Making Driverless Cars Safer 140
colinneagle writes "Several autonomous cars have been developed elsewhere, most famously by Google, and they are generally capable of identifying objects in the road directly ahead of or behind them. The challenge undertaken by MIT researchers is making these cars aware of dangers lurking around corners and behind buildings. MIT PhD student Swarun Kumar showed a video of a test run by the MIT researchers in which an autonomous golf cart running the technology, called CarSpeak (PDF), encountered a pedestrian walking from the entrance of a building to a crosswalk. The golf cart stopped roughly five yards ahead of the crosswalk and waited long enough for the pedestrian to walk to the other side of the road. The vehicle then continued driving automatically. The solution Kumar presented is based on a method of communications that is intended to expand the vehicle's field of view. This can be accomplished by compressing and sharing the data that autonomous vehicles generate while they're in motion, which Kumar says can amount to gigabits per second. In a comparison test, a car using CarSpeak's MAC-based communications was able to stop with a maximum average delay of 0.45 seconds, compared to the minimum average delay time of 2.14 seconds for a car running 802.11, the report noted."
Does not 802.11 a (wireless) Ethernet... (Score:2)
... thus the notion of MAC still applies?
Just curious...
Paul B.
Re: (Score:3)
Does not 802.11 a (wireless) Ethernet... ... thus the notion of MAC still applies?
Uh... what?
Re: (Score:2)
Well, first word in Subject should have been "Is"... Sorry, edited from something else... ;(
MAC, as in, "Media Access Control" address, or Ethernet address, which every Ethernet card, wired or wireless, has. To quote Wikipedia on IEEE 802.11, "Current 802.11 standards define "frame" types for use in transmission of data as well as management and control of wireless links. Frames are divided into very specific and standardized sections. Each frame consists of a MAC header...".
So, I was totally confused by th
Re: (Score:3)
I checked TFA, and found this:
As noted in this report on the project, standard 802.11 networks cannot accommodate the data transmission needs for communication between autonomous vehicles because they generate more data than the available bandwidth can handle. CarSpeak instead uses a content-centric MAC protocol for transmitting data, in which data pertaining to specifically requested roads and regions contends for space in the medium, as opposed to the cars sending requests for information. This ensures the network only displays relevant data, avoiding a flood of data pertaining to open roads.
So, yes, apparently they're talking about low-level wireless networking protocols, but... it's like saying that your revolutionary new Web search engine is "copper-based." I mean, that's what the conductors on the server CPU are made of, and without copper none of it would work, but it hardly captures what's unique or noteworthy about the technology.
Re:Does not 802.11 a (wireless) Ethernet... (Score:4, Informative)
From the OP: "a car using CarSpeak's MAC-based communications was able to stop with a maximum average delay of 0.45 seconds"
This acronym 'MAC' is not used or explained anywhere else in TFS, so it's unclear whether they mean Media Access Control from the IEEE 802 spec (which probably is employed in moving data wirelessly from car to car, but has little to do with the specific problem of detecting or responding to safety hazards) or something else entirely.
Re:Does not 802.11 a (wireless) Ethernet... (Score:5, Informative)
From the OP: "a car using CarSpeak's MAC-based communications was able to stop with a maximum average delay of 0.45 seconds"
This acronym 'MAC' is not used or explained anywhere else in TFS, so it's unclear whether they mean Media Access Control from the IEEE 802 spec (which probably is employed in moving data wirelessly from car to car, but has little to do with the specific problem of detecting or responding to safety hazards) or something else entirely.
They explain MAC right in the paper ( which is linked in the article ). It's MAC just like you think MAC is (Media Access Control).
Really, the gist of the paper, is instead of each car being the source or identity of a packet, via normal MAC 'addressing' and trying to communicate some important information ( like soft squish target...er.. human at X,Y,Z moving Z-> Y-> Z-> at such and such a rate ) via the full OSI model ( like packaging that info in UDP or TCP), You instead break down the 3D space around the car ( and other cars do the same thing ) using an octree graph ( just like visibility systems in 3D game engines), and send out this info with the MAC layer altered to show which region of the octree your information is pertinent too.
So if you are Car A, and Car B,C,D,E,F,G,H,I are all in your broadcast range, and they are dumping out gigabytes of network aware info based on their laser scanners, you can quickly and at a very low level (hardware) pick out the packets that are important to you (from the air).
tl;dr:
It's a clustered index for wireless packets based on GPS location of events stuffed into the MAC (data link) layer. It's a complicated QOS scheme that has been crafted around a specific engineering latency problem.
Re: (Score:2)
Interesting! So it uses spatial addressing rather than hardware addressing! That is actually a pretty clever approach, come to think of it. Thank you for breaking this down for me.
Re: (Score:2)
I think the fundamental problem here is that if you DEPEND on other cars to detect collisions, you have a HUGE problem for all cars if one car has faulty detection.
Also, depending on other cars -- does that mean if you are the only Autonomous car on the road, you don't have the processing power?
What needs to happen is that dedicated GPUs could be programmed to merely deal with what objects are in motion towards the car's destination, and then they merely push size, vector and speed data to the main unit --
Re: (Score:2)
I think the fundamental problem here is that if you DEPEND on other cars to detect collisions, you have a HUGE problem for all cars if one car has faulty detection.
What's the problem if one car has faulty detection? If you are in a car, and a kid unexpectedly runs out between two parked cars and in front of you, what is the negative effect if one of those two cars was "linked" but not detecting? The fault did not reduce safety at all. The lack of a detector vs a detector with faulty detection seems to have no effect. So I ask again, how does that cause a HUGE problem?
Re: (Score:2)
Create small solar powered 360 degree, 3D imaging devices at street corners, alley ways, sidewalks and store fronts. Have the car road state computer be a node of a network that includes all the other cars in, let's say a 2 mile radius and the imaging devices. Moving bodies are models for position and velocity. Your cloud map of the road includes the information about the 18 years old going 75 mph on a surface street, cutting in and out of traffic and will almost certain run the up coming red light. With su
Re: (Score:2)
The "solution" should be a complete re-write of the idea of information transmitted. The problem is that the solution provided here would *only* work if everyone played nice. Know those mirrors some peo
Weather Conditions (Score:5, Interesting)
How do these things perform in weather? ex. Blizzards
I'd hate to wind up in a snow drift in the middle of the road rather than backing up and finding an alt. route... or going home.
Re: (Score:1)
Re:Weather Conditions (Score:5, Informative)
How do these things perform in weather? ex. Blizzards
The same way cars driven by people do: They get stuck in snowdrifts several feet wide and thick. And that's before you back out of the driveway. Disclaimer: I'm from Minnesota. Autonomous vehicles can't unbury your car, and any visual sensor would be as blind as you are in a blizzard. For that matter, even radar operating at microwave frequencies would be... snow is made of water, and water attenuates it. That's why you're supposed to stay inside during a blizzard... It's suicidally stupid to try driving in conditions where, should your vehicle become disabled, not only are you at risk yourself, but others have to risk themselves to come rescue your sorry, impatient ass. And incase you're wondering, no -- your cell phone doesn't work very well in a blizzard and GPS is straight out too, so if you don't know exactly where you are, emergency workers may not find you even with E911 capability; It's only accurate to within 50 meters. In a blizzard... you have trouble even seeing a couple meters in front of you.
Take it from someone who lives and breathes the fluffy white death from above -- Never, ever, trust a vehicle with your life. Any vehicle, even ones connected to Skynet with an IQ of a billion and a hundred different types of sensors. If you can't walk 10 miles in the weather, don't go out in it.
sensors on the car can become dirty (Score:2)
sensors on the car can become dirty and then what do you hot shot??
remember coming to a dead stop can be unsafe or it can block up traffic.
Go to a fail safe speed?? can be very unsafe on some high speed roads and even a slow speed can do some big damage if it hits something.
Re: (Score:2)
I'm sure there is a self diagnostic and calibration that can sense when a sensor is dirty and prevent the car from operating. I also don't imagine that the car will be programmed to stop any more suddenly than is safe for all concerned. But, what's the alternative - plow into the pedestrian? Any of us would slam on our brakes anyway and stop as rapidly as we can. The thing about autonomous cars is that they will see the pedestrian and implement braking the microsecond he becomes visible - unlike human opera
Re: (Score:2)
Re: (Score:2)
The same way cars driven by people do: They get stuck in snowdrifts several feet wide and thick. And that's before you back out of the driveway. Disclaimer: I'm from Minnesota. Autonomous vehicles can't unbury your car, and any visual sensor would be as blind as you are in a blizzard. For that matter, even radar operating at microwave frequencies would be... snow is made of water, and water attenuates it. That's why you're supposed to stay inside during a blizzard... It's suicidally stupid to try driving in conditions where, should your vehicle become disabled, not only are you at risk yourself, but others have to risk themselves to come rescue your sorry, impatient ass. And incase you're wondering, no -- your cell phone doesn't work very well in a blizzard and GPS is straight out too, so if you don't know exactly where you are, emergency workers may not find you even with E911 capability; It's only accurate to within 50 meters. In a blizzard... you have trouble even seeing a couple meters in front of you.
In other countries with severe winter weather, people drive in convoys during blizzards. Only the last car in the convoy is really vulnerable, and only until it isn't counted at the end, or until the second-last car can signal.
Drivers also make sure they have their cars equipped for winter, including studded tires, chains, spades, ballast, blankets and everything else that goes with winter, and they won't even get a permanent driver's license without slick training.
Never mind that what Americans in general
Re: (Score:2)
Excuse me, but when I use the term blizzard, I mean it. We get 1 or 2 a year up here, and the "persistent strong winds" can exceed 50 MPH up here, creating white-out conditions where visibility barely exceeds the reach of your arm. We host one of the largest international airports in the country - MSP, and the only birds we put in the air during then are military during those times, to support search and rescue teams.
So please, don't get condescending -- when I say blizzard, I mean it. And our schools don'
Re: (Score:2)
How do these things perform in weather? ex. Blizzards
I'd hate to wind up in a snow drift in the middle of the road rather than backing up and finding an alt. route... or going home.
Golf Carts have never traditionally done well in snow blizzards.
Re: (Score:2)
I wonder if it might help to record video... (Score:4, Insightful)
Even with faster stopping, there will be those who deliberately jump in front of cars in order to get hit, hopefully to score a big jury verdict.
The solution -- a camera that turns on and records encounters with pedestrians, bicyclists, etc, with a timer in place. That way, if there is a wreck, there is documented proof that the other party jaywalked or violated traffic laws.
Of course, if it is the car's fault, it will be documented as well, but assuming a fully automatic vehicle which obeys all traffic signals, it likely won't be the vehicle that caused the collision.
Re: (Score:2)
Privacy concerns aside.
Re: (Score:3)
What privacy concerns are there about recording a public street?
Re: (Score:2)
Well for one thing there's the security of the device doing the recording - if it's remotely exploitable then that data is practically exposed to everyone at all times, not just to law enforcement when a dispute needs resolving.
For another thing, existing in a public place != consent to being recorded.
Re: (Score:3)
I can understand that you can be in a public place but still be unhappy if someone videotapes you. An otherwise empty nude beach, with no one in sight for miles, is one such example. There is no public good that would result from taping you on such a beach :-)
However there is a clear public good that results from videotaping all incidents on roads. You don't have to record everything that ever happens in 360 degrees around your car. It's enough to have a ring buffer for 20-30 minutes. If an accident happ
Re: (Score:2)
If you could trust the police to only videotape incidents, that would be OK.
I don't particularly welcome systems that let police economically map where people not suspected of a crime are going. That's outside of their mandate. One thing I'm sure of is that it will be abused. If we're creating more criminals, let's fix that problem, not just make it easier to track them down.
Re: (Score:2)
If we're creating more criminals, let's fix that problem, not just make it easier to track them down.
It is not politically correct to "fix that problem". Only a small percentage of criminals get caught - only when they do a serious crime (like murder) and when they get unlucky. A typical vandal will not even be stopped; locals will be afraid to step in (see George Zimmerman) and police cannot be bothered. Besides, LEO response time would be 10 minutes at best - and how long does it take to tag a storefro
Re: (Score:2)
Today crime prevention does not exist; this means that at least one innocent must become a sacrificial bait for detection of a criminal; often that is needed more than once, and one criminal can destroy many lives over his "career."
Returning to my point, that's why it's so important to have a system that works for "everyone"; to create less of the kind of crime that has actual victims.
With respect to your signature (Score:1)
Re: (Score:2)
Maybe Dr. Seuss was sicker than I previously thought?
Re:I wonder if it might help to record video... (Score:5, Informative)
Re: (Score:3)
That kind of thing has been around for a while for your non-driverless car. You can get cameras that continuously record, only saving the last few minutes if you hit a button (or, with some systems I expect, a pedestrian).
This is off topic, but what are some good brands of these consumer dash cams? I've seen lots of $30 models for sale (which I'm assuming are crap), and a few $500+ models with GPS tracking and optional "phone home" tracking that seem to be targeted toward trucks and commercial vehicles.
What is a good, relatively inexpensive, "set it and forget it" dash cam that I can mount in my car and have it keep a continuous video log of my last few hours of driving?
Re: (Score:2)
What is a good, relatively inexpensive, "set it and forget it" dash cam that I can mount in my car and have it keep a continuous video log of my last few hours of driving?
I don't have an answer for you but one way to find out is to browse reviews on dealextreme. The main reason why I keep using them (and correspondingly end up sounding like an ad) is that their fora are of high quality. If you post "don't buy this, buy sku x instead) they leave the post there even though it means they wind up sitting on more stock.
Re: (Score:2)
Re: (Score:2)
In Russia insurance companies give you deductions if you have such. The camera is always on, and I suppose it overwrites old stuff as it goes. I do not knowhow long the buffer is.
Search from Youtube (e.g. russia car insurance scam).
Not safer, just faster (Score:2)
So in other words, instead of slowing to what would be a reasonable and appropriate speed [ca.gov], the cars are able to maintain high speeds without sacrificing safety by informing each other of hidden hazards.
It's an interesting solution, but it could actually sacrifi
Re:Not safer, just faster (Score:4, Funny)
Re: (Score:2)
It's a fact that slowing down will always be correlated to safety. If cars creeped around at 1 MPH, almost no one would ever die. This technology will improve matters, no matter what is deemed "reasonable" or "appropriate".
Re: (Score:2)
So in my words
FTFY. You've inferred that this must all be about increasing vehicle velocity, but I don't see that implied anywhere.
Re: (Score:2)
elsewhere (Score:2)
"Several autonomous cars have been developed elsewhere"
Elsewhere? Elsewhere from where?
you know what? (Score:3, Insightful)
Do not want. It's obvious at this point that the real deal with all these innovations is to retain more and more control over what people do and where they go. They entice us with convenience as they remove the control. I realize this article is about technical minutiae, but I have no desire to help this project along.. Until society matures such that those in charge don't have insatiable desires to micromanage individual choice as much as possible, I'd rather deal with driving my own vehicles around, thanks. Besides, with the right fit, driving a car is enjoyable.
Re: (Score:2)
I can foresee absolutely no such problem whatsoever in getting into my autocar and instructing it to take me to the nearest crack den.
Re: (Score:3)
the problem with these analogies is that they don't address the issue. Do you lock your doors even though it is illegal to trespass/steal? do you out of your way to avoid police even when you are fairly sure (as sure as our fucked up legal code allows us to be) you haven't done anything? Do you want to test it and find out? No? Why not? There's a dirty line that only gets crossed when the power balance is severely lopsided. Sure, in theory, you could hand over all control to the state, live a legally (
Re: (Score:2)
well yeah, a government that punishes power seeking and respects civil liberties IS a seige..
Re:you know what? (Score:5, Insightful)
I have the opposite reaction. I think we are entirely too cavalier about the unbelievable human toll that our current reliance on human-guided cars takes. Tens of thousands die in the US every single year. Look at the way the country responds to something like war casualties at 1/10 the scale and ask if this situation makes any sense.
Re: (Score:2)
I can't comment on your statistics, but generally, when 'studies' are published, they are insanely biased to push particular political agendas, so I take them with huge grains of salt.
Whether you trust the drivers of today, or the cut-rate programmers who direct the auto-cars of tomorrow, you still have to assume failure. Centralized systems sound great except when they fail, because when they do, it's catastrophic. A computer will happily cause the deaths of 1000 drivers if it misreads a situation since,
Re: (Score:2)
I can't comment on your statistics, but generally, when 'studies' are published, they are insanely biased to push particular political agendas, so I take them with huge grains of salt.
I didn't quote any statistics? The only vague number I mentioned was automobile deaths, which aren't particularly controversial - you just count dead bodies. This Wikipedia article has them by year - in 2010 it was 32,885 deaths. [wikipedia.org]
Centralized systems
We aren't talking about a centralized system - this article is about a distributed system.
We can't even automate our trains completely without catastrophic failures, so as far as safety goes, I still trust my situational awareness over that of a computer.
Trains are the safest transportation system we have today, so I'm not sure why you would use those as an example. In any case, almost all train accidents have been human in origin. Which makes
Re: (Score:2)
What I meant by risk is that any form of travel comes with it. I do not want the extra bit of theoretical safety if it means I have to give up gobs of freedom and control over my equipment to have it. Our system is fairly safe today, and a good balance between safety and freedom of movement. It could be better if it was based on reality and not the coffers of insurance companies, law enforcement budgets, and institutionalized control freakery.
1. Todays cars are whisper quiet inside compared to what they w
Re: (Score:2)
I do not want the extra bit of theoretical safety if it means I have to give up gobs of freedom and control over my equipment to have it.
And I do not want your feeling of freedom over your equipment to come at the expense of thousands of lives. By the way, you already don't have complete freedom over your equipment. You are required to wear a safety belt. You are required to have working safety equipment. You must have safe tires and sufficient brakes. You even have your emissions regulated. These are all measures to improve the health and safety of the population at the expense of your freedom over your equipment.
Our system is fairly safe today
It is horrifyingly unsafe.
Re: (Score:2)
1 death is a tragedy. A thousand deaths is a statistic. People are affected disproportionally by low probability but unusual events than they are by the everyday dangers. Thus fear of flying, where you're safer on average than driving.
I've calced that a self-driving feature could be worth $30k, assuming the thing lasts at least 5 years. And this is a system that's in the top 90% of drivers, no where near 'perfect'. Note this would be a 'total' self driving feature - you give it the destination, perhaps
Re: (Score:2)
I wonder at what point they'll stop requiring breathalyzers and start requiring that you simply use a self-driving car with any manual operation controls disabled
Probably right around the same time that they actually license self-driving cars without a human backup.
Re: (Score:2)
Really - tens of thousands per year? - I though that exaggeration, but apparently 33,808 died in the US from car accidents in the last year. Wow. ... and many places don't record the distances, so could be even worse. Maybe try Iceland, at 3.5?).
8.5 deaths per billion car kilometres. (Mind you, don't try driving in the United Arab Emirates - most dangerous roads with 310 deaths per billion km - 36 times as dangerous
Yes, clearly humans in control = carnage.
Looking at our society, it is scary how dramatic the
Re: (Score:2)
That's a great theory, except for two inconvenient facts:
(a) Our army is all-volunteer.
(b) Many poor people have little choice but to drive if they want to get to work.
Re: (Score:2)
Re: (Score:2)
Don't forget those blasted teenagers! Stay off my lawn!
Fine my me (Score:2)
Do not want. It's obvious at this point that the real deal with all these innovations is to retain more and more control over what people do and where they go.
The savings in time will result in everyone else in society being more productive and more successful than you. Good luck competing in that scenario:
Re: (Score:2)
sure, I acknowledged the convenience.. but it will come at a heavy price.
Re: (Score:2)
That's what they grunted about fire.
Re: (Score:2)
I understand that the complaints are implied "slippery slope" complaints, but every argument against would have applied to cruise control, yet none of them were true then. So why should I be worried about the same tired and and false concerns now?
Automated delivery. (Score:2)
4. Order online and have the grocery store's automated delivery vehicle bring you your groceries, complete with refridgerator and freezer compartments.
If they're reasonably busy that would allow them to make a stop or five in addition to your delivery while still delivering faster, reducing overall miles traveled per delivery. It gets loads more efficient if you're willing to wait until the next day. Remember, if you send your car it has to go there first, it's your fuel that's being burned, you don't no
Food truck laws may get in the way pizza oven insi (Score:2)
Food truck laws may get in the way pizza oven inside maybe fire laws as well.
Also what about fuel for the oven???
Re: (Score:2)
Pardon, I don't live in an area with crazy amounts of regulation about businesses, so 'food truck laws' are mostly the same as 'food establishment laws'.
As for the fire rules - wouldn't the engine/gas tank be bigger? I'm not thinking of just jamming one in there. As for fuel I was thinking mostly electric, though a vehicle completely powered by natural gas might be interesting. Heck, you can even power the thing via gasoline, though that might cost a touch more money.
Worst case a 20# propane tank will pr
Re: (Score:2)
Around here the food trucks complained about having to obey the "establishment" laws on grill temperature and working area around the grill and stuff so they got special rules.
Now they complain about not being allowed to park in streets and clog up traffic.
Of course its not just them, there's also whiney busybodies (probably the establishments) complaining about their use of lots they bought for parking.
Automated pizza cooking delivery car (Score:2)
Let's see, you quote grill temperature and working area. For our theoretical pizza delivery car, there's not going to be a grill, there's going to be an oven. If the oven isn't at the proper temperature, the pizza won't cook property, so I don't think that's going to be an issue.
As for working area, "what working area"? The pizza would be made at the store as it usually is, then loaded, uncooked, into the oven of the car, from the outside. Well, it might be precooked and the oven more of a 'finisher'.
Good point. (Score:2)
You make a good point - that would be an even more efficient way to deliver goods.
I don't see online ordering for things like fresh vegetables - people still want to squeeze the tomatoes and pick the best from the lot. Despite this, most of what comes from a supermarket could be auto-delivered.
One problem with this scheme (yours and mine) is that someone or something has to be home to receive the delivery. If the supermarket scheduler delays by a day for greater efficiency, it may not be convenient for the
Re: (Score:2)
I don't see online ordering for things like fresh vegetables - people still want to squeeze the tomatoes and pick the best from the lot. Despite this, most of what comes from a supermarket could be auto-delivered.
There are grocery delivery services already, and they do deliver fresh fruit and vegetables. From what I understand the online groceries have pretty much a 'no questions asked' guarantee, and are very good at delivering high quality produce*, thus people are satisfied. Worst case, remember that most food service establishments have everything delivered, and because they're getting a whole crate of X, it's fresh. The service might divy the crate up, but it's still going to often be fresher than what peopl
Re: (Score:2)
Never mind that this kind of technology could allow you to sleep while going to work, or sleep while going cross-country, or moon the people from the other cars, or pee outside the car drivers' window while doing 70 MPH.
Now, that's freedom! Having more options to do what you want is freedom. Now don't get me wrong, I'd also want to be able to turn off the automatic-pilot when I'd want. I just think your viewpoint on how to preserve that right is a little bit too extreme.
Re: (Score:2)
In the end it's not the technology or the convenience, it's the state and corporate middlemen who will demand that it come with some heavy strings attached.
Re: (Score:3)
Depends who has control, doesn't it? currently, we have control over our vehicles.. Sure, there are laws, but laws aren't control necessarily and they can't dictate reality. Having the ability to remote stop or disable a car that does not belong to you is another thing entirely. I guarantee this power will come with your shiny automatic car, and the list for the usage of this feature will explode as insurance companies, law enforcement, and the IRS clamor for access.
Driver interaction? (Score:2)
So if the driverless car stops in the road for a perceived pedestrian that may-or-may not be crossing the street will it give me the electronic finger when I lean on the horn?
Seriously, I see a lot of people standing close to the edge of the sidewalk that I think might be going to cross. Usually it turns out they are just chatting and aren't going anywhere. I suspect there will be a lot of false positives resulting in the driverless car slowing or stopping in traffic for someone who isn't actually crossing
Re: (Score:3)
I can also anticipate kids having fun with this by "faking out" the autonomous vehicles for a laugh.
Given the system uses unauthenticated inter-car communications to report to eachother on safety issues ... when it comes to faking out autonomous vehicles, forget standing close to the edge of a sidewalk:
"There's an app for that."
[parse error] (Score:2)
summary written by spam robot
marked as unreadable
parse error,
core dumped
*What the bloody fucking fuck* is this about? If you run wifi you cannot brake in time? What how why is this correlated to brakes?
Re: (Score:2)
Communication (Score:5, Insightful)
Yes, and this communication is not authenticated... which means you now have up to seven tons of machinery barreling around a corner... and if it's told that the way is clear, instead of blocked, instead of a gentle deceleration and safe crossing you get human hamburger. Up next on CSI... hacking GPS signals and inter-car communication to create the perfect murder: No forensic evidence, looks just like an accident.
I do not like the idea of autonomous cars depending on or accepting unauthenticated inputs, or having two-way communication abilities while in operation. We already have a pile of broken nuclear facilities in Iran caused entirely by malicious digital communications, the source of which can't be proven. Most systems rely on GPS and network communication for route planning, which is problematic enough but can probably be made reasonably secure... but when you start processing realtime data from unauthenticated sources to make operating decisions, not just navigation decisions, I just don't see it as being possible to secure because of the wide number of variables which could be influenced independently or collectively to create an unsafe condition.
Re: (Score:2)
I do not like the idea of autonomous cars depending on or accepting unauthenticated inputs, or having two-way communication abilities while in operation.
I tend to agree, given how appallingly bad computer security remains. Seeing the "cloud-based internet enabled crowdsourced" people getting involved with automatic driving worries me.
Re: (Score:2)
Re: (Score:2)
The paper references IEEE 1609.2 for trust based system. Such that all inter-vehicular networks use digital signatures and verify all messages.
Still, it's outside the scope of the paper.
Re: (Score:2)
More information in general? (Score:1)
Adversarial Implications of sharing information (Score:2)
What happens if any of those systems are hacked either for nefarious reasons or just so that the driver of the hacked car can gain some advantage by sharing misinformation. ?
In this setup of autonomous vehicles, they become essentially computers on wheels. The issues that are faced in network security can manife
Re: (Score:2)
That problem is solved if cars act as if all the information they can trust is their own, and only add "potential dangerous situations" reported by others to their own list, but never discarding them purely based on another machine's information.
Re: (Score:2)
Re: (Score:2)
But if your point is that it is still not perfect, I agree.
Car Insurance? (Score:1)
Re: (Score:3)
Theoretically there is always a chance of a crash. I would expect the chances of crash to be less than human driven cars and would expect insurance to be way cheaper. But yes, the car would need insurance. And yes, it would be the cars fault, and insurance premium for all cars of the model would increase.
Whats the point ? (Score:5, Funny)
Why do we need driverless cars?
The largest use of automobiles is to transport the driver (the sole occupant) around a city, or between cities.
So if you take the driver out of the vehicle, why does it need to go anywhere?
Re: (Score:3)
Was watching some show recently which depicted late 1800's London, and the guy was in his horse drawn buggy headed to the country on a small road and was asleep. And I thought to myself, at some point we transitioned from intelligent vehicles which reacted to the enviornment around them, and had basic collision avoidance and guidance, to mechanical systems where were we sit directly in the loop all of the time. Nod off on a horse carriage and you'll not likely end up in a tree, try doing it in your car.
I ca
Re: (Score:2)
Was watching some show recently which depicted late 1800's London, and the guy was in his horse drawn buggy headed to the country on a small road and was asleep. And I thought to myself, at some point we transitioned from intelligent vehicles which reacted to the enviornment around them, and had basic collision avoidance and guidance, to mechanical systems where were we sit directly in the loop all of the time.
At some point we transitioned from using horses which can be panicked and run people down, to toyotas. Er, wait.
Re: (Score:2)
Re: (Score:2)
With complex routing software, cities especially could become far more efficient, yet still very convenient (unlike much human driven public transport). It can help with congestion too. Much of todays gridlocks come from behaviours of humans that can be suppressed and overcome in machine. Forcing clean flow. Also
It's kind of a shame (Score:2)
That robot cars will be required to be a million times safer than people thereby negating any practical benefit. A robot that can do as well as 70% of the stupid rednecks, 90 year olds, soccer moms, illegal Mexican, drunks, teenagers, women yakking on the phone etc is way way way ahead of the game.
Re: (Score:2)
Re: (Score:2)
My only complaint about Mexicans is that they almost never pull over for you while they're driving slowly to avoid being pulled over and shaken down by the man. I don't know if my willingness to let people by is a form of reparation for my peoples, or because I'm only a quarter Mexican.
Re: (Score:2)
No, No, and No. (Score:2)
If your solution to moving object and in-motion safety is analyzing gigabits of data per second, you're already incorrect. It already won't work. It already won't be secure, it won't be safe, and it'll only work at STP. Safety has never been about longer checklists and more data. It's always been about learning which few of those billion bits are the important ones, learning to identify them, recognize them, and act upon them, in spite of the enormous amount of data surrounding them.
That's not a process
Do we really need this? (Score:2)
I just realised a Major problem ... (Score:3)
Re: (Score:2)
Mod parent +5, This Is Why We Can't Have Nice Things