Slashdot Mirror
Self-Driving Cars' Shortcomings Revealed in DMV Reports (mercurynews.com)
A demand from the California DMV of eight companies testing self-driving cars has highlighted a number of areas where the technology falls short of being safe to operate with no human backup. From a report: All companies testing autonomous vehicles on the state's public roads must provide annual reports to the DMV about "disengagements" that occur when a human backup driver has to take over from the robotic system. The DMV told eight companies with testing permits to provide clarification about their reports. More than 50 companies have permits to test autonomous vehicles with backup drivers on California roads but not all of them have deployed vehicles.
It turns out that a number of the issues reported are shared across technology from different companies. Some of the problems had to do with the way the cars sense the environment around them. Others had to do with how the vehicles maneuver on the road. And some had to do with what you might expect from systems made up of networked gadgets: hardware and software failures. The disengagement reports themselves identify other problems some self-driving vehicles struggle with, for example heavy pedestrian traffic or poorly marked lanes.
It turns out that a number of the issues reported are shared across technology from different companies. Some of the problems had to do with the way the cars sense the environment around them. Others had to do with how the vehicles maneuver on the road. And some had to do with what you might expect from systems made up of networked gadgets: hardware and software failures. The disengagement reports themselves identify other problems some self-driving vehicles struggle with, for example heavy pedestrian traffic or poorly marked lanes.
Reliability will become ever more important as full AD becomes standard and humans' driving skills deteriorate (or never develop)
Holy shit, they've just described, you know ... driving.
This shit is what happens pretty much daily, and for which the act of driving requires you to have a high degree of situational awareness.
Just this morning as I was driving into work, some clown turning off a street into the road I was driving on ... he hesitated, then apparently said "fuck it" and went anyway. Unfortunately he didn't seem to be aware enough or intelligent enough to have noticed me. The end result was I had to pretty much do a panic stop behind some idiot who unsafely pulled out into oncoming traffic, and was pretty much suddenly in front of me and driving at half my speed (which was the posted limit).
Why the hell are these companies acting like they have self-driving cars when they clearly can't handle driving in the real world.
This shit is never going to work if it can't handle random, illegal, and stupid behaviour from the humans on the road. And if they think the world is going to replace all cars with autonomous vehicles, then clearly they expect the rest of us to pay for that future.
This is just hubris from the tech industry who are pretending they're closer to solutions that work in the real world than they really are. The fact of the matter is, we're not all going to run out and buy new cars to allow this awesome future as envisioned by corporations to actually ever happen.
Sorry, but all of the stuff in the above quote is pretty much mandatory for driving a car.
It is not like anyone knowning anything about AI could have told them it isn't ready for general roads, and only for special roads. It is not like allowing them on special roads first is exactly the plan in Europe where government listens to experts.
It would be like the donkey which dies of thirst AND hunger when placed equal distance between food and water.
Add a failing test, add code to fix failing test, deploy to all vehicles - fixed. Humans rearly learn in that fashion.
A self-driving car's software has the priority to minimize litigation of its creating company.
“Common sense is not so common.” — Voltaire
Seldom (if ever) is there the rather obvious suggestion to limit autonomous vehicles to simple point to point 'highway' trips; but that's exactly where and how it should be done for the foreseeable future, if it happens at all. That is, the (literally) lethal mistake is to introduce autonomous vehicles into the complex and chaotic world of city driving. The next time you drive in the city consider how many of your decisions are predicated on understanding subtleties (some might occasion "stupidities") of human nature: "Is that guy looking at the person as they're talking on the corner? If so, they aren't as likely to start across the street" "Is that a child's toy which just bumped a bit into the road (to be chased by a child) or just a blown leaf?" "OK ...four way stop: it's that guys turn, but, he's got a cell phone in his hand he's consulting" ...etc. So, start out with truck loads from freeway exit 113 to 114, then if that works, exit 117...
Humans are unlikely to improve at driving. Software can potentially be improved on. That's not the same as saying today's software is safer than a human though.
“Common sense is not so common.” — Voltaire
It is going to run a damned lot of steering scenarios, pick the one with the smallest projected casualty, and save a log of its calculations for the inevitable trial.
Trust me, the car is perfectly capable of seeing options you as a driver cannot, simply because a CPU runs faster than a human brain for mathematical computations.
-=This sig has nothing to do with my comment. Move along now=-
In the US we define experts as those who are paid by industry leaders to say the things we like to hear.
“Common sense is not so common.” — Voltaire
so if I'm a self driving car with no backup operator, do I prioritize the safety of my passengers? if I have to run down 5 people to keep my rider safe, do I do that?
The car should run over the 5 pedestrians, then the litter of puppies across the street and finally drive off of a bridge. Bonus points if the horn sounds like a maniacal laugh or plays Dixie.
what if I have to do the whole run over your mother / a baby / a nun or run over a bunch of assholes? how are they ever going to solve for this, because whatever it chooses will be wrong
Run over my mother, sideswipe the baby so it goes airborne into a dumpster, and back over the 5 assholes. Preferably spinning the tires on their dismembered corpses. The nun is difficult. I'm not sure if it will piss off more people to hit the nun, or avoid her after all of the other carnage.
You're right, some situations are difficult. Oh wait, avoid the nun and run a bus full of orphans off of a cliff. That will probably piss the nun off to. Problem solved.
Reducing your kinetic energy should be your first priority.
Any sufficiently unpopular but cohesive argument is indistinguishable from trolling.
So the self driving cars act just like an inexperienced millennial who can't drive a stick or read a map, and texts constantly?
Please describe how, in any kind of physically plausible scenario where one is travelling at a lawful speeds that are actually commensurate with any reduced visibility they may have, one would actually be faced with being unable to stop for five people on the road.
File under 'M' for 'Manic ranting'
After a passenger eats at Chipotle, the vehicle will automatically set course for the nearest hospital, notify the CDC, and engage biowarfare protection protocols.
https://www.dmv.ca.gov/portal/dmv/detail/vr/autonomous/testing
What if increasing it would avoid the accident entirely?
It's odd to me how so many people think the brakes are always the answer. I guess being a guy who rides motorcycles, I understand better than most how getting the hell out of the way is just as important as stopping.
I told you so.
Trust me, the car is perfectly capable of seeing options you as a driver cannot
That's the opposite of what the problem is. The backing up truck in Vegas, the woman cyclist, the concrete barrier... why can they NOT see things that people can clearly see? Don't use the excuse that *sometimes* people don't see them either; because people can be distracted, and these cars aren't supposed to be.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
Self-driving motorcycles will operate under a different set of rules.
Any sufficiently unpopular but cohesive argument is indistinguishable from trolling.
GASP... who would have thought driving was actually hard for a computer to do? Maybe they can keep paying attention EXCEPT when it comes to concrete barriers, or backing up trucks, or pedestrians in the street. But millions of cycles a second means nothing if they're not successfully driving with them.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
No, because no trial ever took place.
APK is a fine upstanding citizen and a beloved, trusted member of society and this community. Why are you trying to tarnish that good manâ(TM)s reputation? You need to look deep inside of yourself and confront your inner monster lest you become one.
so if I'm a self driving car with no backup operator, do I prioritize the safety of my passengers? if I have to run down 5 people to keep my rider safe, do I do that? what if I have to do the whole run over your mother / a baby / a nun or run over a bunch of assholes? how are they ever going to solve for this, because whatever it chooses will be wrong
I don't think it counts. I think it's a simple set of tradeoffs, health > property, life > health, occupants > bystanders.
Ie, if given a choice between damaging the car or hurting a bystander it will damage the car.
If given the choice between smoking a pedestrian or a dangerous swerve into the ditch it chooses the ditch.
But if a pedestrian materializes in the middle of a bridge with no guardrails, well goodbye pedestrian.
Of course, that's a choice 1 or 2 generations down the line, right now it's a challenge just to keep them from killing people for no reason whatsoever.
I stole this Sig
They will work great until some prankster repaints the lines and leads them off the road.
Because mv^2? Velocity is squared to mass in energy tranference. Also breaking will deescalate. During an incident you should already be slowing down to preserve options. There are few exceptions such as rail tracks, but speed is not safety increasing in the first place!
In their last month of reported testing ( November 2017 ) Waymo only had 1 disengagement, in 30+k miles of driving, because the car couldn't understand the behaviour of another car.
I am pretty sure I've never driven a stretch of more then 2k miles where I didn't encounter at least one instance of 'unexpected' behavior by another driver.
I'd say that - even today - there at least some companies which have autonomous vehicles that are safer then most, if not all, drivers. I think it's not an exageration to think that, within 36 months, there will be autonomous cars better then all drivers in most situations.
"Trust me..."
You do realize that's the lead in every single con man in the world uses right? Why should I trust you?
Also remember- these cars are solely being tested in some of the BEST conditions for these systems. In real world driving they will be ABYSMAL!!!!
Except that the ability to actually comprehend that what is being computed, in your example math, is beyond that of a CPU. It has no comprehension of anything it computes, it just does it. They inherent value of those computations lie in the in the moral fabric of humanity. To choose one scenario over another is not always something that is best decides by a computer devoid of morality.
I am wondering if self driving cars take into account probabilities when making decisions? it seems to me that you can never be 100% safe when driving, and are constantly weighing probabilities constantly in your head. like something similar to a 3d markov chain?? i dunno
I'm curious what your results on the self-driving car quiz would be. Hey, if you do well enough, they could use you to program the next generation of self-driving algorithms! Probably on the 'criminally insane' setting.
Brake as hard as you can, and keep your lane unless an adjacent lane is clear and safe to swerve into. A simple strategy that requires no moral choice and is also fair towards other road users: any other rash action increases the chance you won't hit whatever you were going to hit and probably had no business being in the road in the first place (otherwise you weren't going to collide with it), and increases the chance of hitting an innocent bystander who had nothing to do with you or the obstacle.
If construction was anything like programming, an incorrectly fitted lock would bring down the entire building...
I'm not familiar with all of the cases you mentioned, but for the one that I am, it's my understanding that the sensors that were used on the self-driving automobile were inferior to LIDAR, which I suspect may have prevented the accident.
But modern sensors, despite being superior to human senses, still have some technological limitations. You can't blame the software for what the hardware doesn't see.... at least we know that hardware exists with better acuity than human senses which, when it used correctly, should reduce accidents considerably.
File under 'M' for 'Manic ranting'
But, less safe than any driver in the 10^30 situations it hasn't encountered before. Meatbag drivers handle that shit casual, avoid the worst and get on with their business.
.... ever written any piece of code ? Bugs are not suppose to happen... and yet....
We will probably never have streets filled with fully autonomous vehicles. It isn't a technology that can realistically be applied to something like driving a car. Isn't going to happen. It was a cool and very expensive tech demo, that's it. We may see them for specialized uses in controlled environments, maybe. People in the Valley really, really need to grow up and join the rest of us in reality.
There are times that they can't see the faded lines in the road or the stop lights? Well, yeh. I agree. Same here. And I don't consider that safe either. There are many different inherently unsafe intersections near me. The only difference is that the self-driving vehicle can avoid taking the chance by turning it over to me to take the chance. I honestly don't know if it would be better if it kept control and tried it than handing control to me. But, the buck is passed.
The biggest problem that autonomous vehicles have to overcome is that they expose the faults in a very bad system through extensive data collection. Exposing faults in something people depend on is never an easy road. The messenger often becomes the victim.
Like most who have lived more than half a century, I've been in a number of accidents including:
Driving is dangerous. Four out of four of the accidents above were because of driving while impaired in some fashion - a teenager unprepared to drive in rain, driving while tired, blinded by the sun while driving (should have stopped or greatly slowed), and driving while distraught. And there are numerous other incidents that didn't rise to the level that I would term an accident that would be reported by these self-driving vehicle regulations.
Road maintenance is also atrocious in this country and human drivers die because of it every day. Human drivers are also prone to complain that others shouldn't drive while impaired and then make exceptions for their own needs.
Maybe we should start this conversion by just shining a very bright light on reality - require every new car to be equipped with the sensors to record the reality and disclose every bit of it to the DMV - every solid line crossed, every rolling stop, every time the light is red and we're stil
Well, you can blame the autonomous vehicle company for not using hardware that will give the software enough data to process so that it results in a safe driver. That is the equivalence of allowing grandma to drive after she can no longer see properly from cataracts. You can also blame the company for not correctly interpreting the data from the hardware, which is the equivalent of allowing grandma to drive after she develops serious Alzheimer's.
It would be one thing if these companies didn't already know their cars have certain flaws. But the fact is, they know they have vast weaknesses and put them out on the road anyway.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
Facebook just trained their image recognition "AI" with over 3 billion instagram images
They then only scored 85% in a test.
Self-driving cars aren't going to be able to recognise what they're looking at. Is it a person crossing the road wearing a big raincoat, or is it a newspaper being blown around in the wind?
What if increasing it would avoid the accident entirely?
A common response from people who think they're a better driver than everyone else. Just drive (ride, whatever) safely, and not at excessive speed. It's not that hard, and you're not in a rush.
One of the more recent google self-driving car accidents where they ran into a city bus - the car pulled from a right turn only lane into a traffic lane and didn't see a 30 ton city bus traveling normal speeds.
While I admit I do have blind spots - busses usually don't hide in them - especially when performing what is essentially an illegal moving violation (if I'm going to try to pull off something that dumb I'm usually a bit more alert about what I'm going to hit).
Until all the cars on the road have LIDAR, and they start interfering with each other. That's going to be a problem with all types of active sensors - LIDAR, RADAR, ultrasonic. etc.
They all work by sending out a signal and measuring the reflections. That works until there is too much other noise because all the cars around you are also shining their lasers on to the same things you're measuring. Or the laser projections from other cars directly hit the image sensor and throw out the signal to noise ratio it can work with. Or the sun is shining on it, also flooding it with a similar wavelength to the infrared lasers they use.
So the self driving cars act just like an inexperienced millennial who can't drive a stick or read a map, and texts constantly?
I was born in 1991, which makes me Gen Y or "millennial" (uggghhhhhhh). I've been married for 6 years and I have a mortgage. I think you mean Gen-Z. Now get off my non-existent lawn.
I'm curious what your results on the self-driving car quiz would be. Hey, if you do well enough, they could use you to program the next generation of self-driving algorithms!
That was a stupid quiz. I wasn't once offered any of the following:
C. Kill everyone
D. Set off a dirty bomb
E. spread Anthrax through the tailpipe of the car
Probably on the 'criminally insane' setting.
I'm looking at it as population control a la Death Race 2000.
Preservation of human life should be your first priority; whatever is required for that (which, admittedly, is typically going to involve reduction in kinetic energy, but will also often involve modification of the path of said kinetic energy). This brings up a really interesting possibility for self-driving cars as the technology improves: the ability for the car to take into account its own safety features (crumple zones, airbags, etc) and the safety features of various things it could crash into in the event of an unavoidable wreck when deciding how to proceed. For example, a direct head-on collision is vastly less dangerous for human occupants than offset head-on collisions, so two autonomous vehicles could potentially negotiate with one another to steer directly into each other (totally counter-intuitive) if circumstances made that the option most likely to preserve life.
Where might this come into play? Every been on a fairly narrow, windy mountain road? (think coastal mountain highways in California as an example) What happens when an object (be it falling rocks or a reckless cyclist) suddenly drops into the roadway in a completely unpredictable way and there's no guardrail or shoulder on the cliff side of the road? Humans will typically panic in such a situation and react in unpredictable ways, potentially sending someone off the road into an uncontrolled rollover situation or even running over the cyclist. Two autonomous vehicles may mutually decide and communicate to one another in a fraction of a second that by braking to just below the steerable friction threshold while turning directly into one another, odds are extreme high that a simple fender-bender will be the extent of the damage, both vehicles will remain fully operational, and no human beings will sustain any serious injuries. Still an accident, but a nice, controlled wreck where everyone leaves unharmed. Should those circumstance be between a car and a motorcycle, perhaps the autonomous car decides that braking hard while steering slightly into the cliff wall best preserves human life.
In any case, whatever the challenges are that exist today, the naysayers will inevitably lose in the end because humans are incredibly poorly adapted for operating motor vehicles. I say this as someone who loves to drive and will probably continue to drive manually so long as it's legal to do so. But if computers aren't already better overall drivers than humans, they absolutely will be within a decade or less.
-- "Government is the great fiction through which everybody endeavors to live at the expense of everybody else."
I am not going to trust you. Sure, a computer can do the math faster, but a human brain is not evaluating the problem as a series of calculations. Both require training to know what options are possible, good and what the side effects are. From what I have seen, Humans are better at making leaps of logic between them (and sometime logic leaps off the proverbial cliff).
It must be nice to be this optimistic. The numbers will inevitably prove you wrong, of course. The companies who are working on autonomous vehicles think this will just require a certain amount of engineering know-how to make it work, without taking into account how much effort already goes into making roads and highways safe for human drivers. The companies are developing a technology for a system that relies on constant road maintenance and repair, lane markings, car safety regulations, driver training, complex right-of-way rules, and constant policing. Of course you say "a solution is only 36 months away", because billions of dollars are spent every year keeping the roads as safe as they currently are.
The research is only possible because of the externalized costs of the research environment, but the safety of that environment will be built into the technology, ensuring that it does not actually work in those pesky situations where it really matters. And as the number of people killed by autonomous goes up, the justifications for the technology will become more difficult to swallow.
I think I have seen this account name before. You are that guy who loves AI, aren't you?
One laser can intersect another, and they don't interfere with eachother. They can each have their own modulation and so can be detected individually, regardless of other vehicles using the same technology. Current technology can do millions of these per second.
If car is mistaking sunlight for its own LIDAR signal, then the software is at fault, not the hardware.
File under 'M' for 'Manic ranting'
Back in 86 at Michigan State I took an experimental c++ course. We were given a bunch of data points and required to drive a car around a course with obstacles etc. I was so proud of my work. I settled in for my demo. The professor said your course is a triangle. Oh god my car flipped out and just did a hard right off the track. I didn't code for that. Ugh!!! Not sure my point. But it was awful.
In this situation it was a failure of the AI to understand what a bus is and how it inherently behaves. The SDC assumed the bus would stop quickly to allow right of way, but of course buses can't just stop like that because of the riders inside; something humans inherently know.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
Every kind of sensor has a weakness. Lidar doesn't work in fog, heavy snow and rain or if there is dirt on the sensor. Cameras don't work if they are dirty. The question is whether a few different types of sensors will ever be put together in a way that works all the time and is still affordable to a consumer.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
In a swerve / recover situation, the part of the correct response is to use the accelerator.
So yes actually, you just proved you, not the OP are one of those that doesnt know about driving.
No, it's like putting a perfectly competent driver in a car when, for only lots of added expense, you could put in two like planes or driver training cars.
It's the equivalent of complaining when something is 5x better that it's not 6x better.
If the cyclist qualifies as a MAMIL* then run him down. You are doing us all a favour.
*Middle aged man in lycra.
New Zealanders are well balanced with a chip on each shoulder. One represents Australia, the other the rest of the world
You are already in a bad situation that you have lost control of. By definition you don't know all the variables/outcomes. Making quick decisions without all the data is the most likely way to cause an accident, and you are already in one.
What could possibly go wrong.... Yes instead of slowing down and having a minor collision with a subcompact, try speeding up and going around it, and totalling yourself on that semi you didn't notice until too late.
Humans are unlikely to improve at driving. Software can potentially be improved on.
And the halting problem can potentially be solved too.
Tell you what, lets wait until the potential is achieved before assuming that it will be achieved.
I'm a minority race. Save your vitriol for white people.
Humans can easily improve at driving, but no politicians want to be kicked out of office when they attempt to mandate additional drivers license requirements. Have you taken a defense driving class? Been able to practice swerves, tire blowouts, etc... in controlled environments? Did your tester walk around your car to ensure all your mirrors were aligned correctly? Do gas stations require you to check all tires before pumping gas? There are tons of room for improvements.
Software can making driving near perfect. It'll take a while, but it could do it. It would be easier if we modified road ways to help it.
In their last month of reported testing ( November 2017 ) Waymo only had 1 disengagement, in 30+k miles of driving, because the car couldn't understand the behaviour of another car.
The average driver averages 1 accident every 250000km and 500000km, not 1 accident every 30000 miles. This includes the pool of all drivers, by the way: those who are drunk, speeding, texting, phoning, etc while driving.
I'm waiting for SDCs performance to be as good as the average driver.
I'm a minority race. Save your vitriol for white people.
A company would have to be run by complete and absolute morons to ever even suggest trying to make the AI perform "moral choices." In nearly 100% of situations, you slow down, and, maybe if possible without hitting another detected object, swerve around the obstacle if it is completely stationary or already moving in the opposite direction of the potential swerve. If there's nowhere to easily and safely swerve, you just reduce your speed as much as possible before the likely collision to limit kinetic energy (and hence damage/loss of life) as much as possible. At no point should a self driving car ever make any "moral choice" period (and frankly, people probably shouldn't try to either. Thinking about it WILL cause you to hesitate, and that moment of hesitation could easily make the difference between avoiding a collision entirely and failing to avoid one).
A disengagement is not necessarily a would-be-accident.
This space intentionally left blank
Driving into the sun, any kind of snow, hydroplaning, hard to identify objects in the road, road construction, pedestrians doing stupid stuff, other drivers not following predictable actions or laws, driving in the rain, sensors failing due to age and wear, and about 100 other items are unsolvable problems that our level of technology, especially AI, is nowhere near yet. This is like all those "before their time" inventions in the 50's. The technology just wasn't there to make it real no matter how convincingly they presented it. AI would have to be at human 3D and visual processing level or better and the best AI in the world is around cockroach level intelligence currently, with the best measurement estimate.
Too.
I'm not familiar with all of the cases you mentioned
The truck in Vegas was a case of a self driving bus stopping in the blind spot of a truck as it could not pass. The truck was slowly backing out of a driveway (illegally) and the bus just stood there waiting, not taking any measure to evade or signal the truck driver. Basically it was hit in a situation that any human driver could have easily avoided.
the woman cyclist
That is the case were people believe that Uber wasn't using LIDAR.
the concrete barrier
A Tesla driver trusting the Autopilot feature to just work on the same road it worked flawlessly in the past. I think they issued a software update the day before and his car made a beeline for the barrier. It basically boils down to another Tesla driver trusting the hype, ignoring the "keep your hands on the wheel" requirement that Tesla doesn't seem to enforce.
A disengagement is the equivalent of my wife warning me about an approaching object before I see it, even though I keep my eyes glued to the task at hand even on a long trip.
Yes, I see the bicycle, thank you. Yes, I see the pedestrian, thank you. Yes, I see that person in front of us driving like an idiot; there's one behind us, too! Eventually we get to, oh yes, thank you, I didn't see that goat on the shoulder. That's the equivalent of a disengagement. But a goat is unlikely to jump in the street.
If you look at the reports, some companies the disengagements are situations where they almost killed somebody. Waymo disengagements are more like the goat on the shoulder. Eventually the government will figure the testing out, and then the wheat will be separated from the chaff.
Some humans are probably very unlikely to improve. I took at Tesla Model S taxi to the Airport yesterday, some 20 km. The driver drove like a car thief and almost got into two collisions with equally reckless drivers. If the driver hade left the car in autopilot mode with oranges on the wheel, I would probably have been safer.
The sooner we can have self-driving cars, the better.
Humans are unlikely to improve at driving.
Which human drivers, though?
Looking at road death statistics would seem to suggest that German drivers are a lot better than American drivers, while for instance Russian drivers are a lot worse. Now it's not all due to quality of driving (but also quality of roads, law enforcement, car safety standards, etc.) and the quality of driving itself is impacted by many factors, but we can safely say for instance that drunk driving is a huge problem in Russia that contributes a lot to their death rate being higher than America's, while the far more stringent process of driver training and driver certification in Germany contributes a lot to their death rate being lower.
The companies developing self-driving cars, being either American or doing most of their testing in America, seem to take American drivers as the baseline. However, looking around the world one can conclude that American drivers could indeed improve quite a bit. Not to mention that historically, drivers have improved...most places in North America, any moron can easily get a driver's license. Even that's an improvement based on the situation several decades ago, surely it can be improved still.
well what would be your plan ?
Comment removed based on user account deletion
If you can drive in fog, rain or snow where LIDAR fails, then you won't need it in fair weather at all and there is no need for it.
Why? Driving in adverse conditions is harder than driving on a sunny day. So if you can manage that without LIDAR, why install it at all?
On the other hand, if you can't drive without LIDAR, the idea of autonomous cars is already dead since people will not buy/use a car that is unable to drive in rainy weather.
So in short, LIDAR is not necessary.
I'm sorry but autonomous cars should never need a connection to an Active Directory server.
how are they ever going to solve for this, because whatever it chooses will be wrong
Do what humans do. Step on the brake and hope for the best. The goal of self driving cars is not to solve moral question that don't actually apply in the real world anyway.
Oh wait, avoid the nun and run a bus full of orphans off of a cliff.
Yep, username checks out.
If you've ever driven in high wind situations, you know that they require a constant battle to stay on course.
Also, especially for higher profile vehicles, one must drive extra cautiously. Furthermore, the cruise-control
will require constant adjustment especially in variable winds. I would like to know if they can do all of
that.
I'm British so I am not fully cognisant of US driving rules. Why would a car in a right turn only lane have right of way over a bus in a lane to its left? In the UK, if you change lanes, you have to give way to traffic already in that lane.
It seems to me more likely that the self driving car simply did not detect the bus. Either that or it didn't understand the rules of the road.
All I want is a secure system where it's easy to do anything I want. Is that too much to ask ~~ Randall Munroe
Those weren't autonomous cars. Those were incidents of stupid drivers mis-using cruise control.
So if a self-driving car can work in America, it'll probably do just fine in Germany. It'll probably struggle in Russia though.
Even if there are failure cases for autonomous cars, if the overall safety is significantly less than human drivers reason says we should implement post haste. But the reality is humans are emotional thinkers, and we likely won't accept that. The first time a self-driving car runs over a kid we will be pulling them all off the road even though ten other kids were run over by humans the same day.
This posting is provided 'AS IS' without warranty of any kind, implied or otherwise.
It can be a whole lot safer and Elon was on the right track with the Boring company, just not on target. That is an underground system, in tunnels, a controlled environment, where you would catch a vehicle much like a turbolift and it will take you where you want to go. You could even own your own lift and have a parking spot, would not be cheap. So replace, taxis, public and private transport in major metropolitan areas, every current road would have an underground tunnel system, proving that controlled automated transport environment. It would not necessarily have to be magnetic suspension but could be wheels in enclosed tracks, with track switching to change direction.
Far more sensible than trying to make roads work under increasing traffic loads.
Chaos - everything, everywhere, everywhen
I'm sorry but autonomous cars should never need a connection to an Active Directory server.
How else will cars know who is and isn't an authorised user? Are you honestly suggesting we use OpenLDAP instead... what's next, Sendmail to replace the CANBUS?
Calling someone a "hater" only means you can not rationally rebut their argument.
The lane markings were murky, because it was a parking lane that the automated car was partly in. The automated car had to steer around a sandbag by the curb though. The automated car thought it was already in the same lane as the bus (not merging into a lane) and so had right of way since it was in front. Or, it could not have seen the bus at all... possible. I would really hope it saw the bus.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
Trees block traffic lights and GPS signals. Pedestrians aren't always seen by the bot-cars. So unless you live in a desert, once in a while your car is going to try to kill you or a pedestrian.
There is no way all traffic lights are going to be networked to talk to vehicles in the US. I predict "bot-car friendly roads" will be marked as such, and cars will have to shut down autopilot type driving in all other areas.
But we all know that autonomous driving is safer than humans. The tech companies have told us that. They even have whitepapers and actual numbers based on tens of thousands of logged miles. Don't you trust them?
None of what you have said is inconsistent with there being issues to work on with autonomous driving.
Streets and roads are a huge factor as well.
Streets and roads in the US are designed to be very forgiving to drivers, and tries to optimize for automobile speed and minimize stopping. To use one local example, pedestrians complained about drivers at an intersection, so the city put in a small concrete median - then when too many cars were hitting the median, they removed it. Think about that for a moment - the city put in a safety device and then when drivers couldn't stop hitting it, the solution was to remove the safety device and revert the street back to how it was before.
can LIDAR read expressions? Can it tell that the pedestrian is turning head left right in an obvious posture preceding stepping out into the street? Can it tell that the pedestrians head is tracking a vehicle and infer what that implies? Can it observe that the driver in front of you is texting or otherwise behaving dangerously and back off to increase safety distance? Can it avoid leaving too much safety distance to avoid enticing a driver with different risk tolerance from abruptly cutting them off?
This sort of overall situational awareness is not just "there's an object at position x,y moving with velocity i in direction j."
It would be one thing if all road conditions were fixed and all vehicles became self-driving at a single instant. Having to share the road with human drivers means they have to have the same degree of flexibility -- which at present seems vanishingly unlikely.
Comments about interstate only driving miss the problems with rush hour traffic -- don't expect interstate driving to be safe for a self-driving car anywhere near a significant urban center during certain times.
Your argument would make sense if we hadn't already seen many accidents that are clearly due to the vehicle not seeing or interpreting things as well as a human.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
Your argument would make more sense if people didn't crash every single day.
Lets start building the vacuum-powered hamster tubes from Futurama.
so if I'm a self driving car with no backup operator, do I prioritize the safety of my passengers? if I have to run down 5 people to keep my rider safe, do I do that? what if I have to do the whole run over your mother / a baby / a nun or run over a bunch of assholes? how are they ever going to solve for this, because whatever it chooses will be wrong
No need for moral choices (and no capability for it anyway with machines) ... how about just don't hit stuff?
I think that pretty much covers it? ;)
People crash once every 57,000 miles. No automation can touch that.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
And how many of those crashes would be avoided if the car also had better sensors or simply more competent drivers?
I don't see what this has to do with the fact that automated cars aren't really ready to be on the road yet.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
A disengagement is the equivalent of my wife warning me about an approaching object before I see it
No. A disengagement is when the car has to be actively prevented from killing someone.
(What makes you think your characterisation is the correct one? A disengagement is basically the car saying "I dunno").
I'm a minority race. Save your vitriol for white people.
And the halting problem can potentially be solved too.
I believe it has been proven not to be solvable for general cases. As there is a simple example where halting is indeterminate so we know that the general solution doesn't exist.
Tell you what, lets wait until the potential is achieved before assuming that it will be achieved.
You overlooked the hedge word in my statement. Words like potentially, maybe, or might, indicate a lack of commitment by the speaker. In this case it because I feel there is a lack of concrete proof, and not because I'm trying to use weasel words to make vague statements.
“Common sense is not so common.” — Voltaire
According to some research, automated cars crash less than human drivers. 1.6 compared to 2.5 crashes per million miles for the most severe category of accidents, and 5.6 compared to 14.4 for the least. The problem is those miles are not really comparable to the entire set of miles driven by humans. I would say the answer is nobody really knows how safe they are compared to people.
https://www.vtti.vt.edu/featur...
I've accelerated to successfully avoid an accident before, although not all that frequently. If there's an accident waiting to happen in one place, and I'm somewhere else by then, no problem.
"When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
I'm not sure that's the best strategy. Consider traveling on a highway at 65 mph. The (human driven) car ahead swerves aside to reveal a nearly stationary vehicle or other large object. Both lanes to the sides are occupied. Is it better to hit the vehicle ahead at whatever speed you'll be traveling after max braking, or gently sideswipe one of the cars to the side? Maybe the car to the side can get over into an empty lane or shoulder and it will be a very mild accident.
The good thing about this is that there are tons of poorly engineered roads, intersections, and road signs all over. When you complain about one of these, the city traffic engineers tell you to go fuck yourself. However, Google and its buddies have enough clout to get these things fixed. Currently, Calif is making freeway stripes wider to be more visible to self-driving cars.
Unfortunately, there are going to be plenty of issues that bother ONLY self-driving cars. Google and its buddies also have the clout to get those things fixed. But, guess who is going to be paying for that. Not Google, my friend; the bell tolls for thee.
......global warming
LIDAR's primary advantage over human senses is its ability to see things in the dark. It can be combined with other types of imaging technologies in other situations to create an impression of the surrounding environment that,while it might be technically more limited than what is ordinarily capable of by any extant adverse conditions, will still generally be superior to what a human being is capable of in the exact same conditions.
File under 'M' for 'Manic ranting'
There's always going to be exceptions and unforeseeable consequences. Maybe you'll squeeze by, or just maybe you'll gently nudge the car next to you into an oncoming bus full of nuns. If the adjacent lanes are not clear and there's no room to squeeze by (something the car should be able to determine in microseconds), attempting to slow down as much as you can might still be the best strategy even in this scenario.
If construction was anything like programming, an incorrectly fitted lock would bring down the entire building...
I've been wondering - is there a single autonomous vehicle out there that recognises hand signals from cyclists?
attempting to slow down as much as you can might still be the best strategy even in this scenario.
It certainly might, and clearly often will be. But it won't always be the best strategy, which is what it sounded like you were saying.
You were complaining that they weren't perfect, they are still closer to perfect than people. (In the situations they are being used in) They aren't ready yet to take over.
>automated cars crash less than human drivers. 1.6 compared to 2.5
they drive in IDEAL circumstances and are BARELY 50% better.
Its the same rules in the US (except reversed) - if you watch the video:
https://www.youtube.com/watch?...
The car is in a right turn only lane - the bus didn't have to yield at all - however there aren't clear lane markings. Also the bus was only going 15 mph. Very clearly a moving violation.
Even if it was a parking lane like others have suggested (which its also illegal to be parked so close to an intersection) - the person parked has to yield to moving traffic.
Yes, so they are better, tell fluffy nuts.
And since when is 50% more of anything barely, in all caps even...