8 ms·
So the scenario (as described by waymo, surely with some spin): "Driving through a 4 way intersection, while a large truck is also going through the intersecti
by km3r 3y ago
So the scenario (as described by waymo, surely with some spin):
"Driving through a 4 way intersection, while a large truck is also going through the intersection. As you go through, a biker quickly passes behind the truck with minimal time to react once in sightlines."
A human driver would take significantly longer to react, and could be deadly.
Waymo managed to hit the breaks before the collision, potentially leading to only minor scrapes. Feels much better than human would fare in a similar scenario.
What I hope Waymo learbx for this, as well applies as similar scenarios: never cross through these zones of high bike traffic when the obstruction from another vehicle prevents a wide enough view to brake before this zone.
- nerdponx 3y agoLet's wait until there's a neutral 3rd-party description. https://www.caranddriver.com/news/a35844915/ntsb-letter-nhtsa-self-driving-vehicles/ https://www.caranddriver.com/news/a35844915/ntsb-letter-nhts...
- zone411 3y agoWe shouldn't have to wait and listen to the spin from both sites. The video should be released immediately. What's Waymo's excuse?
- ugh123 3y ago>What I hope Waymo learbx for this, as well applies as similar scenarios: never cross through these zones of high bike traffic when the obstruction from another vehicle prevents a wide enough view to brake before this zone. I hope it's as simple as that, and that regulators will understand just how little fine-tuning could yield substantial safety increases not possible with human drivers.
- clbrmbr 3y agoI hope regulators watch video from the vehicle and think, “damn I would not have been able to react that fast. Sure we can make these things even safer but it would be immoral to block waymo’s operations as it would certainly increase harm.”
- weaksauce 3y ago> damn I would not have been able to react that fast. i apply my own standard of driving to it: I wouldn't get in that position in the first place.
- seanmcdirmid 3y agoI too dislike driving in SF.
- harimau777 3y agoIt seems to me that the best metric might be whether the average driver would get into that position or react faster if they did.
- weaksauce 3y agoI don't know if i'd ever feel good enough to get in an autonomous car if they were just a bit better than the average driver... the average driver is not great.
- nicbou 3y agoNot only that, but fine-tuning is cumulative. Test situations can be created and tested against forever. Lessons can be learned and applied across the whole fleet. This is radically different from sharing the roads with a bunch of Sunday drivers, some of which might be distracted or impaired.
- henriquez 3y agoYes, it totally makes sense for bike commuters to be run over and sent to the hospital by robotic driverless vehicles while billionaire entrepreneurs refine the AI models powering them — it’s a beta test after all, you can’t actually expect it to work on public streets. And I bet those cyclists weren’t even paying their road tax. The needs of the 95th percentile that the model was tuned for outweigh the needs of everyone else, who can’t even afford a basic RWD Tesla Model 3 (which is pathetic and the most pedestrian of EV options since I drive an EV GM Hummer and I post on Hacker News)
- brian-armstrong 3y agoHow many cyclists are hit per million Uber rides?
- henriquez 3y ago[flagged]
- harimau777 3y agoI think that line of reasoning would only apply if the driverless vehicles are not better than human drivers. That is to say, it's perfectly reasonable that bike commuters would die as the software is improved as long as less bike commuters are die than would without the driverless cars. To be clear, I don't know that I necessarily think that driverless cars currently are better than human drivers. I'm just pointing out that the logic changes if they are.
- henriquez 3y ago
- jsyang00 3y ago> What I hope Waymo learbx for this, as well applies as similar scenarios: never cross through these zones of high bike traffic when the obstruction from another vehicle prevents a wide enough view to brake before this zone. This might be fair, but at some point, if you are applying this consistent level of caution, there isn't going to be anywhere in the real world left to actually drive
- whyenot 3y agoYes. I appreciate the political reasons for wanting zero accidents, but fewer, less deadly accidents than if a human were behind the wheel is what the goal really should be.
- lex-lightning 3y agoThat’s such a low bar though. Zero is achievable. Part of it is to slow cars down and take more cars off the road rather than changing who drives. Invest in public transit, redesign problematic infrastructure, and create mixed use urban areas so people don’t feel the need to travel through a city. A city should be a safe place. More deaths in a slower zone than on freeways is preposterous. I understand the reasons why this is true, but the deeper reasons for the reasons are that we gave up. Fewer, less deadly automobiles is what the goal really should be.
- petesergeant 3y ago> A city should be a safe place Is that a fundamental attribute of a city?
- illirik 3y agoOf course; that's why our ancestors first built walls around granaries.
- lex-lightning 3y ago
- itake 3y ago> When they became fully visible, our vehicle applied heavy braking but was not able to avoid the collision. The "fully visible" part made me thinks humans can identify a bicycle and rider with just partial visibility, which means they would stop faster than the car.
- roenxi 3y agoBased on what? Humans run in to cyclists too and although I have know idea about Waymo's tech, in a research setting computers are better than humans at figuring out what objects are based on visual data.
- al_borland 3y agoIf computers are so good at this, why am I (a human) asked to identify bicycles and motorcycles to train these computers when confronted with a captcha? A human needs far less training and can identify a person on a bike without ever seeing an actual bike. They can also infer what is likely to happen based on a lot of context clues. Even if an event doesn’t seem likely, I can still prepare. For example, if I see kids playing basketball in a driveway, I’m going to slow down, incase the ball bounces into the street and a kid takes off after it without looking. I don’t need to wait for the kid to be in the street to start reacting and prepare for different possibilities.
- ganieldackle 3y agoA computer doesn't need a multi million dollar marketing campaign convincing it not to text while driving or drive drunk. >I don’t need to wait for the kid to be in the street to start reacting and prepare for different possibilities. Neither does a computer.
- patrick451 3y ago> A computer doesn't need a multi million dollar marketing campaign convincing it not to text while driving or drive drunk. A human doesn't get random bugs every week from OTA updates. The human mind is not a viable target for hackers. The risks of human drivers are individual. The risks of computer drivers are systemic and massive. One bad bug or hack could kill millions if self driving cars become ubiquitous.
- belltaco 3y agoLooks like the Waymo vehicle was turning after an all-way stop sign? Is this a good representation with the blue car being the Waymo and the truck being the red car? https://taylorkinglaw.com/wp-content/uploads/2019/09/4-Way-Stop-4-1024x578.jpg https://taylorkinglaw.com/wp-content/uploads/2019/09/4-Way-S... Wonder why it was cutting so close to the large truck's back while accelerating during the turn that it couldn't see the cyclist till it was too late. Them saying the car braked only after the cyclist was fully visible is additional cause for concern. Those are precious fractions of seconds. Doesn't Waymo use LIDAR? Shouldn't it detect the partial cyclist as a solid obstruction and start braking before the object recognition kicked in and recognized it as a cyclist? What if it was a trailer?
- TaylorAlexander 3y agoYeah. Seems like good driving means "don't drive in to places you cannot see". If you are just accelerating after a stop sign, there is no reason to have such high speed and urgency when making the turn.
- cozzyd 3y agoRight there can also easily be pedestrians in the crosswalk who are also occluded, though I guess with a bit more stopping distance (but then you may be blocking a lane that would otherwise be able to go).
- lokar 3y agoWell, it sounds like the cyclist was doing the same while failing to stop or yield at a stop sign.
- belltaco 3y agoIf the description is correct then the cyclist was going into the space that they could see in front of them, unlike the Waymo. Someone else said that stop signs are legally equivalent to yield signs for bicycles in California. They probably couldn't see the Waymo because it was occluded by the truck, so how could they yield to it. This is why straight going traffic always has the right of way if the roads are similarly sized, and it's on the turning vehicle to stop and wait at the turn. I would also hope that self driving cars are programmed to drive defensively and conservatively over trying to shave off a second or two of drive time.
- manzanarama 3y agoI feel like you are injecting facts with a lot of assumptions. How do we know a human driver would have taken longer? What if the human would have seen the bike 200 yards up the road 45 seconds ago, and since bikes don't disappear, waited a half second before proceeding into the intersection. Or what if a small portion of the bike was visible? It could have been absolutely unavoidable but I don't think we know that now.
- dharmab 3y agoIn my experience with motorcycling and bicycling as a primary mode of transport I'm happy if the human drivers attempt to stop at all
- websap 3y agoCyclists are supposed to stop at stop signs, if not stop at least yield. At least in this case (if true), the cyclist followed the truck through a 4 way intersection (assuming it's a 4-way stop), that would put them in the wrong. Hope waymo releases the footage.
- yellow_postit 3y agoThere’s rules as written and then rules as followed — I’d much prefer a system that recognizes the rules that people tend to follow/bend/break — as a cyclist I too will often “convoy” with a bigger vehicle as it provides some additional protection most of the time (though obviously not here)
- fortran77 3y ago[flagged]
- thaumasiotes 3y ago[flagged]
- adrr 3y agoWaymo and Truck in intersection both heading in opposite directions. I dont get where the bike was. It says behind the truck. So did bike run the stop sign and cross into waymos lane because that is way it sounds. Article doesnt state who's at fault.
- tfehring 3y agoYeah, my reading was that the bike ran the stop sign and then attempted to make a left turn in front of the Waymo as it was going straight through the intersection.
- belltaco 3y agoMy reading was that the Waymo was turning left. Remember, the description is from Waymo's PR, so that's why it's vague on purpose and needs to be taken with a grain of salt. According to another comment bikes can treat stop signs as yield signs in California. If the bike illegally ran a stop sign I would think Waymo would say or imply that.
- adrr 3y agoDid a quick google and DMV states they must stop at all stop signs. https://www.dmv.ca.gov/portal/driver-education-and-safety/educational-materials/fast-facts/sharing-the-road-ffdl-37/#:~:text=Obey%20Traffic%20Signs%20and%20Signals,the%20light%20changes%20to%20red https://www.dmv.ca.gov/portal/driver-education-and-safety/ed....
- tfehring 3y agoI agree that there's still lots of ambiguity and Waymo's PR comment provides limited and one-sided detail. I stand by my previous comment as the most likely course of events, but we won't know for sure until video is released. The comment that claims bikes can treat stop signs as yield signs in California is incorrect. From the CA DMV website [0]: > Bicyclists must obey STOP signs and red signal lights, and follow basic right-of-way rules. There was an attempt to change this a few years ago, but it didn't pass. [0] https://www.dmv.ca.gov/portal/driver-education-and-safety/educational-materials/fast-facts/sharing-the-road-ffdl-37/ https://www.dmv.ca.gov/portal/driver-education-and-safety/ed...
- Retric 3y agoI think the most critical bits are left out here: “The cyclist was occluded by the truck and quickly followed behind it, crossing into the Waymo vehicle’s path.” Was the Waymo car turning or was the cyclist? Because if the cyclist swerved into oncoming traffic there’s likely little any driver could do, but if the waymo car was ignoring the possibility that a cyclist was behind the truck that seems like a significant safety issue.
- cozzyd 3y agoI feel like Waymo would have said something if it was the former, no? Since there's a bike lane on 17th street, it's possible that the cyclist entered the intersection parallel to the truck on the bike lane but the truck overtook it. It's even possible that the cyclist was in the crosswalk for some reason... edit: according to https://www.theverge.com/2024/2/7/24065063/waymo-driverless-car-strikes-bicyclist-san-francisco-injuries https://www.theverge.com/2024/2/7/24065063/waymo-driverless-... the truck was turning?
- jiveturkey 3y agoAgree with sibling. You are speaking out of turn in that you have no idea how long a human would take to react, as we don't know the situation completely enough from this article. That said, my takeaway is that the cyclist likely ran their stop sign. This is a very common accident situation even between cars, where the view is obstructed and the at-fault driver imagines that their path is clear. It seems very likely that the cyclist did a similar kind of thing, but with the additional aspect of ignoring the traffic control. Since I'm here: brakes not breaks.
- tdeck 3y agoWaymo probably has cameras from every angle here, so it seems kind of odd that we have to rely on a textual description.
- tialaramex 3y agoPrivacy concerns mean that Waymo has a strong reason not to release video to us, the general public. My guess is the most likely two routes you get video is either there's a lawsuit and so it's evidence or a government agency insists on having video and that gets either deliberately released or leaked to excited local TV news. In both those cases Waymo can say they had no choice, even if it's good for them.