Live data from Hacker News

Mercedes beats Tesla to autonomous driving in California

theregister.com

341–350 of 778 posts

Re: Mercedes beats Tesla to autonomous driving in California

#341
post #262

Earlier quoted context omitted.

> Tesla FSD navigating the complex city streets of LA for 60 minutes with zero human intervention. 17 fatalities, 736 crashes: The shocking toll of Tesla’s Autopilot https://www.washingtonpost.com/technology/2023/06/10/tesla-a...

Without any comparison to humans those numbers are completely meaningless

> Without any comparison to humans those numbers are completely meaningless

I disagree.

What I want to know, is is it safe? Not is it safer than the average human driver, but is it safe in an absolute sense.

When a human driver crashes we can almost always pin the cause down to human error. Errors caused by some human failing, being less than they could be. The promise of machines is that they are consistent. They are never "less than they could be", but are consistently at their best.

Comparing humans and machines is comparing dogs and roses - it is interesting a rose smells better, interesting a dog is more loving (or fierce), but not a valid comparison

Self driving cars stand or fall on their own capabilities.

Re: Mercedes beats Tesla to autonomous driving in California

#342

Earlier quoted context omitted.

Working in the AV space, it's really frustrating how confidently people who have no idea about what's hard and what isn't go off about Tesla right now. Mercedes has soundly beaten the last decade of Tesla efforts by reaching L3. I've personally watched FSD go off the rails and into a crash situation within 60 seconds of being turned on three times this month (I have a friend who loves to try it in San Francisco) Had…

> People also don't seem to understand Mercedes has a separate L2 system that works above 45 mph that already scores better than AP by consumer reports Isn't Mercedes' L2 system non-upgradable, though? i.e. people with Merc's L2 system today won't be able to upgrade to L3+ later on, they'll need to buy a whole new car - right? I bring this up only because I bought my X in 2018 with FSD prepaid - it's 5 years old now,…

Drive Pilot uses their MB.OS which is updatable so assume it also will be.

Unless you have some source that suggests otherwise.

And it uses LiDAR which in my opinion is mandatory for FSD.

Re: Mercedes beats Tesla to autonomous driving in California

#343

Earlier quoted context omitted.

Now why would the EU mandate details that make for a less safe driving? You are implying they are either malicious or stunningly incompetent.

> stunningly incompetent Not stunningly, just moderately. But generally yes, that's a good description for EU bureaucrats.

>> stunningly incompetent > Not stunningly, just moderately. But generally yes, that's a good description for EU bureaucrats.

I disagree.

Wordy, worthy, self interested, rule fetish... All of that. But not incompetent like this.

Teslas being optimised for L.A. makes more sense in this case.

Unusually it looks to me like conspiracy rather than cock up.

Re: Mercedes beats Tesla to autonomous driving in California

#345

Earlier quoted context omitted.

Mercedes accepts legal liability for collisions if the system is used under these stringent conditions. Tesla doesn't do this. It also doesn't mean it doesn't have the technical capability to drive through the streets of LA for 60 minutes. https://www.thatcham.org/mercedes-to-accept-liability-for-ac...

Working in the AV space, it's really frustrating how confidently people who have no idea about what's hard and what isn't go off about Tesla right now. Mercedes has soundly beaten the last decade of Tesla efforts by reaching L3. I've personally watched FSD go off the rails and into a crash situation within 60 seconds of being turned on three times this month (I have a friend who loves to try it in San Francisco) Had…

This FUD is insane.

You make it sounds like Tesla's with FSD on are crashing all the time. This makes no sense at all, because... they aren't?

Re: Mercedes beats Tesla to autonomous driving in California

#346
post #115

We’ve actually had nearly self driving vehicles for over a century. They’re called trains. We should invest in some. The navigation problem gets way easier

Driverless trains are actually a pretty new thing, being on tracks is not as set/forget as people assume. If you mean "transportation where I'm not driving" then sure but taxis also fit that bill.

Paris had it's first fully automated subway line in 1998.

Today there are 2 fully automated ones and one other mixing manual and automated.

The sofware was developped by Matra transport international (now Siemens Mobility).

It can even dynamically add/remove trains in service to best match the demand !

It also apparently uses a 48 bit cpu architecture.

[0] https://fr.m.wikipedia.org/wiki/Syst%C3%A8me_d%27automatisat...

Re: Mercedes beats Tesla to autonomous driving in California

#347
post #331

Earlier quoted context omitted.

Anecdotally, in LA, I turn off FSD basically every minute while trying to use it, due to it doing something slightly “inhuman” or not ideal/too to the letter of the law, signaling incorrectly while staying in a lane that curves slightly, etc. I can’t imagine letting it go for a full hour without causing some road rage or traffic blockage. To be clear there is a definite driving “culture” in LA that is very permissive…

Every time you disengage it invites you to leave immediate voice feedback as to why, and presumably they are using all this feedback in conjunction with camera and data feeds from cars that are opted in (which includes all FSD beta cars I believe). So, they are getting what they need to make it better.

The problem with Tesla is the lack of LiDAR not training data.

You need to be able to accurately do bounding box detection in order to determine whether that billboard of a person is real or not or if that dog with a hat should be avoided.

Research has conclusively shown that vision only systems simply can't match LiDAR for this task.

Re: Mercedes beats Tesla to autonomous driving in California

#348
post #71

Earlier quoted context omitted.

https://www.teslaoracle.com/2023/06/07/tesla-model-3-highlan...

That’s not shipping… If you buy a model 3 TODAY you get V3

Not on the 3. But they are shipping it with the Model Y.

https://driveteslacanada.ca/news/tesla-now-shipping-model-y-...

Re: Mercedes beats Tesla to autonomous driving in California

#349

Earlier quoted context omitted.

Anecdotally, in LA, I turn off FSD basically every minute while trying to use it, due to it doing something slightly “inhuman” or not ideal/too to the letter of the law, signaling incorrectly while staying in a lane that curves slightly, etc. I can’t imagine letting it go for a full hour without causing some road rage or traffic blockage. To be clear there is a definite driving “culture” in LA that is very permissive…

> signaling incorrectly while staying in a lane that curves slightly This is super annoying, but the most recent FSD update seems to have mostly fixed that. It's still pretty crappy in a lot of other ways, but at least that improved.

> mostly fixed that

The problem is that issues may be fixed one week but re-appear later.

Self driving is not an exercise in fixing bugs despite what many people think.

Re: Mercedes beats Tesla to autonomous driving in California

#350

Earlier quoted context omitted.

These statistics are meaningless. How many Tesla cars with full self driving are there compared to regular cars? A Tesla with FSD will only be using FSD a small fraction of the time. If we had statistics on the number of hours FSD was active compared to the number of hours all other car driving of all cars we might be able to compared these numbers. Hour wise, I think normal driving is way above 7500:1.

Why are people willing to excuse things because they happen in a car? If an AI power tool were going haywire once in a while and chopping the arms off of bystanders near by, would we find that acceptable because people using non-AI powered tools also chop off their body parts sometimes? 'We can't know if it would have happened if a person were in control' is not an argument that would fly for anything else just as da…

> would we find that acceptable because people using non-AI powered tools also chop off their body parts sometimes?

Yes, if the AI power tools chop off less limbs than the human controlled power tools, I would find that acceptable.

Post reply on HN