Autonomous Waymo Car Runs Over Dog In San Francisco (arstechnica.com) 204
An anonymous reader quotes a report from Ars Technica: One of Alphabet's Waymo autonomous cars has killed a pet dog. TechCrunch spotted the public report of the incident, which says one of the Waymo Jaguar I-Pace cars ran over a dog in San Francisco while in autonomous mode with a safety driver behind the wheel.
Waymo's collision report says: "On May 21, 2023 at 10:56 AM PT a Waymo Autonomous Vehicle ("Waymo AV") operating in San Francisco, California was in a collision involving a small dog on Toland Street at Toland Place. The Waymo AV was traveling southwest on Toland Street when a small dog ran into the street in front of the Waymo AV. The Waymo AV then made contact with the dog, which did not survive. At the time of the impact, the Waymo AV's Level 4 ADS was engaged in autonomous mode, and a test driver was present (in the driver's seating position). The Waymo AV sustained damage."
The collision was a block from Waymo's Toland Depot, a 120,000-square-foot warehouse that houses at least 50 autonomous cars. The speed limit on Toland Street is 25 mph, according to posted signs viewable on Google Maps. From that Street View link, the road looks like a busy industrial area with many warehouses, truck delivery areas, and barbed-wire fences. The incident is Waymo's first reported fatality. Waymo sent along a statement: "On May 21 in San Francisco, a small dog ran in front of one of our vehicles with an autonomous specialist present in the driver's seat, and, unfortunately, contact was made. The investigation is ongoing, however the initial review confirmed that the system correctly identified the dog which ran out from behind a parked vehicle but was not able to avoid contact. We send our sincere condolences to the dog's owner. The trust and safety of the communities we are in is the most important thing to us and we're continuing to look into this on our end."
In early 2018, an autonomous Uber vehicle in Tempe, Arizona, hit and killed a woman. "According to Tempe PD, the car was in autonomous mode at the time of the incident, with a vehicle operator sitting behind the wheel," reported Gizmodo at the time. The company went on to suspend self-driving car tests in all North American cities after the fatal accident.
Waymo's collision report says: "On May 21, 2023 at 10:56 AM PT a Waymo Autonomous Vehicle ("Waymo AV") operating in San Francisco, California was in a collision involving a small dog on Toland Street at Toland Place. The Waymo AV was traveling southwest on Toland Street when a small dog ran into the street in front of the Waymo AV. The Waymo AV then made contact with the dog, which did not survive. At the time of the impact, the Waymo AV's Level 4 ADS was engaged in autonomous mode, and a test driver was present (in the driver's seating position). The Waymo AV sustained damage."
The collision was a block from Waymo's Toland Depot, a 120,000-square-foot warehouse that houses at least 50 autonomous cars. The speed limit on Toland Street is 25 mph, according to posted signs viewable on Google Maps. From that Street View link, the road looks like a busy industrial area with many warehouses, truck delivery areas, and barbed-wire fences. The incident is Waymo's first reported fatality. Waymo sent along a statement: "On May 21 in San Francisco, a small dog ran in front of one of our vehicles with an autonomous specialist present in the driver's seat, and, unfortunately, contact was made. The investigation is ongoing, however the initial review confirmed that the system correctly identified the dog which ran out from behind a parked vehicle but was not able to avoid contact. We send our sincere condolences to the dog's owner. The trust and safety of the communities we are in is the most important thing to us and we're continuing to look into this on our end."
In early 2018, an autonomous Uber vehicle in Tempe, Arizona, hit and killed a woman. "According to Tempe PD, the car was in autonomous mode at the time of the incident, with a vehicle operator sitting behind the wheel," reported Gizmodo at the time. The company went on to suspend self-driving car tests in all North American cities after the fatal accident.
Autonomous, not prescient (Score:5, Insightful)
Re: (Score:2)
Maybe there was plenty of room to stop safely. Maybe the dog was visibly unleashed and could be easily interpreted to be ready to jump into the street.
Impossible to tell without dashcam footage, but it's a bit miserly to demand that for a dog. So I will give them and the SF authorities the benefit of the doubt.
Re: (Score:3)
Re: (Score:2)
Impossible to tell without dashcam footage, but it's a bit miserly to demand that for a dog. So I will give them and the SF authorities the benefit of the doubt.
I don't agree there. I think it's important to know if it can recognize things outside of what's expected. You might not care if it can recognize a dog and stop, but what about a crawling infant or a toddler? Can it recognize those, or will it only stop for things it recognizes as adults? For that matter, just objects in the road in general. They might not be alive, but they certainly can potentially damage your car. Then there's the ancient debate about whether you should stop for small animals in the road
Re: (Score:3, Insightful)
I don't agree there. I think it's important to know if it can recognize things outside of what's expected.
What? It sounds like the car recognized the dog faster than a human would have, but apparently you expect AI to break the laws of physics (friction between tires and road). I mean, I support our robotic overlords as much as the next guy, but that seems rather odd.
Re: (Score:2)
So... all roads with people or animals within a few km or so? Yeah, that's a strategy.
You could use AI to scale the distance with how far the creatures travel daily... no, still stupid.
Re: (Score:2)
Re: (Score:3)
I never suggested anywhere that anything ever makes it okay to run dogs over. Not sure how you could read that into what I wrote, but you did. Is this place turning into reddit??
Here is what I wrote:
> If the dog was unleashed then the owner was feeling entitled or lazy
Here's what you wrote:
> Even well kept dogs can get off leash sometimes.
The part where I suggested the owner was "lazy" is accurate. If someone doesn't have a harness
Re: (Score:2)
> If the dog was unleashed then the owner was feeling entitled or lazy
That "lazy" at the end there covers dogs "slipping out", jumping a fence, zigging when the owner was zagging, children opening a door, and a whole constellation of methods dogs use to evade their owners to run out in the street.
Re: (Score:2)
Re: (Score:2)
Accidents happen when people are being lazy. It's really as simple as that, but you're needlessly complicating this because you're a troll and you have nothing fucking better to do. This is the last reply I'm making in this ridiculous, pointless thread. Go be pedantic somewhere else, your replies are pointless drivel.
Re: (Score:2)
Nah. I believe you are the troll. The sort of person who, though far from perfect, expects perfection from others. Enjoy your bridge, watch out for billy goats.
Re: (Score:2)
Jesus, your supporting argument is "what if there was an earthquake"?
If you have to go to such an uncommon scenario you really should be doubting your own point. Of course there are extremely low odds ways a dog could escape without it being the fault of the owner, that doesn't change the fact that the vast majority of time what the above said is accurate. All you're doing is trying to find extreme outlier scenarios so you can claim your petty victory.
Re: (Score:2)
Re: (Score:2)
If I hide and jump out in front of any car, Iâ€(TM)m going to get hit.
Dude! You alright? Looks like you already got hit. :-)
Re: (Score:3)
If I hide and jump out in front of any car, I & # 2 2 6 ; & # 8 3 6 4 ; ( TM ) m NO CARRIER
Re: (Score:2)
Re: (Score:2)
Nope,. It can stop exactly as fast as a human (unless you think AIs can affect universal physical constants), but they can usually recognize issues faster than humans and start braking earlier. Of course, that only matters if almost-instant reflexes are enough to avoid the problem.
Why do people think autonomous cars are Superman? It's creepy.
Re: (Score:2)
Re: (Score:2)
That's easy. Because, as I said, they can start braking earlier. But more importantly, because the cars don't become tired, distracted, or assume that they are more competent than they are. They follow road rules way better than humans, and don't put themselves into the Trolley Problem. They make safe choices way more than humans do. Face it, we're idiots. All of us. Those who believe they are not idiots are the worst idiots of all. And the injuries from cars proves this beyond a doubt.
There are oth
Re: (Score:2)
As for the dog. it is an industrial area and there is no mention of any parked car, or any weather condition or obstruction of the view. Is it possible that a human would have seen the dog 20 feet away and running towards the road but the car's sens
Re: (Score:2)
Indeed. Also, autonomous cars only need to be significantly safer than the notoriously unsafe average human driver to make sense. They do not need to be perfect at all.
Re: (Score:2)
This seems to be a fact-free story unless we can take a US corporation's word for it. I can't think of any I trust.
Just one? (Score:5, Insightful)
One car running over one dog, out of all the autonomous miles these things have already driven, is pretty damned good, I would say. Far safer than human drivers, at least so far.
Re:Just one? (Score:4, Insightful)
All the autonomous miles on nice wide straight US roads mostly on grid systems.
Lets get to one navigate through Rome or Mumbai in the rush hour and see how well it does.
Re: (Score:2)
Or Boston... San Diego...
Re: (Score:2)
Re: (Score:3)
In Mumbai it would stop and get no where in all the confusion... which makes it at least as good as human drivers in Mumbai.
But in any case Rome wouldn't be a challenge. Being able to stay in a lane was the first thing that they solved for autonomous driving. Paris would be a challenge, specifically the Arc de Triumph roundabout.
Re:Just one? (Score:5, Insightful)
Uhm. Maybe you have never been to San Francisco - where this accident happened? One end of Toland streed is a 5 way intersection. The city is a maze of one way streets, steep hills, not-grid intersections, no left turn streets, service and main streets, alleys, etc. This accident MAY have taken place at a grid-like area (I don't know offhand), but certainly the testing done in the city is some of the trickiest in the US.
I'm not saying that it'll be good enough to navigate Mumbai or China. But I'd give it solid odds in Rome. That is, once it's good enough at US roads - which may be a long time from now.
The tail is *infinite*.
Re: Just one? (Score:2)
Now add in 2 way streets with cars parked so only enough room for one vehicle in either direction at a time and drivers who ignore the rules. Good luck.
Re: (Score:2)
Re: Just one? (Score:3)
TFS says the system correctly identified the object as a dog.
Re: (Score:2)
TFS says the system correctly identified the object as a dog.
That's not correct. It only says it "correctly identified" the dog, not correctly identified it as a dog. More likely, it correctly identified that something had appeared in the street in front of it. Very unlikely it identified it as a dog. Probably not distinguishable from a plastic bag blowing into the street.
Re: (Score:2)
It only says it "correctly identified" the dog, not correctly identified it as a dog.
This has to be one of the dumbest arguments I've seen on Slashdot. It doesn't matter what it was. Call it hazard X, and move on.
Re: Just one? (Score:2)
Good point. It could be corporate doublespeak ... Without annotated dashcam footage we won't know. I hope the NTSB investigates all autonomous-drive collisions.
Re: (Score:2)
> A human would likely notice the dog as something more than a binary on/off blob on the side of the road.
- "The human operator didn’t see the dog, but the vehicle’s autonomous system did."
- "The vehicle was operating in autonomous mode, and a human safety operator was in the driver’s seat at the time of the accident."
- " Neither the safety operator nor the autonomous system braked to avoid collision, according to Waymo. In both cases, that’s because of the “unusual path
Re: (Score:2)
One car running over one dog, out of all the autonomous miles these things have already driven, is pretty damned good, I would say. Far safer than human drivers, at least so far.
Yep. How many dogs and small children were killed by human drivers that day?
Strangely enough: The press didn't report that datum.
The company went on to suspend self-driving car tests in all North American cities after the fatal accident.
They could hear the lawyers salivating.
Re: (Score:2)
Yep. How many dogs and small children were killed by human drivers that day?
Let’s not fall into the trap of saying that one sort of accidental death of a human child or pet is better or worse than another kind of accidental death. Each pet or human child killed in an accident is a waste of delicious meat that could be feasted on by always thankful trolls. Even lesser trolls deserve to devour human children as much or more than they deserve to be crushed at the hands of greater trolls.
Re: (Score:2)
Yep. How many dogs and small children were killed by human drivers that day?
Let’s not fall into the trap of saying that one sort of accidental death of a human child or pet is better or worse than another kind of accidental death. Each pet or human child killed in an accident is a waste of delicious meat that could be feasted on by always thankful trolls.
Won't someone think of the pet humans?
Meanwhile (and I do mean "mean") my Artificially Insatiable Robocar hunts the most dangerous prey of all...other cars.
Re: (Score:2)
> Yep. How many dogs and small children were killed by human drivers that day?
Hard one to answer, but on average, 510 children per day in the world, and 3,287 dogs per day in the USA.
Re: (Score:2)
Not only that, but there was a human in the driver's seat as well and it STILL happened. Accidents happen.
Re: (Score:2)
That's not a good argument. If the human is not actively controlling the car, it will take several seconds for the human to switch to car-controlling mode. That's one of the reasons Level-3 automated cars are a really bad idea.
Re: (Score:2)
While tragic for the dog, some accidents cannot be avoided. The litmus test should be are cars safer when they are automated or under human control.
Re: (Score:2)
fatal and injury accidents involving teslas are up significantly.
Re: (Score:2)
sigh link fail https://www.washingtonpost.com... [washingtonpost.com]
Contact Was Made (Score:4, Insightful)
I was mildly aggravated by the story until I read this irresponsible passive-voice press release, and now I think nobody at Waymo possesses the moral capability to be writing autopilot algorithms.
This situation required aggressive acceptance of responsibility and contrition. Did they have Bard write this?
Somebody's gonna make contact, alright, with one of these people when more than a purse dog is killed, and probably the aggrieved will pick the wrong target. Very sad.
Re: (Score:2)
>"I was mildly aggravated by the story until I read this irresponsible passive-voice press release"
I agree. The "made contact with" is just so slimy-sounding. The car hit and killed a dog. The end. It *is* going to happen, autonomous or not. Interesting, it is kinda the same irritation I have when people say "he passed over" or "passed away" instead of just saying he "died" or "is dead."
>"This situation required aggressive acceptance of responsibility and contrition."
Yes and no. It is quite poss
Re: Contact Was Made (Score:2)
I agree. The "made contact with" is just so slimy-sounding. The car hit and killed a dog. The end. It *is* going to happen, autonomous or not. Interesting, it is kinda the same irritation I have when people say "he passed over" or "passed away" instead of just saying he "died" or "is dead."
Why? It reads like a typical incident report where the focus is centered on one particular subject, and everything else is described as it relates to that subject. This writing style is common in everything from medical autopsies to NTSB reports. This report in particular is about the car, and what the car was doing. So naturally, they'll describe all events as they relate to the car.
Re: (Score:3)
Re: (Score:3)
They 'made contact with' sounds like it was just a harmless little nudge. They then go on to much more emotional language with how the car 'sustained damage'. It's obvious where their interests lie, and it's not with the dog that was killed.
Re: (Score:2)
Re: (Score:3)
Re: Contact Was Made (Score:2)
purse dog
Rat dog, please. Let's try to use the correct terminology lest some AI reads this and is trained incorrectly.
Re: (Score:2)
A rat dog is a dog that is bred to hunt rats. A purse dog is a dog that is bred to sit in purses. They aren't the same thing, and we don't know which this was. (I'd guess it was a rat dog, as they run around more, but it's a guess.)
OTOH, some rat dogs are rather large. One example is the Doberman Pinscher. According to what I heard they were first raised by a garbageman around New York to keep the rats in his dump under control.
Re:Contact Was Made (Score:5, Interesting)
Sigh.
So you slow down to below 25 MPH when there's cars parked along the road? I call bullshit.
Because that's the the situation here, 25 MPH along an industrial street.
A human would also have hit that dog. This is a complete non-story.
Re: (Score:2)
Along a 25 mph road? Of course I do, because like I said, it's the law. [ca.gov]
Remember, All speed limits are based on ideal driving conditions. [driversed.com]
Autonomous vehicles are supposed to be safer than human drivers.
Re: (Score:2)
A human would also have hit that dog.
Autonomous vehicles are supposed to be safer than human drivers.
Safer but not magical, they still have to conform to the laws of physics, which dictate that all cars have a finite stopping distance. It is sloppy magical thinking to make the implied assumption that safer = infinitely safe. There are two parties involved in this collision, and if the circumstances make the collisions unavoidable to the driver, then the other party can still choose to avoid it. Aka if you love your dog, leash your dog and keep it short near roads. Dogs don't know better but you do.
Re: (Score:2)
Also you have to add in reaction time. It is for these reasons that you are legally required to slow down when your vision is obstructed. [driversed.com]
Re: (Score:2)
It's not altogether clear that this is a complete non-story. We don't know how far away the dog was when it was first detected running onto/towards/along the street. There's too much uncertainty to make a real determination.
My tendency is to blame Waymo until they release better evidence, but to acknowledge that there might be no real problem.
Re: (Score:2)
Sigh.
So you slow down to below 25 MPH when there's cars parked along the road? I call bullshit.
Yes, maybe? Depends on the exact road but I adjust the speed to what is reasonably safe and not the posted speed limit. If it's tight or it looks like there are people around for example.
Technically here the limit is 50km/h but you'd have to be nuts to go that fast here for instance: https://i.imgur.com/vCnGGyU.pn... [imgur.com]
Re: (Score:2)
Are you going to slow to a complete stop? Because there's no viable speed at which it's possible to avoid hitting a small dog darting out from between parked cars right in front of you. No machine could do that either.
Re: (Score:2)
>I thought the computer was going to be programmed to be better at following the law than a human,
They are, and do. They are programmed to never exceed the posted speed limit, and also to slow down from that speed when conditions dictate. Can't say the same for most humans.
>and therefore actually be safer.
Again, they are.
"When compared to national crash rate estimates that control for unreported crashes (4.2 per million miles), the crash rates for the Self-Driving Car operating in autonomous mode when
Deer, foxes, cats, dogs, weasels (Score:2)
slow news day (Score:5, Funny)
In other news: a dragonfly was killed today by an autonomous car as it flew into the radiator grill while the car was rounding the corner of 22nd and Lincoln St. The name of the dragonfly has been withheld until its next of kin can be notified, although there is some suspicion its next of kin is implanted in the same grill. Experts say this kind of death will be commonplace with the new autonomous vehicles. Opponents of autonomous vehicles are planning to boycott the company in a joint press release with PETA.
Re: (Score:3)
Netcraft confirms it, DragonFly BSD is dead!
Re: (Score:2)
Was the dog owner cited for letting their dog run free?
More info needed (Score:2, Troll)
I would like to know a lot more detail about this collision. Autonomous vehicles are being pushed as 'better than stupid human drivers' because of their better all-round vision, not being tired or distracted, better reflexes etc.
Waymo reckon their robocar recognised the dog as such, but was unable to stop in time on a road with a 25mph speed limit. The braking distance at that speed with a modern vehicle should be in the region of one vehicle length.
So, unless the dog was unseen until within one car length
Re:More info needed (Score:5, Insightful)
Swerving away from an animal that runs into the road is the most common mistake that causes major accidents in this kind of incident. The best response is slam in the brakes if you're sure there isn't another car too close behind, otherwise chance takes its course. A little animal can change direction much faster than a car anyways, swerving won't even change the probability of hitting it appreciably.
Re: More info needed (Score:2)
The autonomous vehicle would, of course, be monitoring all the traffic around it. 'Swerving away from the dog's path would put the vehicle in danger of hitting this specific other vehicle' would be covered by the "why not" part of my question.
Re: (Score:2)
What if i am not sure whether someone is following closer than they should?
I am not sure I am following your reasoning here.
Re: (Score:3)
What if i am not sure whether someone is following closer than they should?
Then you are a shitty driver. You should know that already, there is a mirror hanging from your windscreen that will tell you. Or if you are in a vehicle with an obscured rear view like a cargo van or a lamborghini countach, you can use a side mirror. Anyone who isn't sufficiently aware to know whether someone is up their ass at all times has no. business. whatsoever. driving.
Also, just lean back and rest your head on the whiplash arrestor and hit the fucking brakes anyway. If you do it on purpose just to c
Re: (Score:2)
Yeah... your condescending tone is so grating that even if you were right, you'd still not get your point across.
Not only you were wrong though, you were completely off topic.
The question was rhetorical. There is no thinking involved when something jumps out in front of the car. I hit the brakes reflexively before I recognize what that thing is, much less before I think about whether there is someone behind me. Even as a passenger my foot has reflexively hit brakes that are not there.
Re: (Score:2)
Yeah... your condescending tone is so grating that even if you were right, you'd still not get your point across.
Found the shitty driver in denial. Feelings hurt? Good. In the future, pay attention to your surroundings. Do more of that, and there will be less stomping on the brakes.
Re:More info needed (Score:5, Informative)
Then you are a shitty driver
Actually no. A good driver doesn't take into consideration whether someone is behind them when attempting to avoid a hazard. Slam on the breaks. If you get rear ended you get rear ended, driver behind is 100% at fault. If you don't slam on your brakes and hit something, you're at fault. If you stop and think about how close someone is behind you when making a decision, you're at fault.
All these people saying you need a condition or to think about correct action have never taken a safe driving course. The recommendation is always brake. Don't consider. Don't think. Don't question. Just brake. Whether someone is up your arse or not is not a consideration a good driver makes.
Re: (Score:2)
What if i am not sure whether someone is following closer than they should?
Then you are a shitty driver
Actually no. A good driver doesn't take into consideration whether someone is behind them when attempting to avoid a hazard.
A good driver already knows whether someone is close behind them whether they are taking it into account or not. If you don't think so, guess what?
Re: (Score:2)
go take safe driving lessons from a reputable school
they can demonstrate what and why you are doing wrong
Re: (Score:2)
You think a driving school instructor is going to be upset at me for using my mirrors and having situational awareness while driving? You are an oblivious menace who should not even be allowed to ride in a car, much less drive one.
Re: (Score:2)
Re: (Score:2)
A good driver doesn't take into consideration whether someone is behind them when attempting to avoid a hazard. Slam on the breaks. If you get rear ended you get rear ended, driver behind is 100% at fault.
You're not wrong, but you also have to remember that it doesn't matter who is at fault if you die. I mean, it might matter to your estate, but you're not going to get a warm, fuzzy feeling as you bleed out and your last thoughts are about how everything is ok because the other guy is at fault. So, yes, if you slam on the brakes and someone hits you from behind, they're the bad driver, not you, but there are lots of bad drivers out there, so you have to watch out.
Re: More info needed (Score:2)
My line of reasoning is that an autonomous vehicle should already know the distance and closing speed of any following or oncoming vehicles in its vicinity and be able to incorporate this information in its dangerous-collision-avoidance strategy. A human might not emergency brake for a dog for fear of being rear-ended because the human driver does not have time to check behind.
Maybe in this particular case it was not possible to avoid hitting the dog, but the telemetry and logs from the vehicle should be b
Re: (Score:2)
it was the dog owner's fault.
of course, the law is on the side of the pet, what with pedestrians having almost absolute right-of-way, the insurance will satisfy the dog owner's demands as far as legal stuff goes.
There's not much one can do in situations like this.
Re: (Score:2)
The best response is slam in the brakes if you're sure there isn't another car too close behind,
If you ever took drivers ed, one of the first things you should have learned is you never let pressure from behind you dictate your driving decisions.
Re: (Score:2)
A dog was moving at âoea high rate of speed" WRT a car? Somebody is lying.
There are lots of plausible scenarios where the Waymo car is not at fault, but that' statement is not just wrong, but silly.
Re: (Score:2)
It doesn't require lying at all. Just somebody with different standards for "high rate of speed". A race car driver will have different standards for "high rate of speed" than, say, your average grandmother out on the road. Or, for the top gear episode I'm watching right now, peugeot drivers. ;)
A quick googling suggests that the top speed for a "small dog" is ~30mph. In a 25mph zone, appearing from a blind spot, that would indeed be a high rate of speed. Would also potentially signal a dumber than avera
Re: More info needed (Score:2)
Interesting.
Are _all_ side impacts ignored?
Would the robocar be as nonchalent on a level crossing?
Is it just side impact from a small object which is ignored?
Is it the suddenness which is beyond the robocar's ability to identify, track and predict?
How far away does a threat have to be for the robocar to try and react to it?
What does that mean for a robocar in heavy multilane traffic?
I look forward to the developer's detailed and data-supported replies.
"the system correctly identified the dog" (Score:2)
Fuckin' nerds.
And why wasn't the dog on a leash? (Score:5, Informative)
While many owners teach their dogs to stay close when off leash, not all dogs are predictable, especially when around other dogs. In order to prevent potential issues San Francisco introduced Sec. 41.12 of the city’s health code requiring that whenever off your own personal property, a dog must be on a leash no longer than 8 feet in length.
FSD finally moving in the right direction (Score:2)
hopefully we can soon be rid of this canine menace
Root cause (Score:2)
Obviously needs more LIDARs. Moar.
Snark aside I once killed a cat. Stray probably feral. Came out of a low hedge less than four feet from the roadway, completely undetectable until a fraction of a second before it hit the front bumper. Maybe an IR sensor could have seen it but its motion was unpredictable. It isn't clear to me what any autonomous system could have done.
Horrible experience. In the rear view mirror I saw it thrashing around with a broken back. I stopped immediately and went back
Re: (Score:2)
Hell, I killed a bird a few months ago. I'm driving down the street, and it suddenly dived and swooped right into the side of my car. DRT.
Years and years ago a deer ran into the side of my car - it's head basically hit my driver's side mirror, snapping it's neck. Any slower, I'd have probably plowed into it. Faster, it would have maybe hit the side of my car instead, or maybe I'd have passed before it went through.
But I basically had no time from seeing the deer jump from the woods to the impact.
Black box recording (Score:2)
Just love the careful wording. (Score:2)
Note how the dog was an active participant. The dog ran out. The dog died.
Apparently, all the car did was "make contact"
Dog, Child Next (Score:2)
Re: (Score:2)
Re: Liability (Score:2)
Re: Liability (Score:2)
Re: (Score:2)
Re: (Score:2)
I'd be surprised if waymo or any of the current systems can:
1) recognize a small child over the hood of a parked car bouncing her ball
2) understand the odds of child chasing ball into the street are high enough to take precautions
3) take those precautions and already be going slow enough to jam the break when she disappeared from view, because that's when I hammered the breaks. I saw the parent's expression of horror but couldn't see the little girl anymore. I knew she had to be in between the cars, hidde
Re: (Score:2)
I'd be more surprised if these systems *didn't* have "bouncing ball" procedures and algorithms. I imagine it's an easier solution than many other things that have had to be added to the systems for safety. If not deduced during the myriad meetings and analyses of hundreds of people during design and test, occasionally programmers breed and have personal experience of what kids can do.
I congratulate you on your observational skills and intuition. That's not a snark, it's appreciation of a skillful driver. No
Re: (Score:2)
it isn't true, and Teslas are worse than the average human driver.