US Presses the 'Reset Button' On Technology That Lets Cars Talk To Each Other (npr.org) 95
An anonymous reader quotes a report from NPR: Safety advocates have been touting the potential of technology that allows vehicles to communicate wirelessly for years. So far, the rollout has been slow and uneven. Now the U.S. Department of Transportation is releasing a roadmap it hopes will speed up deployment of that technology -- and save thousands of lives in the process. "This is proven technology that works," Shailen Bhatt, head of the Federal Highway Administration, said at an event Friday to mark the release of the deployment plan (PDF) for vehicle-to-everything, or V2X, technology across U.S. roads and highways. V2X allows cars and trucks to exchange location information with each other, and potentially cyclists and pedestrians, as well as with the roadway infrastructure itself. Users could send and receive frequent messages to and from each other, continuously sharing information about speed, position, and road conditions -- even in situations with poor visibility, including around corners or in dense fog or heavy rain. [...]
Despite enthusiasm from safety advocates and federal regulators, the technology has faced a bumpy rollout. During the Obama administration, the National Highway Traffic Safety Administration proposed making the technology mandatory on cars and light trucks. But the agency later dropped that idea during the Trump administration. The deployment of V2X has been "hampered by regulatory uncertainty," said John Bozzella, president and CEO of the Alliance for Automotive Innovation, a trade group that represents automakers. But he's optimistic that the new plan will help. "This is the reset button," Bozzella said at Friday's announcement. "This deployment plan is a big deal. It is a crucial piece of this V2X puzzle." The plan lays out some goals and targets for the new technology. In the short-term, the plan aims to have V2X infrastructure in place on 20% of the National Highway System by 2028, and for 25% of the nation's largest metro areas to have V2X enabled at signalized intersections. V2X technology still faces some daunting questions, including how to pay for the rollout of critical infrastructure and how to protect connected vehicles from cyberattack. But safety advocates say it's past time to find the answers.
Despite enthusiasm from safety advocates and federal regulators, the technology has faced a bumpy rollout. During the Obama administration, the National Highway Traffic Safety Administration proposed making the technology mandatory on cars and light trucks. But the agency later dropped that idea during the Trump administration. The deployment of V2X has been "hampered by regulatory uncertainty," said John Bozzella, president and CEO of the Alliance for Automotive Innovation, a trade group that represents automakers. But he's optimistic that the new plan will help. "This is the reset button," Bozzella said at Friday's announcement. "This deployment plan is a big deal. It is a crucial piece of this V2X puzzle." The plan lays out some goals and targets for the new technology. In the short-term, the plan aims to have V2X infrastructure in place on 20% of the National Highway System by 2028, and for 25% of the nation's largest metro areas to have V2X enabled at signalized intersections. V2X technology still faces some daunting questions, including how to pay for the rollout of critical infrastructure and how to protect connected vehicles from cyberattack. But safety advocates say it's past time to find the answers.
Not useful tech. (Score:5, Insightful)
Any vehicle to vehicle system is inherently flawed by design. You can't trust another vehicle. As soon as you do anything more than slightly increasing following distance in response to external data that could be forged, somebody will start broadcasting fake data and cause a pile-up.
Even with massive server infrastructure and centralized algorithmic processing of the data, even companies like Apple and Google can sometimes be fooled by a wagon full of cell phones [seattletimes.com]. Your car has a tiny fraction of the data that those sorts of companies do.
Re: Not useful tech. (Score:2)
Pile up, or force the car to stop and steal whatever the occupants have on them or kidnap them.
Re: Not useful tech. (Score:4, Insightful)
force the car to stop and steal whatever the occupants have
You don't need V2X for that.
Just wait at a traffic light with a shotgun and rob people when they stop.
Re: (Score:2)
That's true, but doesn't take away from the fact that this would be a problem. Traffic lights are in predictable places and don't move much, which makes them much easier to police. If someone starts this as a crime they repeat you could much more likely catch them. If they do it in the middle of the countryside, they are much more likely to be able to do it with little chance of someone being there. Also they are much more likely to be able to set up an ambush with multiple people and decent long guns where
V2X your data is the product (Score:3)
Can't they even get one expert to discuss privacy, monetization of your car driving habits and such in the article?
Where are the who, what, when and how questions even asked?
As in, who profits, how do they profit, when will they incrementally get to overreach and more control?
Is this media confirmation bias where softball questions and softball reporting for favored topics are accepted?
Re: (Score:2)
'You don't need V2X for that.
Just wait at a traffic light with a shotgun and rob people when they stop.'
No comparison.
That's S2V, shotgun to vehicle tech.
Re: (Score:2)
Standing at a traffic light with a shotgun sounds like the quickest way to get the police called.
Being able to stop any car, at any location, at will, would make it considerably easier not only to commit robbery, but also targeted attacks like murder and kidnapping, where hitting a random car won't do.
Re: Not useful tech. ha ha (Score:2)
"Standing at a traffic light with a shotgun sounds like the quickest way to get the police called", for what, standing, with a shotgun?
Re: (Score:2)
Yes. If you don't believe me, I encourage you to try it. You will swiftly learn how the world works.
Retro sales pitch. (Score:2)
Pile up, or force the car to stop and steal whatever the occupants have on them or kidnap them.
(Classic car dealer) ”Yeesss. That’s it. Give into it. See the evil. Feel the evil. Fear the evil. Know what’s good. Buy claaa-sic.”
Between this and new car prices, my great-grandkids will still be yelling “punch buggy!”
Re: (Score:3, Insightful)
I don't want a car that "phones home" to the mothership, or the govt...or other cars.
This is just TOO MUCH of an intrusion by the surveillance state.
Re: (Score:2)
Yeah, even though people keep forgetting, it's Security 101 - don't trust your inputs.
Re: (Score:2)
I didn't know we have so many murderers around. Have they been setting up turrets and shooting people on highways? Seems like that would be much easier. Unlike highway shooters, people broadcasting fake info will be caught since the transmission location can be extrapolated data recorded on multiple cars in the vicinity.
Re: (Score:2)
I didn't know we have so many murderers around. Have they been setting up turrets and shooting people on highways? Seems like that would be much easier.
It's like the problem of trans rapists in prisons. In fact there are (were) almost none and the people at most risk of assault are the trans people themselves. However, when you create a security hole that people can exploit and criminals learn about it, then the calculation changes because the criminals start working on how to do that and take advantage of the security hole and suddenly that category increases. If you don't protect against that then the media will blame you when there start to be problems
It's OK if the media doesn't blab? (Score:2)
So the situation in your example is not one from which inmates of women's prisons need protection as a general principle of just and humane treatment of persons who end up in prison? But it is only a problem when the rare cases stir up public controversy from accounts in the media?
One of the arguments against the death penalty is that innocent people get executed. Is the rebuttal to that argument that an innocent person being put to death is rare?
Re: (Score:2)
So the situation in your example is not one from which inmates of women's prisons need protection as a general principle of just and humane treatment of persons who end up in prison? But it is only a problem when the rare cases stir up public controversy from accounts in the media?
Prisons are dangerous places that need to be made safe. That applies even if there are just women together with women. That applies especially if there are penises about, which tends to make sexual assault more likely both in the case of men with men and some form of mixing. I don't see that the problem of true normal, non violent (no history of or reason to believe there is an inclination to), pre-operative trans women in female population which accepts then is a particularly big problem more than putting
Re: (Score:2)
So right now people can use a drone to drop a paint can on windshields, yet they don't do that. At least I haven't heard of it, so it must be very rare. Besides, I doubt cars can't be coaxed into accidents by V2X unless they foolishly make V2X authoritative over other sensors.
Re: (Score:2)
There may not be that many murderers around. There are plenty of social jerks and bored teenage boys who get enjoyment from causing mayhem and loss to others, and they will 100.1% absolutely exploit the inevitable security holes in vehicle-to-vehicle transmission to do that. It would be the public park toilet problem writ large: maybe only 0.0001% of the population gets their jollies from trashing public toilets yet every one that is built is generally trashed within a year, often in ingenious but offensiv
Re: (Score:2)
Stuff like this has been faked [digitaltrends.com] in the past. V2V needs to be secure in some way, and some bogus ID needs to be able to be pushed down on its importance. Otherwise, someone can just broadcast, "hard brake, hard brake, guy coming in from the right" to every vehicle, and now you have some smash-ups. Security through obscurity won't do it because this is going to be a windfall for saboteurs, especially if there is a disaster [1].
V2X is going to need to be worked at from the ground up, and that means the actua
Re: (Score:2)
'Otherwise, someone can just broadcast, "hard brake, hard brake, guy coming in from the right" to every vehicle,'
More like 'giant accident, deep new sinkhole' or whatever from people who want their peace again from traffic in their street.
Like they do it with Waze since forever.
Re: (Score:3)
V2X is going to need to be worked at from the ground up, and that means the actual hardware is going to have to have a physically tamper resistant component, and every segment needs authentication and the ability to tune a non-functional item out.
You mean like Blu-Ray keys in tamper-resistant chips, which have gotten cracked how many times? Remember that all you need is a single ECU from any car that supports this, and now you have something you can uncap with acid to read the codes out of the hardware. Rinse and repeat as needed. All it takes is one compromised key, and people can spoof it at a hundred locations across the country the next day. And once they figure out how to extract the keys, as soon as you ban one key, somebody will steal ano
Re: (Score:2)
You can't trust another vehicle.
When you see the vehicle in front's brake lights come on, do you assume they have been hacked and speed up? Or when the vehicle indicates, is that just the driver being malicious and trying to disrupt the flow of traffic?
Existing vehicle to vehicle communication is already well regulated and works fine. It's massively vulnerable, just re-wire it, but for some mysterious reason that's not a widespread problem. It would be helpful for the car behind to know when the one in front is about to brake hard, or int
Re: (Score:2)
Windows 11 (Score:2)
'nuff said.
Re: (Score:2)
10 year old cars still work perfectly well with modern OBD2 diagnostics. In fact cars from the 90s do.
Even decade old wifi devices work with modern wifi. CDs from the 1980s still play in modern BluRay drives.
The worst that will happen is a safety feature gets degraded. That's more likely to happen with speed limit warnings based on out-of-date maps.
Re: Not useful tech. (Score:2)
Re: (Score:2)
Of course, assuming that *among all possible contingencies* the only one that must occur, when you see the break lights turn on, is a network hack followed by the car speeding up: that's not smart. Your argum
Not worthy of driving on the road unsupervised (Score:2)
Many states have "slow down or move over laws" to protect police, fire and tow-truck drivers and their vehicles and the motorists stopped by the road they are either helping or scolding.
So you are on a long drive trying to stay alert, you crest a hill and come upon flashing emergency vehicle lights on the side of the road, you check your mirrors and signal to move over. The motorist in the left-lane decides to "close the gap" to not allow your lane change, and that vehicle is followed by a long line of
Re: (Score:2)
When you see the vehicle in front's brake lights come on, do you assume they have been hacked and speed up?
Those brake lights tell me that they are in the process of doing some amount of braking from 0-100%. I don't speed up, because the assumption that the brake light switch has failed on is the least plausible. But by the same token the messages from V2V can only be used in an advisory capacity, because the most important thing is reacting appropriately to what the vehicle is actually doing regardless of what the lights are doing. Two different people signaled a left merge and then exited the highway to the ri
Re: (Score:2)
And still enormously useful.
Airplanes have their own systems, called "VHF radio" and "ADS-B." They have also not been "massively hacked" and both have proven to be quite useful.
Re: (Score:3)
You can't trust another vehicle.
When you see the vehicle in front's brake lights come on, do you assume they have been hacked and speed up?
No, of course not. But the only thing it does is alert me to the fact that they *might* slow down. Reacting much at all is usually a bad idea unless you're tailgating. You should react when you see the car start to actually slow down. It's useful as a signal to humans, but self-driving tech will ignore that signal, in all likelihood, because it has a better source of truth — the actual speed of the car in front of them, as observed using radar, lidar, or camera parallax.
Turn signals, of course, sh
Re: (Score:2)
Any vehicle to vehicle system is inherently flawed by design. You can't trust another vehicle. As soon as you do anything more than slightly increasing following distance in response to external data that could be forged, somebody will start broadcasting fake data and cause a pile-up.
The thing, which makes "trust data from an outside source" so dangerous is, that this "outside source" could be anywhere on this planet, and would not have to follow any rules. There are enough jurisdictions, which are either uncooperative or even openly hostile.
Car traffic, on the other side, is extremely local. As long as you can ensure 100%, that any computer interacting with your vehicle is nearby and well identified, this sort of car to car interaction could become quite reliable and dependable. Anyone
Re: (Score:2)
There are enough jurisdictions, which are either uncooperative or even openly hostile.
Can we please just leave Texas alone?!
Re: (Score:3)
When I looked into the tech, the discussion was between car-to-roadside unicasting and car-to-car multicasting.
You're right, there's no way to make this secure for vehicular data. Even if each car had a signed digital cert, it would only be a matter of time before thieves stole certs from cars and thus be able to fake transmissions that appeared legit.
But, already, many cars have links to online services that have access to doors, the engine, etc. Sooner or later, cyber criminals will have the means to inst
Re: (Score:2)
'You're right, there's no way to make this secure for vehicular data'
It already happens right now.
Car owner types accident into Waze.
Waze tells other Waze-users.
Cars get routed around.
Re: (Score:2)
That people send data doesn't mean it's secure, and a vehicle-to-vehicle adhoc mesh is very different from a car sending to the nearest cell tower.
Re: (Score:2)
You don't even have to have malevolence involved:
You'd want to be informed if there's a blockage: say, a car stopped, around a blind corner. Hey, V2X can do that for you! So, you start to rely on this ability. ... which works until the car blocking the road does not have that ability enabled (currently or at all). You know, say due to a collision.
And then you blithely drive around the curve at full speed, assured in your safety, and have time for just that one panicked scream...
Only signed messages (Score:2)
The vehicle to vehicle communications need to have a cryptographic signature, one that is hard to get, and registered to a car. That should be good enough to deter pranks; but there's still the possibility of a group using stolen keys for a widescale traffic catastrophe. Not sure it's worth it. I imagine the main motivation is so the car always blabs to the mothership.
Re: (Score:2)
The vehicle to vehicle communications need to have a cryptographic signature, one that is hard to get, and registered to a car. That should be good enough to deter pranks; but there's still the possibility of a group using stolen keys for a widescale traffic catastrophe.
The whole "registered to a car" thing goes out the window as soon as a car has to be repaired or gets sold for scrap, because those parts are going to get sold for reuse in some other car.
Remember how long it took folks to break Blu-Ray crypto? It took only ten months to crack a system specifically designed to prevent people from being able to get to the keys, created by people with many years of experience doing exactly that sort of work. Now think about the quality of software that the automobile indust
Re: (Score:2)
Well, taking ten months to crack your car's key should be enough to deter pranks, especially since using it would lead straight back to you, and then the key would be banned and you'd need a new one after your jail sentence.
Anonymity feels like the issue here... (Score:2)
If there were reputations in place, and it wasn't easy to just throw one identity away for another... Wouldn't that change the dynamics? Ignore any trolls, and give reputation for any claims you can confirm (even later).
Sure there needs to be protections in place to make any punishment fit the action/result. We don't want a Black Mirror episode here where annoying enough people means you can't drive to your job, which gets you fired, which means you're kicked out of your housing immediately, ... etc
I don
Re: (Score:2)
If there were reputations in place, and it wasn't easy to just throw one identity away for another... Wouldn't that change the dynamics?
It would make it borderline useless, because you're unlikely to keep encountering the same car over and over again. Unless you mean that the centralized authority would not trust data until it has seen data from that car previously, in which case that would work right up until the part gets sold at auction and the key extracted for mayhem, at which point the key is trusted, and it can lie for quite a while before it gets caught. Then throw away that stolen key and use a different stolen key. It isn't rea
Re: (Score:2)
How is that more disruptive than caltrops?
If it's not, why would it be more prevalent?
Re: (Score:2)
How is that more disruptive than caltrops?
If it's not, why would it be more prevalent?
Tire puncturing hardware will be easily seen and cleaned up, and if it keeps happening in a particular spot, somebody will stick a camera there to find out who is doing it. A radio transmitter won't necessarily be easily seen and removed, and can wait arbitrary amounts of time before re-triggering, making it hard to track the actual source using radio triangulation.
Re: (Score:2)
Any vehicle to vehicle system is inherently flawed by design. You can't trust another vehicle. As soon as you do anything more than slightly increasing following distance in response to external data that could be forged, somebody will start broadcasting fake data and cause a pile-up.
Even with massive server infrastructure and centralized algorithmic processing of the data, even companies like Apple and Google can sometimes be fooled by a wagon full of cell phones [seattletimes.com]. Your car has a tiny fraction of the data that those sorts of companies do.
This, I've always said that vehicle to vehicle communications were never going to work due to the speed of authentication, A.K.A. the reason why I cant tell your car I'm an ambulance and it needs to get out of my way. Even with a massive back end in the same city, the latency on the signal is going to take far longer than the time the vehicle has to react in.
Same with fully autonomous cars, another pipe dream as you're never going to get a system small enough to perform real time path tracing of multiple
Good and bad (Score:3)
Good:
The ability for cars to communicate with the roadway and each other enables improved safety features. Know when a car is coming, or there's traffic, or alert about an issue. Follow paths in roadways to implement driverless cars without the uncertainty of a computer being able to read signs and lane markings. Alert cars when vehicle speed limits have changed, potentially even slowing cars down in an emergency. Implement alerts or even geo-fencing to protect emergency workers, cyclists and pedestrians on shoulders and in roadways. Alert of obstacles in roadway. Alert a vehicle of an amber or silver alert, or a weather emergency. Notify other cars where your car is in traffic and vice versa so they can avoid each other. Alert the vehicle or even stop it at red lights (running through red lights and stop lights is responsible for a ton of accidents).
Bad:
Unreliable and faulty sensors. Hackers. Software glitches. Targeted advertising, and thus privacy issues. Incompatibilities. Lack of adoption. Shitty implementation.
Ultimately there's a lot more good than bad. Cars already have enough advanced technology that a sufficiently advanced attacker could already wreak havoc on millions of vehicles, but that's never happened yet. So we should be cautious, but there seems to be a lot of potential for good.
Re: (Score:3)
Cars already have enough advanced technology that a sufficiently advanced attacker could already wreak havoc on millions of vehicles, but that's never happened yet.
Many, many Kia and Hyundai owners would beg to differ.
Yes that was car theft instead of forced wrecks, but anyone who reads the news knows there are enough crap humans out there that it's obviously just a matter of time. Ransomware gangs shut down emergency rooms. Guys swat other guys because they lost an online game.
Re: (Score:3)
When it works. Is it really helping though? Animals, kids, sports balls etc can just all of the sudden be in the roadway, and not everything has a warning beacon that its approaching. The only real option when driving is to be paying attention at all times. I drove my parents newish car the other day, it has a lot going on to distract the driver. Things that used to be push buttons now you have to navigate through menus. There isn't just a speed display but mpg, how eco friendly
Re: (Score:3)
Animals, kids, sports balls etc can just all of the sudden be in the roadway,
This was my immediate thought. I go out on my mountain bike. Whilst off road I break the transmitter and then, returning home I have nothing. Does that mean that I'm more at risk because the cars are assuming they can detect me?
There can be real uses, for example you could build car trains on motorways (highways) where a series of cars drive bumper to bumper, saving fuel and road space. In the states, you'd let them use the car share lanes. In that case the vehicles are each agreeing to join the train and s
Re: (Score:2)
There can be real uses, for example you could build car trains on motorways (highways) where a series of cars drive bumper to bumper, saving fuel and road space.
What could possibly go wrong!
Re: (Score:2)
Actually, probably less than you think if you are a bit careful. If cars are close enough together then they can't get a big speed difference if they hit each other. If you have a road specifically designed for it (e.g. complete live monitoring of the section where trains are allowed, ban cars without the communications from the car share lane, only allow certified encrypted communications from known customers) then you should be able to ensure that any problem is detected and rectified.
Re: (Score:2)
Advanced safety systems today are silent/invisible until they need to alert you of a problem, so no, I don't think they could cause distraction at all. If anything they would notice when you were distracted and tell you to keep your eyes on the road. My nav system already refuses to let me scroll the interface for more than a few seconds at a time, to cut down on distraction.
And the real purpose (Score:2)
is to facilitate automation so as to eliminate human drivers entirely.
Re: (Score:2, Interesting)
Cars already have enough advanced technology that a sufficiently advanced attacker could already wreak havoc on millions of vehicles, but that's never happened yet.
Never happened yet? Says YOU. Clearly you’re not counting planned obsolescence, factory warranty fine print, forced arbitration, and emissions “limiting” bullshit that more limits the life of the engine than anything. Cars have “advanced” technology alright. They can all go at least 10,000 miles between oil changes according to the maker, and come with convenient little apps to remind you of all those purposely harmful service intervals right “on” time. Meanwhi
Sharing location (Score:2)
"A powerful tool for achieving this ambitious, long-term goal is vehicle-to-everything (V2X) technology, which enables vehicles to communicate with each other, with other road users such as pedestrians, cyclists, individuals with disabilities, and other vulnerable road users, and with roadside infrastructure, through wirelessly exchanged messages. "
And I can totally understand how that would be good and useful. Your vehicle would potentially be able to help sense your vehicular environment and make you safe
Re: (Score:2)
So does this work both ways? ie my car can co-ordinate the movement of a cyclist so they don't make an abrupt turn in front of me? Or move them onto their bike lane? If not, this sounds like a cyclist's wet dream of being able to control a car's movement. They get in the middle of a single lane of traffic and let everyone else deal with it. Who controls my car? It doesn't sound like me. But it sounds like I'm still res
Re: (Score:2)
I think you're looking at it from the wrong perspective. If my vehicle has determined that there aren't any other vehicles between it and my exit 100km ahead, it should increase its speed to its maximum safe speed for this road. In other words, no more speed limits for me on this road until I reach my exit. If, however, a vehicle enters an on-ramp ahead of me, my vehicle can be instructed to move to the fast lane until it has passed the entering vehicle.
I light traffic conditions, vehicles can be moved f
Re: (Score:2)
Re: (Score:2)
I am wondering how this could work too, but in the best case it would make you safer. I see people in their cars yakking on their cellphones all the time and driving badly, can this make them less dangerous to me? Something that would keep some dimwit from T-boning me in an intersection sounds interesting.
Re: (Score:3)
It’s interesting, we have it on boats and aircraft for quite a while now. AIS on boats, and ADS-B on aircraft.
I’m mostly familiar with AIS and the marine world. In that environment, commecial vessels over 100 Gross Tons or any passenger carrying vessel are required to have an AIS transponder. It relays various information, including the vessel’s speed, position, name, MMSI (Maritime Mobile Service Identity) and various other bits of data.
It’s not mandatory for recreational boats, but
Re: (Score:2)
One difference in aviation though ADS-B is information only; there is no element of control. I suspect the same is true for shipping based on your description.
For some reason many people think that adding in remote control channels to automotive is a reasonable thing, when its potential downsides far outweigh the risks it's trying to mitigate.
I wouldn't mind something like ADS-B out and TCAS to be honest, but I would be very hesitant to be in a vehicle that would do more than that.
Also just fun thought pro
Re: (Score:2)
Some Traffic Collision Avoidance Systems (TCAS) use ADS-B inputs (among others) and some offer automated collision avoidance in the autopilot. There may be marine autopilots that do the same thing with AIS. Both can provide aural warnings when they detect a conflict.
The Slashdot catastrophists [youtube.com] have jumped directly to the "carjacker ordering your car to stop" scenario but there are a whole bunch of useful features short of compulsory control overrides. All the smartphone mapping apps already offer some of th
Easily prevent 20% of deaths (Score:2)
The following active safety features can save about 10,000 US lives (out of 40,000 killed in traffic accidents): .. the system doesn't have to be 100% good at it because it can just be a backup to the driver .. therefore an accident would require an inattentive driver PLUS a failure of the system. If the system (
1. Have a car that can in 99% of cases recognizes when it's at a stop sign or red light and slam the brakes if the driver didn't react. (Use image recognition and GPS/map) and apply the brakes. Note
Re: (Score:3)
Until some jerk has a flip down STOP sign in a rear window while on the highway for teh JOKES
Re: (Score:2)
How would that cause a guaranteed accident? ,.. the one-second warning would occur before braking .. also .. the jerk would be caught on video and will go straight to jail for murder/attempted murder. If someone wants to kill people on highways there's other ways .. such as tossing a brick or shooting.
Re: (Score:3)
The AI can be taught to ignore those, so statistically it would have to occur very often for it to trigger. The first time someone does that., they'll be on camera and charged with attempted murder.
Re: (Score:2)
There's a story of a cyclist having a backpack that inadvertently resembled a yield sign close enough for smart cars to detect it as such. The car crept along with the cyclist...
On a slightly related note there is a video of Tesla's FSD giving up when encountering a horse carriage.
Would V2X be mandatory for the Amish?
Re: (Score:2)
Have a car that can in 99% of cases recognizes when it's at a stop sign or red light
Sounds like we'd be teaching cars to solve CAPTCHAs! I certainly have trouble with those...
Underestimating potential (Score:2)
Re: (Score:3)
I did, when I was 14 and was discovering computer networks. Then I grew up...
Re: (Score:2)
Re: (Score:2)
All those you mentioned are "simple" product to be used at a single point of use, by one user at a time. I'm still dreaming of semantic search on personal devices, among other things.
Cars talking to each other is an entirely different category: network connected devices, where the data they transmit is critical. Either this will get regulated to the point you won't be allowed to change a light bulb in your car by yourself, or it won't happen.
When you hear unexpected sounds in your living room at night, you
Not needed (Score:3)
A lot of Slashdot is focusing on the wrong issues. Sensors, hacking, etc, it's all not relevant. The reason this system isn't needed is because we already have it. Communication occurs via lights from one car to another. Additionally we can already judge rate of change between vehicles using sensors.
The weak part for that is the human. We're a poor judge of rate, and we have both lapses in attention and a relatively slow reflex. Computers can virtually instantly identify tail lights going on, they can virtually instantly tell a rate of change ... change in distance between you and the car in front. We already see that in Automatic Emergency Breaking systems which engage so last minute that it's a miracle they don't hit the car in front at all. We already see that in cross motion detection (that loud beep when you put your car in reverse but the car has seen the pedestrian or cyclist but you haven't.
These existing systems already address the big problem with visual communication: the human. And when we're already at a point where without wireless signals my car can come to a perfect stop without hitting the one in front what's the point of throwing another system in.
Re: (Score:2)
Re: (Score:2)
Except the issue with stoplights is there is more than cars on the roads, specifically... humans. Even if all cars spoke to each other, and spoke to busses and trams too, will they also be talking to bicycles, scooters, or people walking around the city? Don't forget the ideal method of transport is not to route car traffic, it's to route and maximise people movement.
We may see what you want in our lifetimes, but it will likely be limited to major arterial expressways, possibly in dedicated smart car lanes.
Re: (Score:2)
Sensors, hacking, etc, it's all not relevant. The reason this system isn't needed is because we already have it. Communication occurs via lights from one car to another. Additionally we can already judge rate of change between vehicles using sensors.
One, two, skip a few, ninety-nine, one hundred.
Getting information from one host to another is only one step of the process. Changing the speed of the vehicle based on that input is another. You've skipped some of the steps in-between, because it's inconvenient for your next statement, but those steps are hard requirements no matter how inconvenient they are for your argument.
The weak part for that is the human.
Funny, the weak part of your argument is the same: The human who wrote it thinks that everyone else is unable to function at thei
Re: (Score:2)
You've skipped some of the steps in-between, because it's inconvenient for your next statement, but those steps are hard requirements no matter how inconvenient they are for your argument.
I didn't skip steps, in fact I gave examples of how those steps are already taken care of.
The human who wrote it thinks that everyone else is unable to function at their level, and believes that all of the technology that they'll use to "compensate" for that "inability" is completely able to trust it's inputs to the extent that their owner's "inability" doesn't matter.
Err no, cars have virtually no trust in their inputs. You seem to really lack any insight into the mindboggling amount of diagnostics which goes into vehicle sensors these days to internally validate results. And even then, these are assistive devices only, just like the smart mesh for cars would have been.
Let's wait on V2X. (Score:2)
Maybe start with Vehicles to Instagram (V2I) and see how that works out. :-)
Re: (Score:2)
I'm holding out for V2.1X, those .0 versions are always buggy!
Wouldn't be used to sell private data....NOT ! (Score:2)
Hacking (Score:2)
From a safety perspective, this sounds like a great idea in the era of smartphone junkies distracted behind the wheel adding to the drunk and drugged driving problem..
..until you realize we also live in a world in which anything and everything that turns “smart”, gets hacked for the dumbest fucking reasons.
Keep that fucking shit far away from me, my cars, and my family.
Re: (Score:2)
New York cars say (Score:2)
"Let's throw more tech at it!" (Score:2)
Tech is only part of the problem (Score:2)
Law would need to change as well
Speed limits are a good example, but there are others. Speed limits are way too low, and the vast majority of safe drivers routinely exceed them. If every tiny violation of the thousands of driving laws was immediately reported to police, the result would be chaos
We need to change the law to punish genuinely unsafe drivers, not harvest revenue from safe drivers who disobey silly rules
Comment removed (Score:3)
How will they do that? (Score:2)
exchange location information with each other, and potentially cyclists and pedestrians
We can't even get pedestrians to respond correctly to the Big Red Hand [istockphoto.com]. What makes people think that a more technologically complex protocol will help?
Some of the detection systems do work. But they are self contained and do not depend on intelligence of competing roadway users. Which for some cyclists is a good assumption.
All of this is fascist/terrorist enabling (Score:2)
This, and self-driving. Musk is on record as wanting to ultimately make manual driving illegal. Any system sufficient to safe cars to the extent they want is sufficient to imprison us. If it doesn't imprison us, it will certainly become a choice target for terrorists who might conceivably shut down transit, perhaps even nationwide for days.
Negative (Score:1)
tweak to spec that addresses some of the concerns? (Score:1)
A 1 dimensional system solves a lot of problems that the proposed 3 dimensional system has: Instead of sending the exact location of each node, it broadcasts a much simpler on-demand beacon with distance determined by multi-path resistant RSSI. 2.4ghz would be used. beacon and beacon/inquisitor modes. 8 node id types (car, bike, pedestrian, hazard, emergency vehicle emergency/non(even more important because people forget to set emergency mode but nearby car AI sees it moving fast and concludes it is an emer
Anyone actually know this technology yet? (Score:2)
How useful is it before 'everyone' has it?
Like you have one car and a sensor or two on an important (often busy) path. Can that be useful still? If not, what level of saturation is needed to extract effective traffic info?
How function is it when our location data often is bad? Or is this going to be better than GPS on my phone (if yes, why)?
Will the need to constantly transmit my location and speed, even as a pedestrian, impact portable users? I know cell phones try to limit transmissions to save batter
Boats already do this (Score:1)
AM Radio? (Score:1)
To what extent do these things transmit data? I don't want my speed broadcasted a mile ahead of me for the next officer to know I'm coming, but I can see how a signal saying there is a car coming could be useful. Then there's the question, can you infer speed from signal strength? because then that does the same thing.
The general idea for using AM radio frequencies though is that they are low-frequency, high ra