upvote
While this is going to be an overly optimistic scenario: Imagine how smooth a hurricane evacuation would go if _everyone_ used a self-driving car to do the evacuation - atleast there might be less gridlock than there is during any usual hurricane evacuations. And assuming the self driving cars don't do something stupid that causes every car behind it to essentially lock up and stop moving

That said, I know a scenario like that would never happen, probably for the best.

reply
The problem is they're not designed for that. They aren't spending resources on some master control networking system because in 99% of use cases that won't be useful anyways as most of the traffic being dealt with isn't other waymo's willing to communicate.

There might be some level of adoption where they would, but honestly we're back to "but what about trains/trucks?".

Half the problem with evacuations is people don't want to leave behind their stuff to get destroyed. You'd basically be better off getting a fleet of semi's with some quick and dirty cube system thrown up than a bunch of automated sedans.

reply
Sort of. There is no built in support for evacuation methods, but the WayMo absolutely does use a master control system for network the cars. This is how the database of streets is kept and is why WayMo vehicles occasionally swarm private non through way ally streets when there is some glitch in the database that indicates private ways are available roads or an ally that looks like a through way turns out to have a fence between properties.
reply
> atleast there might be less gridlock

I've never lived in a hurricane area, but when I think of news coverage of problematic evacuations, they're showing people stuck on highways, not people stuck in urban traffic grids.

It's a throughput problem. Computer controlled "car trains" with shorter following distances can boost traffic throughput, but I don't think that would be enough to make evacuation of large cities actually feasible. The highway system is simply not built for that use case. Especially since evacuation often occurs during inclement weather that reduces capacity.

AFAIK, most places try to figure out how to make shelter in place work, because mass evacuation is likely to end up with many people facing the weather event while on the highway.

You could theoretically do better with busses and trains, things, but there's likely not enough busses that are setup for long distance travel available: lots of municipal bus fleets are setup for alternate fuels which is great for emissions but makes it hard to travel to a neighboring state, because there may not be appropriate fueling opportunities on the way. Etc, etc.

reply
I’m from a flood prone, hurricane prone area - there were some painful lessons from Hurricane Andrew, famously the hurricane tie system for buildings in Florida which quickly spread, but in South Carolina they also learned a very important lesson - reverse all lanes on the interstate so everyone can flee as quickly as possible. The “stuck on I-26” problem no longer exists. I’ve personally driven 100+ miles in “the wrong way” to evacuate. It’s quite fun. They also perform statewide annual drills to make sure all emergency staff can faithfully execute this reversal pattern.

Do other states not do this?

reply
Why would there be less gridlock if people were in a driverless car instead of a regular car?
reply
With human drivers: traffic light turns green. The first car starts driving. The 2nd car waits 2 seconds and then starts driving. The third car waits another 2 seconds (4 seconds total) and then starts driving. The fourth car waits another 2 seconds (6 seconds total) and then starts driving. etc.

With computers driving: traffic light turns green. All cars simultaneously start driving. It'd be like a train but without the efficiency.

Similarly, with human drivers: some jackasses drive into the box and the light turns red. Now perpendicular traffic is either fully blocked or must proceeed slower to maneuver around the jackasses. With computer drivers, they shouldn't intentionally break the law and they should have plenty of sensors to figure out that they cannot make it through the box.

reply
Safety margins still will require some level of delay between cars that aren't mechanically linked. Even with perfect reaction times, the physics of driving (maximum acceleration rates, possible loss of traction) dictate this, it's a non-trivial control theory problem. Besides, it doesn't seem to be a goal of Waymo; I've seen lines of their vehicles before and they all behave the same way as in mixed traffic.
reply
As a sorta informed outsider, conceptually this makes intuitive sense. But in practice, how does this work? It seems a lot of the intuition breaks down if we don't assume it's network (aka 1 vendor). Fundamentally it's a bunch of external actors where we cannot verify trust and in order to solve for the needs of the individual, suboptimal choices must be made. To put it another way, even if computers can drive cars, what _else_ needs to be in place for this vision?
reply
Ideally, robot drivers will some day be better drivers than humans in all road conditions. They'll be able to coordinate fast lane merges and busy intersections by subtly adjusting speed without vehicles having to stop.

Imagine a busy intersection where all the cars fly past one another at 40 miles an hour without stopping but none of them crash. Humans can't do this, but machines could, if, and when the technology gets there. To be clear, there's still a way to go.

reply
Evidence suggests... no, that day is never coming.
reply
Once all cars are autonomous, that day is certainly coming. Even before then, it's very likely we'll see platooning in the future, even if there are still some human drivers.

Also, this already exists in some places. Look at a video of how to cross the street as a pedestrian in Vietnam: You literally just start walking across and people weave around you. Or look at driving in India and similar places.

All I'm saying is never say never

reply
Right… any time now.

If you want to write with such confidence perhaps you should share what the lottery numbers are?

reply
Never is a long time though. Even if it takes 500 years for that to happen, it will still have happened.
reply
busy intersections have more than just cars, my jay walking is going to cause a massive pile up
reply
Traffic is usually caused by adding inefficiencies across a system with little slack - someone brakes too hard or too early, and if all the cars are stacked up, that one brake event can ripple through hundreds of following cars, getting worse and worse because each person brakes more. Self driving cars can perfectly sync up and move like a train. Theoretically there could be no traffic on highways if all cars are self-driving. Rarely is a highway so full that there couldn’t be more cars (eg. The entrance ramps are backed up) which implies the issues are related to the driving flow and not the capacity of the street itself.
reply
> Rarely is a highway so full that there couldn’t be more cars

Yep, here in Chicago you might even go as many as 12 hours between such events

reply
Self driving cars don't panic and drive recklessly. I don't live in hurricane country, but most accidents around here are caused by drivers who are on their phones/spacing out or driving super aggressively.

Most traffic jams are caused by accidents or people slamming the brakes

reply
In principle the driverless cars are more able to organize fleeting, operating in a way that's not actually practical if you don't share a single guiding directive.

I don't know that you'd ever see this in practice, but it's much more practical in theory for almost identical machines running the same software than for a bunch of humans in a variety of vehicles who've maybe only half understood how to do this.

Also, for this specific problem we know humans are idiots. They should all be driving an agreed route to the agreed evacuation point, but some real humans will decide they know a shortcut, they want to drop past Jim's place, or whatever. Just as there's a difference between what the protocol says happens when you have to abandon an aircraft on the tarmac versus the reality that people will decide they want to self-evacuate and they need their carry on bags and chaos ensues and maybe people die.

reply
Same reason there's less gridlock when people obey traffic lights and other rules of the road and don't brake randomly. If every car on the road drove itself then there would never be traffic.
reply
This is literally not true, roads still have finite capacity, and sometimes demand exceeds capacity.
reply
Well, probably not the current generation of driverless cars. Those would be a nightmare. Contrary to what some want to believe self driving cars do random shit all the time.

But in the future, if there is a coordination standard among driverless cars, that could allow much higher density at higher speed. Coordination standards + higher density of self driving should reduce the self driving cars doing random shit too.

reply
"assuming the self driving cars don't do something stupid"

This is a big assumption.

This requires that all cars are self-driving cars capable of complex reasoning on in-car compute without relying on network connection, as network connections can't be assumed reliable in hurricane conditions.

reply
Now imagine if the power is out and cell service is down. We saw that happen in San Francisco and it was chaos.
reply
That's why on-board sensor only systems are the way to go.
reply
It would be a failure. Turns out they do something stupid. People tested this in sf by calling a bunch of waymos at once for a prank, but I guess that is the best case example of what a panicked evacuation on the service might be like. It was like a ddos attack. They ended up gridlocking themselves and turned it into a real life version of one of those rush hour board games. No one got out of the little area they called the waymos in.
reply
I mean the logical conclusion is a dedicated lane for automated cars..

At which point we've reinvented privatized buses with a last mile convenience vs greatly reduced throughput trade-off.

reply
I doubt it's less actual throughput in most cases. In a place like Atlanta there's no place where it's bus after bus. The BRT line they built nearby is a bus every 10 minutes. Which being very generous to the bus usage is equivalent to like 5 cars a minute.
reply
Just take away the sidewalk and bike lane :-/
reply
Evacuation is a use case in my mind. Having a fleet of shuttles on command to move people in preparation of a hurricane would be a benefit. They would obviously need to put weather limitations during actual storms because no one should be driving in a hurricane.
reply
Evacuation you want to prioritized throughput - think of how little road space 100 people in a bus take up vs say 50 cars with 2 people each. Or even 25 cars with 4 people each.
reply
If you have central control you might even be able to get away with changing the rules. i.e. most roads are now one-way leading out of the city. voilà we nearly doubled outbound throughput. Even just for commuting that would be awesome, not that it is happening anytime soon, but one can dream, especially while sitting in gridlock traffic.
reply
Having the middle of five lanes change direction depending on the hour is fairly common. There's even a dedicated machine to move a concrete barrier to support this.

https://www.youtube.com/watch?v=8IwBJPqyoB8

reply
> No one should be driving in a hurricane.

I agree, but there are a number of people here in Florida who will do it or die trying (emphasis on the die trying)

reply
Except the Waymo can do 150 mph bumper to bumper with other Waymos if you let them.
reply
Under an idealized situation sure, but I could get a 150 mph train of cars following me 60 years ago too if anybody had a use for that.
reply
.. well until it hits the flood
reply