is a car that just left a red light - it will most probably get to the next light in (calculate based on the speed limit) so we can turn the light in time. We typically have sensors in the road so we can detect which lane the car is as it goes through the next (now green) light and so make predictions and turn the next light green (sometimes two lights - one if it goes straight, one if it turns). Sure when there are two cars it becomes more difficult, but even then we can do a lot to time lights so most of them are green if you go straight.

Self driving car technology could be (at great expense) be used on all the cameras in a city to get a better picture of where people are really going and thus turn lights green to cause the least waiting.

Most cars sit at a light burning fuel and thus emitting CO2. Then they use a lot of fuel (read CO2) to get up to speed and at the next light that fuel is turned into heat by the brakes (hybrids can do some recharge, but they are not 100% efficient so there is still a lot of CO2 for no work)

That is how the lights in SF work. Presumably the Cruise car was traveling in a place where that was not set up, broken, or perhaps just slightly slower than the normal rate so it would always miss the lights.

I don't think that's true. Often times at night, there are long times for certain lights so the bigger throughput street can have more time on there. Human drives know about this, and will take different routes to get there. It's unclear that Cruise is mapping out light timings and their changes, although you would think they would.

yes lights let higher throughput roads have more time. i don't see why it makes my statement not true.

I take point with your use of 'correctly' because traffic engineering cannot get every car to its destination without stopping it even if it knew every destination.

Cars already do talk to various traffic lights via underground sensors and occasionally a camera. The investment required to have a fully networked traffic light system with a centralized controller would be immense, and would create an organization with a natural government given monopoly.

Your use of 'correctly' paints a world that seems ripe for more corruption and ultimately worse timings across the network. If you wrote 'better', then I wouldn't have commented. I think I'm nitpicking

This is getting into traffic research territory - i.e. given the entire dataset (current and historic) can you shorten average trip time vs individual cars just doing whats best for them. Either way, having a real time data infrastructure seems only positive.

Traffic lights are already remotely controlled by city planning departments, it's the software that's missing. My whole point is that driverless cars, where the city enforces data sharing and maybe even some routing, can have a real impact on throughput times.

The best part is that most of this is just software. And not ~$2B to make a 1.7 mile subway line or 10+years to make a bus lane (see SF)

I agree with you that the idea seems intuitively like it would be better, however I'm skeptical that it will work and that it is the 'correct' way to run traffic.

To quote your own example, this will be controlled by the government, which takes a long time to do physical infrastructure projects. So I would lately just assume it's not worth the investment dollars.

That’s the best part. The pooled driverless EVs doesn’t require any physical infra for the govt to build or buy! It’s one reason I think we’ll get to a more optimal future vs public transit which requires big govt programs and building.

Waymo or Cruise cars on the road today aren't talking to any city infrastructure. The companies are not talking to each other. Heck even cars from the same company aren't communicating to try and coordinate some kind of traffic pattern or share data. The entire current approach to self driving has been a dead end from the start.

Yeah they aren't talking to city infra now. My point was that it will become easier to make that happen once a larger % of the cars on the road are self driving.

When will it become easier? After all these companies have gone bankrupt?

why do you think they will go bankrupt?

Because that's what happens to companies when they run out of cash to burn and still don't have a viable business. What do you think happened with Apple's massive self driving car division? Or Uber's? Lyft's? There are dozens of failed companies in the space already. Ultimately GM and Google are going to run out of patience as well.

i know how companies go bankrupt in general :)

In this specific case: - is there customer demand? Yes. ppl want to move around cheaply, safely and efficiently - is the technology capable? Yes. I've ridden in a waymo, it's great. - what is the first principle cost of a ride? Lower than uber (which is a succesful business btw). Sure there is extra hardware but that goes down a lot at scale and you don't have to pay a driver which actually lowers your costs more than you think (no breaks, a lot less hardware in the car, etc)

I don't see anything structurally wrong with this business that makes me think it has a higher lilekihood of failure than any other random business. Sure a particular company may not execute well and go under, but I think there will be some sucesfful self driving companies in the future.

I think that future is a long way off for lost places. We haven’t seen these really proven to be ready for most urban environments. We haven’t seen them perform well in randomized conditions (road work, emergencies, marathons, critical masses, etc). We see them do well in good weather and predictable traffic conditions.

I’d love to have better transit options, whatever that looks like (including driverless cars), I’m just skeptical of us getting there on a large scale anytime soon.

have you ridden in one?

I have not, they aren’t in my city and I doubt we’ll see them for a while (decent rain, hellish snow conditions in the winter, potholes abound)

then you can't say they are a long way off.

I have and it's working. Right now. In a city with crazy traffic. And construction. And techno pessimists putting cones over their lidars. In the rain. 24/7. It's fucking incredible.

Reply on news.ycombinator.com