Germany Plans Highway Test Track For Self-Driving Cars 90
An anonymous reader writes with news about a new project to test autonomous vehicles in Germany. "The German government wants to convert part of the A9 Autobahn in Bavaria into a test-field for advanced car technology. The project is key to ensuring the country's 'digital sovereignty,' according to its transport minister. The track, part of the 'Digitales Testfeld Autobahn' project, would be launched this year, Alexander Dobrindt said on Monday in an interview (in German) with the Frankfurter Allgemeine Zeitung newspaper. The plan involves equipping the road with infrastructure to allow cars to communicate with each other and the road's own sensors to provide necessary data on traffic. 'Cars with assisted driving and later fully-automated cars will be able to drive there,' Dobrindt said. Germany, a major European car producer, wants to have robotic car technology that's not dependent on foreign companies, the minister said. Domestic producers 'won't rely on Google' he stressed."
Still sounds like early flight... (Score:4, Insightful)
I'm still reminded of what I've read about the Wright Brother's attempts at powered flight, up against dozens of other teams, some with national support.
I look forward to seeing them, because with efforts taking place in Germany, Japan, and the USA to just name the 'big 3', somebody is probably going to succeed in fairly short order(still years though).
Car accidents cost us enough to more than pay for it.
Re:Still sounds like early flight... (Score:4, Interesting)
I think it's a mistake for cars to communicate with each other. I don't trust programmers to write bug-free and exploit-free software to run on those cars, and I don't want someone's car lying about its speed or other characteristics such that it may cause me to crash.
Bott's dots (Score:2)
Huh, actually had to google what Botts' dots [wikipedia.org] are. FTL: rarely used in regions with substantial snowfall, because snow plows damage or dislodge them.
Wouldn't do much good up where I live if they can't take snow plows. Instead of dots to provide 'rumble' we put notches in the pavement.
Now something that you can sink into the pavement, sort of like a concrete screw? That might work.
I don't want someone's car lying about its speed or other characteristics such that it may cause me to crash.
Trust but verify - IE while you trust other cars for providing road condition information, you also don't trust it enough to le
Re: (Score:2)
I can see the headline now: "340 car pileup caused by 'terrorist' who rented a snowplow."
Rented Snowplow (Score:2)
Yeah, except that I predict that it'd be more along the lines of '340 car traffic jam' because they yanked up the dots and rearranged them.
Seriously, 'avoiding collision' is really the most programmed for condition for auto-drive cars. Combine an ever-vigilant computer that's never distracted with defensive driving that 'only' trusts it's redundant sensor suite* and you should have to work very hard to get it to collide with something.
*And refuses to drive if those don't work.
Re: (Score:2)
Here in Texas, we have those, except they are nicknamed "Braille lane markers", and are square, with one side beveled and reflective white, the other side beveled and reflective red (so if one sees a bunch of them red, that should give a hint that one is going the wrong way on a highway.)
I too am leery about depending on other cars. Yes, a module could be made if the core chip was made as secure as the Clipper chip (where the dies were put in a top secret area where the Skipjack algorithm was written onto
Re: (Score:2)
So long as the collision avoidance logic is still in charge to prevent collisions, I'd be more concerned about an OTA update attack that changes the auto-drive programming.
Such as to direct hitting something in certain circumstances...
Re: (Score:2)
I'm actually not so worried about terrorism as I am people attempting to give themselves priority. In some cities there are already problems with non-emergency vehicles activating the sensors that change the traffic lights so that emergency vehicles get the right-of-way, I don't want people tricking self-driving cars into merging right or waiting. That's
Re: (Score:1)
The problem of trust and reputation that such a system has a weakness for is very similar to reputation/trust systems used for protecting computers these days. The idea is that if you have enough data points, you can ignore the anomalies, or flag them up as anomalous (as in: "car data from car 3 up in your lane doesn't match the curve -- pay extra attention to sensor data from that car" being broadcast out to all the other cars in the area).
Now if this was supposed to be some sort of trusted message passin
Re: (Score:2)
The good thing is that this is a solved problem, as the theory is sound. It is the implementation of a trust system and the "if a nail sticks out, pound it down" parts that need to be worked on. That way, a vehicle telling cars to "slam brakes, veer hard left" while everything else around is giving an "all clear" can be ignored or weighted negatively (thank SpamAssassin), with other vehicles passing the "dude, this car over here is on crack; ignore it" messages to others around.
Of course, the hard part is
Re: (Score:1)
This isn't impossible... it just is something that hasn't been a focus by companies since the Cold War.
Actually, it has: this is precisely the same domain being tackled by the smart card industry (think cell phones, satellite TV, ATMs). Some in those industries have gone for the quick buck, but the security analysis and implementation guidelines have been continuously worked out since the 90's (including the voltage input and line emissions broadcast issues).
So really, the only issue is for the engineers to 1) read the right supporting material and 2) give the right pitch to their business units. Of course
Re: (Score:2)
That may not work out well; if a car has declared an emergency then it's more likely to be an actual emergency than it is erroneous crap, and that may be the first or only car to discover that it's an emergency. You can't
Re: (Score:2)
I've been thinking about this for a long time. Early on I thought about embedding a cable in the road in the center o
Re: (Score:2)
Huh, actually had to google what Botts' dots [wikipedia.org] are. FTL: rarely used in regions with substantial snowfall, because snow plows damage or dislodge them.
Wouldn't do much good up where I live if they can't take snow plows. Instead of dots to provide 'rumble' we put notches in the pavement.
Now something that you can sink into the pavement, sort of like a concrete screw? That might work.
Another trick that is becoming more common is notching the pavement with an inset for the Botts' dot so it doesn't stick up into the plow blade.
Re: (Score:1)
While it may be true that cost savings and the injury and grief associated with accidents is considerable. Few people are willing to pay for it. People want immediate intangible benefits.
Reduced insurance rates, reduced traffic jams including the use of the high-speed auto-drive lane, self-parking cars after drop-off, not having to chauffeur the kids, watching TV while "driving"
These are the reasons people will buy self-driving cars.
Reduced Car accidents - Reduce Car Insurance (Score:2)
As you mention, reduced insurance rates is a reason to buy the new vehicles, and that ties in directly with reduced car accidents. The other stuff is bonus.
Though when I figured it out, valuing people's time at $10/hour NOT spent driving, that was the biggest factor.
15k miles/year average per vehicle, figure an average speed of 30 mph, that's 500 hours/year. Or $5k. Even if you only value your time in the car at $5, perhaps because you get motion sickness if you're not driving* so you can't read/watch TV
Re: (Score:2)
Even if you only value your time in the car at $5
Unless I'm currently working, I don't value my time using money in the first place; it's ridiculous.
Time value of money (Score:2)
Do you own a car? If so, do you change it's oil or take it to a place? Why?
Most people don't do it as explicitly as I did, but people still do it. It's one of the major reasons people drive rather than taking the bus. Sure, it's more expensive to drive, but they value their time highly enough that they'd rather spend $5 to get there in 1/3rd the time that the bus would take.
People especially do it when they hire a contractor to fix something in their home, mow their lawn, etc...
Re: (Score:2)
The idea that you need money to place value on your time is what is ridiculous.
Sure, it's more expensive to drive, but they value their time highly enough that they'd rather spend $5 to get there in 1/3rd the time that the bus would take.
This sort of attitude is why I have no sympathy when I lot of people complain about being in debt. There's an easy alternative and yet they make no attempts to educate themselves or invest their money.
Re: (Score:2)
I'm still reminded of what I've read about the Wright Brother's attempts at powered flight, up against dozens of other teams, some with national support.
What is that bullshit about "dozens of other teams"? You obviously spout misinformed nonsense - the only opponent with some national support was Langley. All others (Whitehead, Santos-Dumont, Ellehammer, whoever) paid their work out of their own pocket. But at least they laid all their results open and did not try to sue everyone else on the planet like these trolls from Dayton, OH.
Re: (Score:2)
You obviously spout misinformed nonsense - the only opponent with some national support was Langley.
You spout things you do not know. I never stated that they all had national support. I don't remember who, but Langley wasn't the only one with government support. Which is why I said 'some'.
Re: (Score:2)
Indeed. Safe individual transportation would be a great boon. And it could revolutionize parts of public transportation in general. Example: Need to transport something heavier? Order a self-driving car of any size desired. Of course, the Taxi-industry will likely be a casualty of this, but no historical job-setting lasts forever.
Re: (Score:2)
I don't know, the world's 2 most ancient professions are still around... ;)
Actually, I'd imagine that truck drivers, especially long haul, would suffer before taxi drivers.
The taxi 'industry' would be fine, perhaps even invigorated by this. Fire all the nasty taxi drivers, have a computer do dispatch, etc...
Reduce costs enough and people will be less likely to buy a car rather than just renting one when they need it.
As you say - need a heavier vehicle, rent one, even over the phone. Heck, buy something an
what about liability? and maybe even criminal liab (Score:1)
what about liability? and maybe even criminal liability?
Just think of a auto drive loosing control and plowing through a school crossing killing a dozen children. Who or what is responsible? The passenger? Or the computer?
Re: (Score:1)
Re: (Score:1)
This is Germany, where people know how to drive and the traffic laws are actually made sanely.
Re: (Score:1)
And cars do as they are told, or they will be punished!
Re: (Score:2)
Re: (Score:3)
but I'm guessing the Germans have their share of less than ideal drivers.
Sure, but they're the 1%ers, not the 99%ers, because just getting a license is much harder and the German Police will bust your ass for things like tailgating or not signaling before they will for speeding.
As for the AC's worry about an auto-drive car hitting somebody. I'm sure it will happen, but will be incredibly, incredibly rare.
So rather than 'hit by a drunk driver while crossing the street going to Church Sunday morning', it'd be more along the lines of 'failed to stop in time when pedestrian unexpec
Re: (Score:2)
You are very far off from reality.
Most high cost brands/types of cars already have electronic systems to assist the driver. Like lane recognition, sign recognition and pedestrian recognition. As they are based on cameras and the last 30 seconds are stored liability is no problem.
If a pedestrian runs into your lane and the car does an automatic emergency break AND the car following you crashes into you because of that, the liability issues are clear.
First the pedestrian is liable, for forcing the emergency b
Re: (Score:2)
You are very far off from reality.
Since so much of what you said is the same thing as I said, doesn't that make you far from reality as well?
For example, you said: "If a pedestrian runs into your lane and the car does an automatic emergency break AND the car following you crashes into you because of that, the liability issues are clear."
Which is simply an expansion/different case on my "pedestrian considered at fault for darting into traffic."
You don't address my pointing out that the system maker could be held at fault, but I specified 'po
Re: (Score:2)
Here in Texas, one could go to a school, take a semester offered in high school, then there would be two phases. A written test for a learner's permit, then the test with an officer sitting in the car for the actual drive.
Licenses here need to be renewed every six years, one renewal allowed via the Internet, one with a visit to the DMV for checking vision.
Re: (Score:2)
i'd suggest avoiding school children crossings on the autobahn.
Re: (Score:3)
Just think of a auto drive loosing control and plowing through a school crossing killing a dozen children. Who or what is responsible? The passenger? Or the computer?
The school that put its children on the fucking Autobahn, a high-speed road that is by law off-limits to pedestrians, bicycles and anything else that can't reach and maintain the minimum speed of 60 km/h.
Re: (Score:2)
And what about when a human driver does this? I look forward to the day when humans are banned from driving. I've been driving for 10 years and have had people smash their cars into me from so many angles it's not even funny. Got a concussion from one encounter. All of the people behind the wheel were either high or drunk. To the brain-dead idiots who say that computers will never be as good at driving as humans are (which is just a selfish excuse so they can continue to 'enjoy' the 'thrill' of driving), I
Comment removed (Score:3, Interesting)
Your situation isn't everybody's. (Score:3)
There's still a lot of cars in Germany, something doesn't have to apply to 90% of people to still be highly useful.
Also, self-driving cars can also increase fuel economy through a combination of reduced speeds(no rush if you're reading), more fuel efficient driving(let the car figure out the ideal acceleration rates and such), and perhaps even stuff like predicting the next light to avoid having to stop at all.
Oh, and if you want to keep your export economy...
Re: (Score:2)
Don't forget safe tailgating by reducing the distance between cars as the first car can signal to the trailing car that it's going to break before even the actual brakes are moving. No need to factor in human reaction time anymore.
Re: (Score:2)
Research done has shown that for real mpg improvements you need to be closer than even a computer controlled car can compensate for, and you pay for it by needing to brake so often that you burn off any potential savings.
Computer controlled cars might be able to do it better, but do you trust the signals from the lead car?
Re: (Score:2)
Research done has shown that for real mpg improvements you need to be closer than even a computer controlled car can compensate for, and you pay for it by needing to brake so often that you burn off any potential savings.
Mythbusters isn't "Research". This is. [sartre-project.eu]
Re: (Score:2)
15 meters is the max they measured, you really need to be within 9 meters to realize 10% fuel savings.
That close you're looking at an impact by following vehicles if something happens to make the lead vehicle abruptly stop or slow.
Also, I wasn't considering dedicated 'lead' vehicles like trucks, but other cars, and computer driven by lots of different companies.
IE I don't trust their communications.
Re: (Score:2)
15 meters is the max they measured, you really need to be within 9 meters to realize 10% fuel savings.
No, 8 meters is the closest they have measured with cars, and they only didn't g closer because the build-in proximity sensors (safety standard in the production cars) didn't allow them to go closer without the breaks pre-charging, ruining mileage. And 10% savings with no changes to technology (apart from the platooning system of course) or driving is pretty good, isn't it? Of course the least saving was showing for the big petrol engined car, so it's clear that this isn't for America with its huge engines.
Re: (Score:2)
And 10% savings with no changes to technology (apart from the platooning system of course) or driving is pretty good, isn't it?
Only on a closed track, and remember that my assertion isn't that the gas savings aren't there, it's that even with self-driving cars 8 meters isn't safe once you start trying to move it to production, especially when you'd have cars of different makes, and maintenance levels in the 'platoons'. It'd also be limited(mostly) to the highway systems, which doesn't do much for most commutes.
Re: (Score:2)
And 10% savings with no changes to technology (apart from the platooning system of course) or driving is pretty good, isn't it?
Only on a closed track, and remember that my assertion isn't that the gas savings aren't there, it's that even with self-driving cars 8 meters isn't safe once you start trying to move it to production, especially when you'd have cars of different makes, and maintenance levels in the 'platoons'. It'd also be limited(mostly) to the highway systems, which doesn't do much for most commutes.
The safety brake system (used in production cars) I mentioned was obviously tested to avoid collisions with suddenly breaking cars just 8 meters away without even pre-charging the breaks. Not to mention that the 10% saving is for 15 meters.
But you sure do know more about these things than the people who build them. Volvo. The guys with the car safety record.
Re: (Score:2)
When I worked in Germany, my car was a weekend toy and I took the U-Bahn to work. However, most of the people in my group drove to work as did my "handler" (the guy who dealt with the paperwork so I could work in Germany).
Re: (Score:2)
And yet every time I have been to Germany there seems to be a load of cars on the roads. And trucks as well.
I drove from Bremen to Frankfurt a little over a year ago and Frankfurt was full of cars.
Re: (Score:2)
Frankfurt is indeed full of cars. Still the public transport is full of people. By the way, I was faster from Offenbach center to Frankfurt center on my mountain bike than a Ferrari that really rushed from traffic light to traffic light.
Gives a new meaning... (Score:2)
...to das Auto. Then again, that pun already works in English, so nevermind.
Re: (Score:1)
It means Hack The Auto in Deutsch.
Lack Of Faith (Score:4)
Re: (Score:2)
Are are aware that VW is our low-end brand, yes?
BMW and Mercedes are the high-end brands, as is Porsche.
Re: (Score:2)
Are you aware that BMW and Mercedes reliability has gone into the toilet since the 1980s? I hear most Porsches that don't catch on fire spontaneously are pretty reliable, though.
Re: (Score:2)
Are you aware that BMW and Mercedes reliability has gone into the toilet since the 1980s?
The M3 I drove last year begs to differ. As did the SLK the year before. :-)
Maybe they have problems, I don't know, I don't own a car, I just rent them pretty often, and I'll take one of those every day over almost any brand. At least until my car rental company gets Teslas.
Re: (Score:2)
And Audi :)
Re: (Score:2)
You appear to have consolidated your car industry.
Now they all perform like VWs.
Have maintenance costs like Porsches.
Part availability like Audi.
And ease of maintenance like BMW.
You took the worst traits of your brands and transferred them to all the others.
Re: (Score:2)
FWIW, the Audi dealer still lists... must be at least 95% of the parts for my 80's audi, down to the nut. I don't think any american marque does that..
Of course they charge you for it, but they do carry it.
Re: (Score:2)
Well, except for BMW, all the brands you have listed belong to the same company. That might be the explanation.
Re: (Score:2)
As much as I love the driving dynamics of high-end German cars, I also accept that intrinsic unreliability is part of the cost of ownership. The salesmen at Porsche dealerships laughingly refer to things like rain entering the driver compartment as "The Porsche Experience". BMWs are fantastic...until they are about 5 or 6 years old, at which point they have substantially more problems (at substantially higher repair costs) than 25 year old Toyotas and Hondas. The factors which make some German brands hig
Re: (Score:2)
Could be, as I rent and don't buy, I don't drive cars older than a few years.
I know the Toyotas and Hondas are famous for their reliability. My first car was a used Honda and it had almost no signs of being used before.
That said, old Mercedes cars are also legendarily reliable. My GF wants to buy a used SLK for exactly that reason - they are cute and almost as good as new, for a fraction the price.
Re: (Score:1)
Meh. I drove a Chrysler PT Cruiser, and the door didn't even shut properly. Constant draft whizzing past my ear from the gap between the door and the roof, and water coming in where the seatbelt connects. If we're just giving examples to then blindly extrapolate a trend, then I claim that German cars are far superior to US cars.
"Digital Sovereignty" is part of a bigger issue (Score:2)
By me, a decade ago: http://www.pdfernhout.net/on-f... [pdfernhout.net]
"Consider again the self-driving cars mentioned earlier which now cruise some streets in small numbers. The software "intelligence" doing the driving was primarily developed by public money given to universities, which generally own the copyrights and patents as the contractors. Obviously there are related scientific publications, but in practice these fail to do justice to the complexity of such systems. The truest physical representation of the knowled
Going to be a lot of dead kids and pets (Score:1)
Test tracks rarely allow for what happens in the real world when snow, rain, and fog combine with small kids and pets playing.
How many billions in lawsuits for their lifetime (a kid lives 100 years, and becomes a CEO that means $40 billion each kid) will these Steel Death Automatons rack up before they are outlawed except in retirement communities without kids or pets?
Re: (Score:1)
and then the auto cars will fail big time when they get to the local roads
Re: (Score:2)
Re: (Score:2)
How many billions in lawsuits for their lifetime (a kid lives 100 years, and becomes a CEO that means $40 billion each kid) will these Steel Death Automatons rack up before they are outlawed except in retirement communities without kids or pets?
Zero billions, because the auto companies' lawyers are quite aware of liability issues, and so they aren't going to allow the sale of any self-driving car to the public until they're damn sure it's smart enough to avoid running over pets and children.
So either the automobiles will reliably detect and avoid pets/children, or they will never be released to customers.
Re: (Score:2)
Isn't the point to test automatic cars under real conditions? Google did this years ago. With hand-picked, pre-mapped roads, but still under real conditions with real human-driven traffic. Remember the euro search engine? The euro book digitizing project? Every time Germany/EU tries to copy what Google does, only years later, by government decree and without Google, the result is the same. Burnt money. Next thing they will try to ban undeutsche autonomous cars from deutsche autobahn.
German (and others) scientists did that years before Google even existed. https://en.wikipedia.org/wiki/Eureka_Prometheus_Project [wikipedia.org]
PROMETHEUS profited from the participation of Ernst Dickmanns, the 1980s pioneer of driverless cars, and his team at Bundeswehr Universität München, collaborating with Daimler-Benz. A first culmination point was achieved in 1994, when their twin robot vehicles VaMP and VITA-2 drove more than one thousand kilometers on a Paris multi-lane highway in standard heavy traffic at speeds up to 130 km/h. They demonstrated autonomous driving in free lanes, convoy driving, automatic tracking of other vehicles, and lane changes left and right with autonomous passing of other cars.
https://www.youtube.com/watch?v=2EH3R6c7Ufg [youtube.com]
I have a joke: (Score:1)
It's called the No-onebergring.
Obligatory joke... (Score:1)
In Soviet Germany, car drives *you*!
Re: (Score:2)
The A9 connects Munich to Ingolstadt, which are respectively the main hubs for BMW and Audi, so it makes sense to use that road for testing. Moreover, both cities are in the state of Bavaria, which makes it easier to get things going on the government level.
Bavaria is also the home state of the traffic minister Dobrindt - so pork.