|
They won't.
But self-driving cars only make sense in a closed system where all of the cars talk to each other and nothing squishable gets on the roadway.
As long as they don't interoperate with each other -- between makes as well -- and as long as there are non-affiliated (human-driven) cars on the road and people and animals can cross the roadway -- the system just can't work.
The current research is fine for ironing out the bugs in preparation for making an eventual future city with a closed road system.
I need to watch Logan's Run again.
|
|
|
|
|
A closed system, like a warehouse or construction yard, is not comparable to a city.
Consider what happens if there is an accident in a warehouse.
- Immediate stoppage of most everything.
- Immediate response to the injury
- Any claims of monetary damage, at least in the US, usually are limited to actual provable damages and might even be covered solely by workers compensation.
Additionally, especially in a warehouse, the unexpected is very low. No cows or bears. No high speed chases (cars or foot). No broken water mains.
In a city none of that is true.
|
|
|
|
|
jschell wrote: In a city none of that is true.
Current cities, no. But in an idealized city of the future... maybe. Maybe not.
|
|
|
|
|
Greetings and Kind Regards
I assume Science / Technology march on ever forward. Sooner or later self-driving cars will be more or less perfect. I am rather surprised their legality occurred so quickly. I have always wondered why that was so as I assume a self-driving car would not know what to do in response to this not unusual situation Dancing policeman: America's most entertaining traffic cop - YouTube[^] .
Then of course is the matter of software attacks which I find frightening.
|
|
|
|
|
BernardIE5317 wrote: I assume Science / Technology march on ever forward. Sooner or later self-driving cars will be more or less perfect
That is a fantasy.
The real world is limited not only by physics but by other things as well such as economics and popular perception.
So for example faster than light travel is never going to happen because it is just not possible. Hypotheticals that attempt to circumvent that are even more fantasy and even more so driven by those other factors.
It is not possible to recycle any with a 100% efficiency. It is not possible to create any process that even close to being 100%.
It is not possible to have no one that is poor. Physically not everyone can have a plane. There just is not enough airspace not to mention how to pilot it. And if you avoid the physical limitations then people would still find a way to differentiate themselves. So for example those who can create poetry would be rich and those that can't would be poor.
Nuclear power can not only provide significant power but also significantly reduce pollution. But that requires that you be able to convince the population to let you build them. Especially in the numbers needed.
|
|
|
|
|
Greetings and Kind Regards
I am not certain the relevance of your observations re/ self-driving cars but please permit a few comments of my own.
re/ faster than light travel : In 1000y we will learn how to bend space and time to our will. Just as the visiting Space Aliens are doing. If they can so can we.
re/ recycle : No doubt you are correct if for no other reason the general public is unconcerned and uncooperative. Exempli gratia I learned only today coal can be converted to animal feed. Amazing. "Science Marches On"
re/ poverty : This is not obvious to myself. I imagine a time in future in which robotic economy transforms the world to a lazy man's paradise where all needs are met.
re/ airplanes : No doubt you are correct. I for one do not wish to own one.
re/ poetry : No doubt you are correct. Exempli gratia "Roses are Red Violets are Blue I do not Know How to Fly Please Where is the AirCrew?" I have proven your kind self correct.
re/ Nuclear Power : I am rather optimistic in particular re/ so called "micro-reactors". Otherwise you are of course correct. Things always depend on something. Somehow the Pyramids got built.
A few parting thoughts from Arthur C. Clarke :
"The one fact about the future of which we can be certain is that it will be utterly fantastic."
“The only way to discover the limits of the possible is to go beyond them into the impossible.”
“Any sufficiently advanced technology is indistinguishable from magic.”
|
|
|
|
|
BernardIE5317 wrote: In 1000y we will learn how to bend space and time to our will. Just as the visiting Space Aliens are doing. If they can so can we
There are no aliens here because they can't exceed the limit either.
If they could then they would have populated the planet long before we existed.
BernardIE5317 wrote: coal can be converted to animal feed. Amazing. "Science Marches On"
The sources for those claims are suspect.
BernardIE5317 wrote: I imagine a time in future in which robotic economy transforms the world to a lazy man's paradise where all needs are met.
You can also imagine a world where fairies are enslaved and they use magic to produce everything.
But I already addressed that.
First there are some commodities which cannot physically be allowed for all that want it. Like private planes. Not enough airspace. Not enough runways.
Second in such a society as I said humans will seek to differentiate themselves in other ways. And thus, as an example, creative talents such as poetry might be used to meet that need. And some will not have the talent. So they will be poor.
BernardIE5317 wrote: I am rather optimistic in particular re/ so called "micro-reactors".
They are in fact micro. And they still must be built somewhere.
BernardIE5317 wrote: A few parting thoughts from Arthur C. Clarke :
Few things about quotes is that they are nothing but quotes. They don't change reality.
Technology does not increase forever because reality does not increase forever. As an example the increasing speed of computers which once was described with a quote has now reached a very real physical limit.
|
|
|
|
|
re/ speed limit : Traveling through bent space / time does not exceed c.
re/ aliens : Have you met with them? Is that how you know so much about them?
re/ magical fairies : In 100y robots will be so. They will only require natural resources for manufacture.
re/ micro reactors : Everything needs to be built somewhere. Just like the pyramids and coal burning power plants.
re/ computer speed : I recall reading something about Quantum Computers recently also Photonic computers. "Science Marches On"
Where you Charles H. Duell in a previous life?
btw Why all the blank spaces? Never mind. It seems to be related to quotes something I never do as I consider it rude. Though I do not see why they are inevitable. Here is an experiment.Quote: Here is a quote.
No needless space.
modified 10-Jan-24 2:02am.
|
|
|
|
|
Even if what you're suggesting was possible, mankind will see to it to put an end to himself long before any of that has a chance of ever coming true.
[/story].
|
|
|
|
|
BernardIE5317 wrote: Traveling through bent space / time does not exceed c.
Suggestions about that is completely hypothetical. Not to the extent that it is just waiting to be developed but rather that if something was built then it might allow that.
Suggestions about that often (always?) involve materials that either cannot be created with engineering or are so expensive that the amount needed could never be created.
BernardIE5317 wrote: Have you met with them? Is that how you know so much about them?
Invalid statement. It does not prove anything. Nor lead to proof.
BernardIE5317 wrote: In 100y robots will be so
Actual history suggests otherwise.
Development of robots has been ongoing for at least 70 years.
Compare that to the 70 years after the introduction of the internal combustion engine.
Compare that to the 40 years between the introduction of satellite phones and now.
Technology in not built on 'break throughs'. It is built on incremental improvements on existing technology.
BernardIE5317 wrote: I recall reading something about Quantum Computers recently also Photonic computers
Quantum computers were introduced in the 1980s.
Quantum computers are NOT a replacement for current computers. The problems they solve are different.
IBM, just last year, announced (hoped) that they will deliver a quantum chip in 2033. So 10 years from now.
So not even close to the speed rate that was anticipated both for current computers and even quantum computers.
|
|
|
|
|
Science Marches On. Sometimes w/ breakthroughs sometimes w/ increments. Huge corporations are betting Billions you may perhaps be wrong. As for myself I adhere to he maxim "If it can be imagined Science sooner or later will achieve it."
- Cheerios
|
|
|
|
|
You are both right and wrong -> The timescale plays an important role here.
There will be a very long phase where the cars will be at AD level 2.5 to 3, e.g. only on some specific roads and the driver must be able to take over within seconds. AD will be a standard on highways in the coming 10 years, but that's it. The step to level 4 needs an established level 3, where almost all vehicles are connected AD vehicles. Then, level 4 can be rolled out, and only after we will jump to level 5.
AD in all situation is extremely complex and requires lots of SW (there are already about 100 millions of LOC in SW of an average recent vehicle, this is 10 times what is required to fly a plane, and this is WITHOUT AD).
During this time, people will accept that AD will not solve all crashes, and that they are using a machine that can fail. You sign term and conditions when you drive with the AD function, it is your decision and therefore will remain your responsibility, and it will clearly be put in disclaimers. If AD dies, it would only be because no driver would want to endorse responsibility of the system, but not because people will claim against OEM - at least, not more than today.
|
|
|
|
|
Rage wrote: and the driver must be able to take over within second
That is not 'self driving' then.
Someone who is typing on a computer, while balancing a super sized soda in the crook of their arm, is not going to be doing anything "within seconds".
Look to one of the other posts that suggests the driver can sleep in the car.
What you are referring to is enhanced safety controls on the car.
Rage wrote: You sign term and conditions when you drive with the AD function, it is your decision and therefore will remain your responsibility,
That is not how it works in the US.
|
|
|
|
|
EV's are a given. The fact they're all electronic and "connected" means that over time, the driver will be dumbed down to the point we do have "self driving", and you can sleep in your car (the actual goal).
At some point, you won't be able to go above a certain speed. No crossing double lines. The trafic system will slow you down before the light changes. If you go "off the grid", without a permit, your car won't. Etc.
We're talking years; but it will come.
"Before entering on an understanding, I have meditated for a long time, and have foreseen what might happen. It is not genius which reveals to me suddenly, secretly, what I have to say or to do in a circumstance unexpected by other people; it is reflection, it is meditation." - Napoleon I
|
|
|
|
|
Speed limiters have been a legal requirement in all new cars sold in Europe since July 2022 - and it's only a matter of time before it is illegal to disable them (they default to on when the car is started, but can be disabled by the driver). I don't doubt that they will become a part of annual compulsory vehicle testing (MOT in the UK) once the first such vehicles become old enough to need testing.
"I have no idea what I did, but I'm taking full credit for it." - ThisOldTony
"Common sense is so rare these days, it should be classified as a super power" - Random T-shirt
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
OriginalGriff wrote: Speed limiters have been a legal requirement in all new cars sold in Europe since July 2022
No it isn't.
In July 2022 they signed the 2019/2144 regulation that goes into effect July this year.
And there is no limiter (well not yet anyway). There will be a requirement for all new cars to warn you if you're speeding though.
And the warning/notification should be designed to be easily dismissed or ignored. According to the law, the audible and haptic warnings must be "as short as possible in duration to avoid potential annoyance of the driver."
I read some preparatory work for the regulation a few years ago and a limiter was considered to be out of the question.
Reason being: Assume you're overtaking a slower vehicle and that vehicle suddenly accelerates to the speed limit while you're overtaking, you would end up driving side by side with you being in the wrong lane.
In short, speed limiters would be quite dangerous used wrongly, which they would be.
Well, let's see how long they have that opinion, speed limiters are already in effect for lorries since many years...
|
|
|
|
|
|
Anyone who has read a shrink wrapped license should have an inkling of how this is going to go down.
My prediction:
Car is provided with autonomous driving capability, but it's not enabled by default.
User reads the license (!) and accepts it. License of course places all of the risk back on the owner who agrees.
By accepting this license <blah blah=""> Neither car yard nor the device manufacturer or installer, gives any other express warranties, guarantees, or conditions regarding autonomous driving. <manufacturer> and installer exclude all implied warranties and conditions, including those of merchantability, fitness for a particular purpose. If your local law does not allow the exclusion of implied warranties, then any implied warranties, guarantees, or conditions last only during the term of the limited warranty and are limited as much as your local law allows. <blah blah="">
Final point - car manufacturer identifies the driver/occupant by breaching your privacy and using sensors in the car (weight, seat adjustment etc) or a profile to identify the person who is "in control" of the vehicle.
PS: Elon Musk is not silly. His autonomous driving already takes this approach. Driver must be ready to take control etc. That it's totally impractical is irrelevant. It's all about blame.
|
|
|
|
|
pmauriks wrote: His autonomous driving already takes this approach. Driver must be ready to take control etc.
There is a recall already in effect for Tesla which further restricts the very limited self driving that the car does have. Because that feature has been implicated in several accidents.
The reported fix not only intends to insure that the driver is actively involved (which by definition really means it is not self driving) but failure to be actively involved over time can lead to the car locking that feature out for the driver.
|
|
|
|
|
I think it's targeted wrong in terms of spending too much effort on local city driving. The infrastructure of most cities and towns isn't conducive to AVs. Using a controlled vehicle in an uncontrolled environment is difficult and requires myriad sensors and strong AI analysis and controls. Cruising down a highway where all the vehicles are operating in concert in a controlled space, sure why not?
|
|
|
|
|
Quote: So in the above with a self driving care no person can be at fault. Because they were not driving.
I think this is maybe true, but still a big assumption and all of that just hasn't been sorted out yet (liabilities).
There were autonomous "R/C" aircraft 25 years ago. Yeah really.
The "R/C" is and is not a misnomer. Some of them, maybe even most, you could "take over" just like a Tesla/autopilot sort of works.
Not so strange we'd nail it down in the air first. There's less pedestrians to hit up there.
But liabilities? They always fell on the flier/owner. It didn't matter they didn't write or create the control software or if they'd built their whole setup from a single off-the-shelf kit or some kind of "kit bashing" or what.
I hear where you seem to be going is that there's a sort of DDoS attack using liability lawsuits that suddenly inundates the creators with financial obligations.
Just because you can sue, and probably would, if you lost someone close to you in an accident maybe? That definitely doesn't mean anyone is going to be found at fault and owing anything at all. "Stuff" does happen and people grok that.
Also though, there's a flip side to this whole coin and that's how many of these such accidents do not even happen at all anymore because automated systems start kicking humans' butts at "paying attention" and "not being screwed up behind the wheel" or "trying to text". (They may be already... devil in the details sort of question)
In short, self-driving is not going to die at all.
We're already at a point with tech and affordability/accessibility that anyone can DIY this stuff "easily". Where "easily" is talking about the barrier to entry, especially if we concede that doing it on the scale of a little R/C car is really almost as relevant sensor/software/tech wise and bits of doing such a thing can near directly translate over to "real cars".
So what I see is not really FOSS, per se, just the fact that tons of hobbyists can and will make contributions here in the form of "crowd sourced" ingenuity that finds its way back in the hands of "the big guys" (corps like Tesla).
Now talking Tesla specifically in all this context? Maybe it does effectively "get killed" for them because Elon is just too reckless about many many things. But for the world/country? Nah.
modified 12-Jan-24 15:35pm.
|
|
|
|
|
|
I'd argue that this is not about 'crypto'. this is really no different from other fraudulent companies, hedge funds or startups. There is more fraudulent 'low hanging fruit' because there are more uneducated people putting their money in. But honestly the amount of money being scammed in crypto is dwarfed by the amount of fraud committed by legit companies. If you watch the big short, you get an explanation of how the 2008 banking crisis was caused by those respectable banks and investment companies. A couple of Billion in crypto scams doesn't even register on that radar.
|
|
|
|
|
I've been burning the candle at both ends fighting PyTorch, everything NVIDIA, Python, and my own short attention span (squirrel!). In updating the dashboard for CodeProject.AI I've been refactoring a bunch of truly ugly code and as these things go, you pull on a thread and you just...keep...pulling. After two consecutive nights of dreaming about the code I realised I probably should unplug for a day. Right after I fix just one more thing...
So am I a little odd or does this happen to you guys occasionally? Or am I just lucky to be working on something that grabs me like this?
cheers
Chris Maunder
|
|
|
|
|
I started at 5 this morning. I'll log off in a couple of hours (it's currently 4PM in the UK). So yes, I get the obsession.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
