> telemetry will tell the story
Idk, the death-toll gap between Tesla and Waymo seems to tell a story enough.
Idk, the death-toll gap between Tesla and Waymo seems to tell a story enough.
Note: it’s important to distinguish between Autopilot and FSD here. “Autopilot” is Tesla’s old assisted driving stack that comes free in most vehicles and has had no significant updates in years. “FSD” is an entirely different software stack that only works with newer vehicles and that Tesla charges $$ for. It’s much more advanced and IMO a lot safer.
This article never mentions FSD, only Autopilot.
It’s tough to say given many data sources are aggregated. For what it’s worth, my parents’ car is a Tesla with FSD and I’ve stopped it from, off the top of my head, racing into a red-lit intersection, running over a small dog and running into a closing garage door.
I still use it. It mostly works. But I’m vigilantly monitoring it in a way that isn’t supported by Tesla’s marketing (which frequently shows drivers engaging it hands off).
I’m surprised about the red lights and animals, however: ours seems very cautious around any kind of live animal on the road, even braking and manoeuvring to avoid birds on the road. It’s not so good at avoiding the corpses of already dead ones, however (bump!).
Why not? That’s likely to come up at minimum thousands of times per day, and likely vastly more as the system improves.
They show a 2D representation to the driver because that’s good enough for drivers, but I wouldn’t assume that represents how the system operates internally at every stage. Even navigating requires the concept of bridges crossing roads without intersecting them.
https://news.ycombinator.com/item?id=47783427
AFAIK, there's no issue with bridges that cross roads, however. It will just ignore a road that's above or below the one you're travelling on that doesn't intersect with yours. Just like a human driver would do.
At scale you can “have issues” and still work 99% of the time. If it’s completely incapable of handling railroad crossings that’s relatively easy to verify.
> AFAIK, there's no issue with bridges that cross roads, however. It will just ignore a road that's above or below the one you're travelling on that doesn't intersect with yours. Just like a human driver would do.
As a user of the system that’s worked it all out it’s not an issue for you, but getting that behavior from raw sensor data is non trivial.
Both only happened once. But they were shocking when they did. (To be fair, my Subaru tried to push me into oncoming traffic because it was avoiding a “collision” from the guy in the other lane turning weirdly. Turned at collision-avoidance feature off.)
I'm not sure the point you're making here - that just sounds like "Tesla couldn't care less about updating their software and it's still "not good". People it hits are still dead. "Oh well, it's not like Tesla had updated the software, so you can't blame them".
That way everyone would get the safety advantages of FSD, including the really important stuff like better driver attention monitoring. Unfortunately, Tesla keeps Autopilot bad because it forces more people to pay for FSD, and in this case they seem to care more about profits than safety.
Not saying you're wrong, but I think "it forces people to pay for FSD" is a poor decision if true.
More likely it creates market confusion (as we see in this thread) about exactly which of the two products that Tesla advertises as things which drive your car for you is the more dangerous one and which is "the good one" (though from their taxis, I think the scare quotes are still justified even then).
I'm also not even sure if they do care about profits (or safety), given Musk's made about as much from selling Tesla shares as the company has made in lifetime profit, before getting shareholder approval for a deal to give him more shares. (Conditionally give, yes, but the nature of it all suggests share price is more important than profit or safety, and if you don't care about such things that makes it much easier for them to sell a million robots or whatever).
I ask because he has made about as much from the latter as Tesla has made from the former.
https://theintercept.com/2023/01/10/tesla-crash-footage-auto...
No deaths in this one
Note the comment about Google's Waymo ceasing to use the term "full self-driving" after this accident
Why would they do that
Let the reader decide
See: The software that enabled a secondary AOA sensor on the Boeing 737Max.
It's relevant because it arguably affects the driver's beliefs and behaviour, whether they are using FSD or just Autopilot
It's a separate upgrade but it's reasonable to assume the marketing may affect drivers using only Autopilot
If HN commenters are confusing the two then this only strengthens that argument