US Senate Panel Approves Self-Driving Car Legislation (reuters.com) 123
An anonymous reader quotes a report from Reuters: The U.S. Senate Commerce Committee on Wednesday unanimously approved a bill to speed self-driving cars to market without human controls and bar states from imposing regulatory road blocks. The bill still must be approved by the full Senate. The U.S. House passed a similar version last month unanimously. General Motors Co, Alphabet Inc, Ford Motor Co and others have lobbied for the landmark legislation. Despite some complaints from Republicans, the Senate bill does not speed approval of self-driving technology for large commercial trucks after labor unions raised safety and employment concerns. The measure, the first significant federal legislation aimed at speeding self-driving cars to market, would allow automakers to win exemptions from current safety rules that prohibit vehicles without human controls. States could still set rules on registration, licensing, liability, insurance and safety inspections, but not performance standards.
All I need to Know... (Score:3)
General Motors Co, Alphabet Inc, Ford Motor Co and others have lobbied for the landmark legislation.
Enough said to know where I should stand on this.
Re:All I need to Know... (Score:5, Insightful)
Accidents happen all the time. The only thing needed for this to be a good idea is less average damage to people in accidents with self-driving cars than in human-controlled ones. That should not be hard to achieve...
Re:All I need to Know... (Score:5, Insightful)
People like you who don't have the education or direct experience working with the specific technology in this case are not in any way shape or form qualified to be commenting on how safe or beneficial it'll be. You're just one more bobble-head believing and agreeing with the nonsensical hype you're being spoon-fed about this. The FACT of the matter is this isn't even real "AI" to start with, it's shitty half-assed "machine learning" crap, and at current it has to 'phone home' to have a remote HUMAN operator take over from it when it runs into something it can't handle. This reveals it to be a PIECE OF SHIT that should not be allowed on public roads. There will be traffic problems because of this garbage, there will be DEATHS that need not have occurred and that nobody will be held responsible for. DO NOT WANT!
Similar logical fallacy, circa 1900, used by folks wanting to demonize the horseless carriage.
Re: (Score:2)
"Similar logical fallacy, circa 1900, used by folks wanting to demonize the horseless carriage." and look how many people died (millions) or where injured (hundreds of millions, many more than once) as a result. Now what was your argument again ;P.
Re: (Score:2)
If this doesn't resonate with you, the human drivers where you live must suck considerably less than the ones we are exposed to daily.
*I've been distracted since last posting by a whiskey, some sleep, a dog on the wrong side of a door, and a mosquito bite; so in fair
Re: (Score:2)
The FACT of the matter is this isn't even real "AI" to start with, it's shitty half-assed "machine learning" crap, and at current it has to 'phone home' to have a remote HUMAN operator take over from it when it runs into something it can't handle.
So what?
If it reduces accidents, and the evidence we have so far is that it does, then bring it on. BTW, we have two forms of evidence: the numbers on accidents that we get from companies making self-driving systems and the willingness of at least some of those companies to accept full liability for any accidents caused by their systems. The former could possibly be dismissed. The latter is pretty compelling.
Re: (Score:2)
The FACT of the matter is this isn't even real "AI" to start with, it's shitty half-assed "machine learning" crap, and at current it has to 'phone home' to have a remote HUMAN operator take over from it when it runs into something it can't handle.
So what?
If it reduces accidents, and the evidence we have so far is that it does,
Wait, what? When did "we" get that evidence? All we have evidence of is that the group of alert drivers ready to take over at a moments notice coupled with a rule-based driving software that is only allowed to drive in perfect conditions results in less crashes than the group of all other drivers in all other conditions.
Let's wait until they put general-population drivers in with the software before calling it a success, okay? I want to see science done with a proper control group, which is group of alert d
Re: (Score:2)
You're ignoring all of Google's self-driving data (which includes lots of driving with general-population drivers). And my second -- and more compelling -- point.
I want to see science done with a proper control group, which is group of alert drivers driving in perfect conditions
That's not a proper control group, that's an unrealistic fantasy group.
Re: (Score:2)
You're ignoring all of Google's self-driving data (which includes lots of driving with general-population drivers). And my second -- and more compelling -- point.
I want to see science done with a proper control group, which is group of alert drivers driving in perfect conditions
That's not a proper control group, that's an unrealistic fantasy group.
The "evidence" you referred to in your OP was gained from -always-alert drivers correcting the SD software in perfect conditions. The control for such a group would always-alert drivers in perfect conditions.
It's not unrealistic - whoever is making the claims of "we drove X million miles with Y accidents" should divide their pool of drivers in half and have the control group drive around the same routes with SDC software turned off: that's the only way to tell if the SDC software is an improvement or not.
Re: (Score:2)
The "evidence" you referred to in your OP was gained from -always-alert drivers correcting the SD software in perfect conditions.
Nope. I've seen how the Google testing is done. And you're still ignoring the more compelling second point.
Re: (Score:2)
The "evidence" you referred to in your OP was gained from -always-alert drivers correcting the SD software in perfect conditions.
Nope. I've seen how the Google testing is done. And you're still ignoring the more compelling second point.
I haven't ignored it - I addressed it specifically:
It's not unrealistic - whoever is making the claims of "we drove X million miles with Y accidents" should divide their pool of drivers in half and have the control group drive around the same routes with SDC software turned off
All science is done using a control group in which only a single variable is changed, not one where multiple variables are changed. If studying the SDC capabilities then the control group has to be identical but without the actual SDC. As all the current SDCs in existence require a human anyway it is trivial to simply turn the SDC off and leave the human in place.
Re: (Score:2)
As all the current SDCs in existence require a human anyway
Google's doesn't.
Re: (Score:2)
The point that you're ignoring is that the relevant companies have expressed their intent to accept full liability for any accidents caused by their systems. That's the truly compelling argument.
As all the current SDCs in existence require a human anyway
Google's doesn't.
Perhaps I am being unclear: google has many millions of miles logged with a driver+sdc. To display that the sdc drivers better than the human they should log an equal number of miles on the exact same routes during the same seasons with the same cars and same drivers, but with the sdc turned off.
remember that your claim was that we already have evidence that sdcs are better drivers than humans, my claim was that there was no control group so that doesn't constitute any kind of scientific evidence. My claim
Re: (Score:2)
Automation is actually better for these types of tasks, than strong AI (if we had that) would be, because it is much faster and does not try to understand the situation. If it encounters something not in its profile, it just goes to a safe state as fast as possible.
Re: (Score:2)
Automation is actually better for these types of tasks, than strong AI (if we had that) would be
Yeah, strong AI might get bored and decide to focus on its poetry composition, rather than drive.
Re: (Score:2)
People like you who don't have the education or direct experience working with the specific technology in this case are not in any way shape or form qualified to be commenting on how safe or beneficial it'll be. You're just one more bobble-head believing and agreeing with the nonsensical hype you're being spoon-fed about this. The FACT of the matter is this isn't even real "AI" to start with, it's shitty half-assed "machine learning" crap, and at current it has to 'phone home' to have a remote HUMAN operator take over from it when it runs into something it can't handle. This reveals it to be a PIECE OF SHIT that should not be allowed on public roads. There will be traffic problems because of this garbage, there will be DEATHS that need not have occurred and that nobody will be held responsible for. DO NOT WANT!
Similar logical fallacy, circa whenever, used by folks wanting to demonize the Perpetual Motion Machine
FTFY. Just because something unrelated succeeded in the past does not mean new unrelated thing will succeed.
Re: (Score:2)
He never said how safe it was, he just said how safe it needs to be. But please, don't let reading comprehension get in the way of your talking point.
Re: (Score:2)
Thanks, nice to know that _some_ people here are not functionally illiterate.
Re: (Score:2)
Invalid AdHominem. The rest of your comment is not even worth reading...
Re: (Score:2)
Indeed. The thing is that as soon as safety is adequate, the next thing they will tackle is cost, as in reducing wear&tear, reducing fuel consumption, reducing damage to transported goods, etc. One thing they already have is increased flexibility, because of no rest times and no other biological limitations. Pretty soon a human driver will be entirely outclassed and hugely more expensive.
This stuff is coming. The only sane thing for those it will replace is to make sure they have other marketable skills
Re: (Score:1)
That's a bit simplistic. Of course they would lobby for it. That by itself doesn't make it wrong. Sometimes the opposing side wants an artificial restriction. For instance, the labor unions opposed it because they don't want truckers to lose their jobs. Well of course that's what will eventually happen. But stopping progress to preserve legacy jobs is little different than extended government welfare. The people who would benefit (i.e. everyone else) are de
Unions (Score:5, Insightful)
Because of unions. Next question?
Re: (Score:2)
Self driving trucks don't go on strike
Re: (Score:2, Troll)
Re: (Score:2)
Source? K, thx.
Uh, the article?
Re: (Score:3)
Neither the House nor the Senate bill would speed approval of self-driving technology for vehicles over 10,000 pounds, a step pushed by trucking organizations. Labor unions raised safety and employment concerns, and Democrats resisted that part of the proposal.
Is there a need? (Score:2)
Re: (Score:1)
Is there really a need for self-driving cars?
Yes. Human-driven cars kill 30K people every single year in the United States. That is like the recent Las Vegas shooting, two times every day of the year.
Self-driving won't be perfect, but if it is a major improvement on human-driven, then yes, we do need this. It needs to be phased in and eventually legislated to outlaw the extremely dangerous human driven cars.
It won't take that much to beat humans. It may not be immediate, but it WILL happen. That is a matter of when not if.
Re: (Score:2)
I guess the answer is yes if you stand to profit from removing the human element of your business.
Wages are paid from revenues. Revenues come from price. Consumer buying power is increased by technology. Technology has a single purpose: reduce the working hours put in to get a particular result.
Nations with a high standard-of-living and high GDP-per-capita are operating on high technical progress: they invest fewer labor-hours to produce the same goods. For example: poorer nations may invest 10-20 times the human working hours to produce the same amount of food, and must expend more of their in
Nope. (Score:1)
unanimously approved a bill to speed self-driving cars to market without human controls and bar states from imposing regulatory road blocks.
So, defacto unconstitutional, then.
The federal government has no authority to bar states from passing laws for items not specifically named in the Constitution.
Re: (Score:2)
Re: (Score:2)
So how did they do this? By tying it to federal highway funding. "Hey, state legislatures, that looks like some nice roads you've got there in your state. It would be a shame if anything were to happen to them because you lose your federal highway funding and are unable to
The speed limit had very low compliance, (Score:3)
The speed limit had very low compliance and some states replaced traditional speeding fines with $5–$15 energy wasting fines
Re: (Score:3)
Re: (Score:2)
Interstate trucking would be the least dangerous type of self-driving vehicle usage.
Since interstate truck drivers are already heavily regulated for safety, probably also one of the lowest returns on investment based on safety concerns. But also one of the the highest potentials for damage when the AV fails. Interstate triple-bottom freight haulers going wacko on the interstate highways could take out a lot of humans and infrastructure.
Trucking AV would be a windfall for shipping companies, with no clear guarantee that shipping rates would reflect the savings.
Re: (Score:2)
Republicans are hypocrites (Score:1)
It is amazing how Republicans complain about Democrats passing laws that restrict "state's rights", but then do the same when they want.
Re: (Score:2)
Yes, by all means mod me troll for quoting RationalWiki. Morons.
Re: (Score:2)
Re: (Score:2)
That used to be the case until the definition of "interstate commerce" was expanded with the blessings of the Supreme Court to include pretty much any commercial activity at all
The ICC has certainly been abused plenty, but we have federal motor vehicle standards for a reason, and that reason really is interstate commerce.
Yesterday I asked! Today we have an answer! (Score:2)
"Do they have any liability protections built in to the laws to protect the companies deploying these vehicles?"
What are the Liability ramifications [slashdot.org]
Today we have some answers, and one could maybe? think the public "IS" being used as test dummies
Who in the event of accidents, issues, problems "MAY" have to live with arbitration
Re: (Score:2)
When self-driving cars can navigate a snow covered road, with other snow covered cars, snow covered trees, and a layer of ice at night, I will consider letting it drive for me.
Well, the autonomous terrain vehicles will communicate with other snow-covered vehicles, the snow covered trees should be out of the roadway, and it seems reasonable that regions prone to inclement weather would incorporate sensors into the roadways.
Re: (Score:2)
Well, the autonomous terrain vehicles will communicate with other snow-covered vehicles,
Explain how your Ohlenhausen Smugmobile 3100 AV car will communicate with my 2017 Subaru that is parked on the side of the street covered by snow.
and it seems reasonable that regions prone to inclement weather would incorporate sensors into the roadways.
Based on quotes one and two here, I am assuming you think that AV will only be allowed on the road after all other vehicles are removed from service and every municipality has invested millions of dollars in embedded sensors for all their roads. Until that happens, both of your solutions to the problem do not exist.
Re: (Score:2)
Explain how your Ohlenhausen Smugmobile 3100 AV car will communicate with my 2017 Subaru that is parked on the side of the street covered by snow.
Your 2017 Subaru will (eventually) be required to carry a V2V beacon which uses GPS and whatever the current communications standard is to communicate vehicle position and speed, as well as throttle position and the state of the brake light switch. All of this is trivial to report from most vehicles — whether via OBD-II or simply tapping the relevant signals with high-resistance probes that won't affect the vehicle's behavior. The only vehicles which will be at all difficult to support will be those l
Re: (Score:2)
Your 2017 Subaru will (eventually) be required to carry a V2V beacon which uses GPS
By "eventually" you mean "never". Another example of impossible requirements being required before AV work well in all situations.
Stop pretending that vehicles buried in snowdrifts is a problem everywhere. It isn't.
It is a problem that you were presenting a solution for, a solution that is clearly impossible until two impossible things happen (and now three). It doesn't matter if it doesn't happen everywhere, it will happen and your solution is not.
Re: (Score:2)
Have you never been somewhere that is snows? Where I live, it doesn't snow much. But on several occasions I've seen a drift melt and a car appear.
So SDCs will be suitable there last. So what? Nobody is pretending it isn't a problem. But it is a solvable one, given enough time.
Re: (Score:2)
Apparently you're unaware that radar is actually used to detect snow, and this works because snow reflects radar. So in a heavy snowstorm, your radar is going to be pretty much useless. Same for lasers or other optical means. You need the ability to create models of object moving at high speed in real time based on incomplete information. You also have to compensate for changes to those models using controls that suddenly exhibit non-linear behavior and responses, with the nature of the non-linearity changi
Re: (Score:2)
No commercial truck provisions? (Score:5, Insightful)
Re: (Score:3, Insightful)
This is stupid, long-haul trucking is the industry that would benefit most from self-driving vehicles and it is also one of the easier challenges for the auto-industry.
It is not time to have fully automated class 8 trucks driving around. They are just too big and thus too dangerous. Normally sized vehicles are dangerous enough. Give it a few more years, and you'll see similar legislation for heavy trucks.
It is also not easier to do self-driving trucks, because they have to be able to handle all the same situations that self-driving cars have to be able to handle. Or have you never seen a person or a deer running across a freeway? How about stuff sitting in the road, like
Re: (Score:2)
Re: (Score:2)
It's cute how you think AI is strong enough to pick out a deer behind a bush *and* anticipate that it is about to run out into the road
Well, I actually experienced something similar while riding in one of Google's self-driving cars. It wasn't a deer, he was a cyclist, and it wasn't a bush, it was a hedge, but the car saw it and delayed entering the intersection to let him go past, even though no human driver could have known the cyclist was coming.
I'll grant that knowing which way a deer is going to jump is not achievable with AI -- it's not achievable with human intelligence, either. But what AI could do is to decide to slow a bit, or p
Re: (Score:2)
Re: (Score:2)
Again, there has been no publicity on how realistic it would be for Google's self driving array to be put into, let's say, a Chevy Spark
Why would it not be realistic? The Google-made test cars are even smaller than a Spark. All the system requires is some sensors around the perimeter, the lidar unit on top and a smallish CPU case inside.
so it is unknown whether the vast population will ever be able to afford such functionality.
I doubt that most people will ever buy a self-driving car. Not because they'll be priced out of it -- there's nothing inherently expensive in any of the self-driving equipment; it's expensive now but by the time it's made in volume it won't be -- but because fleets of self-driving taxis will make it so that
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
I doubt that most people will ever buy a self-driving car. Not because they'll be priced out of it -- there's nothing inherently expensive in any of the self-driving equipment;
It's not that the self-driving car is going to be expensive, it's that people won't be able to afford a car, period; their prices are rising faster than inflation, and wages are rising slower.
Re: (Score:2)
Of course, in the case of a semi-tractor, the right decision would be to just maintain course and speed. If the deer jumps in the road, well, too bad for the deer. Semis have big and very strong bumpers for just such occasions.
Nope. A deer can still take out the radiator, if you catch it in mid-jump. The right answer is to brake as much as possible before you hit it, to minimize damage.
Terribly Wrong Link (Score:4, Informative)
Did they even check? Obviously not.
Here's the actual article: http://www.reuters.com/article... [reuters.com]
Re: (Score:2)
I'd mod you up if I had points. Not sure how they missed the link in the summary that leads to something Trump related.
Great article there, but what about cars? (Score:3)
I'm guessing that was a copy/paste error.
Barring states from regulatory road blocks (Score:2)
Can we have a similar ban to protect health-insurance providers from regulatory road blocks?
Unlike the transportation industry in general and the private cars in particular, health insurance is, actually, a disaster in need of addressing.
What argument is there to justify prohibiting me from purchasing health insurance from another state, that would not also justify my state imposing additional requirements to self-driving cars?
Re: (Score:2)
Re: (Score:2)
Most of the insurance industry would likely move to the state that offers the most lenient regulations,
Except under ACA we have federal regulations preempting state. Tell us why insurance that meets the federal standards from a company in Mass., e.g., should not be available to someone in California.
I heard this discussion on a talk program last week, and the person trying to defend state limits tried using the excuse that insurance from other states would probably not provide coverage for the providers in our state. I.e., the Mass. insurance would not have a provider network that included doctors in Calif
Re: (Score:2)
Re: (Score:2)
Products that are sold across state lines are subject to regulations of the state they are sold in. Most of the insurance industry would likely move to the state that offers the most lenient regulations,
Each state has its own laws regarding automobile insurance, and they apply to vehicles registered within the state. Insurance companies can be located anywhere, and it's quite irrelevant. You get a policy which is written for you, and it takes the state of registration into account.
Re: (Score:2)
Re: (Score:2)
Yes, this is true about both cars (human or computer-driven alike) and insurance policies.
Yes, this is also true about car-makers.
My question remains, if my state is disallowed to impose "regulatory road blocks" on sellers of self-driving cars, why is it allowed to impose such blocks on sellers of insurance?
across state lines? with differnt rules? (Score:2)
across state lines? with differnt rules?.
So what will happen with a car that is ok under sate A rules but not sate B rules?
Re: (Score:2)
What happens when a car without insurance leaves New Hampshire (where it is legal to drive without insurance) and enters Vermont, Maine, or Massachusetts (where it is illegal to drive without insurance)? The driver gets a ticket. I guess the only difference here is that instead of "the driver" it will be "the person responsible for the car"
Re: (Score:2)
And the ticket will be issued automatically when the vehicle crosses the state line. Fortunately, with our new self-driving overlords, the ticket will also be paid automatically (just like tolls) by the car when the citation is issued.
Personally, I'd vote for cars that automatically stopped and refused to drive if there was no insurance policy registered against them. Bet we won't see any legislation like that, though.
Re: (Score:2)
Bet we won't see any legislation like that, though.
Because it is not mandatory to have insurance to drive a car. To use it on the public roads, perhaps.
the first child that dies by self-driving truck (Score:2)
This will all fall apart the first time a wealthy white family loses a child to a legal self-driving truck.
Mark my words.
Insurance? (Score:5, Insightful)
If I'm not driving it, it should only need to be insured against damage caused by others. If it causes an accident, that's the manufacturers liability, not mine, as I wasn't driving it. If it's any way made to be my liability, I'm not going to buy it. I'd rather hire a self driving taxi than own the liability of my own self driving car.
Re: (Score:2)
If I'm not driving it, it should only need to be insured against damage caused by others. If it causes an accident, that's the manufacturers liability, not mine, as I wasn't driving it. If it's any way made to be my liability, I'm not going to buy it. I'd rather hire a self driving taxi than own the liability of my own self driving car.
Volvo, Google and Mercedes-Benz have all publicly stated that they will take full liability for accidents caused by their self-driving systems. I expect that this will become the norm, because it just makes sense. If the self-driving system is in control and makes an error that causes an accident, the system is faulty and not fit for its stated purpose, and the maker of the system bears the liability.
However, I predict that many people will never buy a self-driving car. Instead, they'll go from owning a m
Re: (Score:2)
However, I predict that many people will never buy a self-driving car. Instead, they'll go from owning a manually-driven car to using a self-driving car fleet, on a pay per use basis.
If manufacturers are taking on the liability for self driving cars, they are going to want control of maintenance, repairs, and especially modifications. They'll surely not want to be on the hook for an accident that could have been avoided with better maintenance or if the car had not been modified by the owner. I'm expecting manufacturers to either create a rental service (bonus points for the first one to call it Johnny Cab), or partner with someone like Lyft or Uber, so they have better control of ma
What? (Score:2)