Several of Waymo’s autonomous vehicles were seen stuck in the middle of San Francisco streets following a significant power outage that took out the city’s traffic lights. Waymo responded to the power outage by suspending its ride-hailing services in the city, but images and videos on social media showed the self-driving taxis stopped at intersections with hazard lights on.
“We have temporarily suspended our ride-hailing services in the San Francisco Bay Area due to the widespread power outage,” Suzanne Philion, a spokesperson for Waymo, told Engadget in an email. “Our teams are working diligently and in close coordination with city officials, and we are hopeful to bring our services back online soon.”
This is really bad. You need to have emergency service vehicles able to move around the city. Blocking road like that could mean life or death for some. Public road isn’t some playground for doing beta testing. Waymo needs to be heavily fined for putting public at risk.
Police and firefighters would love to have an excuse to go demolition derby on these things I bet.
Honestly, I’m happy they picked this as a default “car doesn’t know what to do” scenario. From what I’ve seen Tesla’s default is to just ignore the unknown thing, I wouldn’t be surprised if Robotaxis would have just treated all the blank lights as green.
What’s the default on regular drivers when the traffic lights are not working?
An intersection with the lights no longer working is treated as a stop sign.
I admit I scratch my head at 4 way intersections with blinking yellows on all 4 though. Usually the bigger road gets the yellows for caution, the adjacent lanes have to yield.
I’ve not seen an intersection with four blinking yellow lights in California; typically two ways are blinking yellow (caution) and the other two are blinking red (stop).
I’ll give you the benefit of the doubt and assume you don’t drive.
A LOT of drivers actually don’t know the answer to this one!
Sure are a lot of things to account for with these self driving car things. It would probably be way easier if they were separate from pedestrians and maybe limited on where they can go.
Trains. I want them to just be trains.
Adam Something, is that you?
I want horses
I think the best solution is more self driving cars and more Generative AI.
My favorite food is chalk.
Would you like some glue with that?
This is one of the many edge cases that I’ve been convinced will keep self driving cars from becoming mainstream unless/until true AGI is achieved.
A few years ago I stopped at a red light next to a construction site. I was watching the traffic light, so at first I didn’t notice a cop at the construction site trying to wave me through the red light. He finally took a few steps towards me and yelled to get my attention. Only then did I realize he was waving me through, so I did just that. I seriously doubt any current self driving car would recognize a police officer (and not just a random pedestrian) that’s overriding the traffic signal like that.
Another edge case, coincidentally at the same intersection a few years earlier was when there was a car fully engulfed in flames as I drove up. I could hear sirens in the distance, and the cars in every direction were making sure to safely get out of the way of the approaching fire trucks. At least one or two cars cautiously crossed on the red to get out of the way. Again, I doubt any current self driving car would have navigated that situation anywhere nearly as well as a human.
Easy. Give police the ability to remotely control driverless cars.
I mean, that is going to be a thing they will implement anyway, and probably not restricted to driverless cars.
If you give police a backdoor to control self-driving cars, somebody is going (to hack it and) use it to kill somebody.
Exactly. Any such remote control would have to be trivial for a cop to use, and also need to directly control only the car(s) the cop is currently interacting with. Think of a situation like this where a traffic light is disabled and a cop is there directing traffic. If driverless cars are approaching from multiple directions then how does the cop direct his commands to only the one he’s focusing on at the given moment? Not all that easy when you think about it…
Yes! Yes it is! Not even close to as big of a problem as giving police full remote control of our vehicles, but, that too!
You can already kill someone with a car without any hacking.
- Get a car
- Drive over someone
- “Oops accident”
It happens every day.
OP argued that giving police such powers is a bad idea because there will be NEW vectors for people to harm others — not that other, easier ways don’t exist already.
Then you come in and say, there are ways to kill people using cars already. Literally, the only thing we can say to that is, “yes, we know and what is your point?”
The commenter before me described the solution, give cops an override, as easy. I wanted to highlight that it isn’t that easy. Unintended side effects ought to be considered before coming up with seemingly easy solutions. And this problem is not dissimilar to the one about encrypted chats and law enforcement wanting a backdoor into that. If you build a backdoor, it’s not guaranteed that only the good guys use it. And that raises questions about privacy on the encryption front or questions about abuse, safety, and liability on the self-driving car front.
I’m not arguing any of that. Giving police access to a car override would open up horrible abuse of power but that wasn’t your argument. Your argument was that people would hack it to kill people.
I’m saying that it’s much easier to kill someone by just driving over them with your car and claiming it’s an accident. It happens every day and people walk away without any repercussions. No hacking necessary.
Do I really trust humans to perform much better when traffic lights are out?
I can recognize when a police officer is directing traffic at a dead traffic light. I can also recognize the intent of other drivers who may wave, flash their headlights, etc. I doubt any current self driving cars can accurately recognize any of those.
If the power had just gone out and the mind was in thought – I could foresee you, me, or some Karen blowing through the intersection.
I’m curious about what cloud service caused a couple of these cars to shit the bed. They’ve been designed to handle a lot of the compute locally. They can’t even been driven from a remote operator - the vehicle mostly troubleshoots sticky situations itself.
I can’t imagine this is the first time their fleet has encountered a traffic light that’s out or cell towers that’s are not responding. They do half a million rides a week in cities that have infrastructure blinking in and out of service constantly.
It wasn’t a cloud failure. The self driving cars are highly dependent on traffic lights being red/yellow/green. With the signals inoperative the cars don’t know what to do. Even if there were police officers directing traffic at intersections, the cars aren’t programmed to recognize & respond to them.
Even ignoring the police officers, aren’t there clear rules for what to do when traffic lights are turned off?
In Germany, an inactive traffic light means that traffic control reverts to any present traffic signs (stop/yield/priority road). If none are present, the default rules for entering an intersection apply (which in Germany are to yield to any traffic coming from your right).
All of those rules already must be implemented for autonomous driving so why the hell couldn’t they implement a hierarchy?
Even if there were police officers directing traffic at intersections, the cars aren’t programmed to recognize & respond to them.
That by itself ought to automatically disqualify any such driverless car for use on public roads.
Weird. If that’s the case, how hasn’t this been more of an issue? This isn’t the first time a light has gone out in SF.
And then on top of that, since a ton of people were then connecting to cell service since their WiFi was out, that meant the cell towers were so overloaded they couldn’t send data to operators that the car requires to be started up again, like multiple camera feeds, a 3d scan of the surroundings, etc.
So there’s a lot of assumptions in this thread, but this specifically is just wrong. The cars do not need to have a constant connection with camera feeds and logic flowing to and from hq. They do nearly all processing on the vehicle, and comms to hq is used for location, status, etc but absolutely does not require the logic sent remotely to actually drive the vehicle.
I can’t list sources because of an NDA (I am the source, nobody else is going to back me up), but I’ve seen the systems, I’ve been inside the AZ waymo hq, seem how hq interacts with the vehicles, the location and size of the compute system that does all the logic, I’ve seen how it works way beyond what the media has seen. I’ve rode in one of their test mules, with techs answering a slew of questions that I posed, and asked about the hardware and debug/test software that the public simply can’t see.
I’m not sure why this happened - most people spouting this or that are just wrong. As of a decade ago, the cars were capable of handling failed stoplight and situations like that. They are also capable of being remotely controlled - someone above claimed otherwise, but they absolutely can be. Only in situations where the car is stuck or acting erratically (you call the hq via a button in the car, and they can pull up the vehicle and see everything about it, and if necessary, take control).
Either someone broke something regarding this situation (handling failed lights, etc) that was previously working, and this is the first time the issue has shown itself… or the power outage hit their ca hq, and when the cars couldn’t stay connected for X amount of time, they went to failsafe mode. I’m leaning heavily towards the latter - there is very little data flowing between vehicles and hq (unless remotely diagnosing or controlling), but there is a bit (location, speed, status…), and maybe when hq went offline for a few minutes, it’s a safety thing (think about someone trying to steal a Waymo car, for example, by trying to sever the connection, physically blocking it in, etc). That’s just speculation though, but it’s all I can think of.
C’mon you don’t even have to read the article, just the blurb OP quoted to figure out it wasn’t cloud.
cell towers not responding definitely not a first time. happened years back because of a music festival, and there were like some dead cellular spots in san francisco that held them back a while back as well.
