Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Hidden Obstacles For Google's Self-Driving Cars

Soulskill posted about 2 months ago | from the not-looking-forward-to-denial-of-driving-attacks dept.

Transportation 289

Paul Fernhout writes: Lee Gomes at MIT's Technology Review wrote an article on the current limits of Google self-driving car technology: "Would you buy a self-driving car that couldn't drive itself in 99 percent of the country? Or that knew nearly nothing about parking, couldn't be taken out in snow or heavy rain, and would drive straight over a gaping pothole? If your answer is yes, then check out the Google Self-Driving Car, model year 2014. Google often leaves the impression that, as a Google executive once wrote, the cars can 'drive anywhere a car can legally drive.' However, that's true only if intricate preparations have been made beforehand, with the car's exact route, including driveways, extensively mapped. Data from multiple passes by a special sensor vehicle must later be pored over, meter by meter, by both computers and humans. It's vastly more effort than what's needed for Google Maps. ... Among other unsolved problems, Google has yet to drive in snow, and Urmson says safety concerns preclude testing during heavy rains. Nor has it tackled big, open parking lots or multilevel garages. ... Pedestrians are detected simply as moving, column-shaped blurs of pixels — meaning, Urmson agrees, that the car wouldn't be able to spot a police officer at the side of the road frantically waving for traffic to stop."Paul continues, 'A deeper issue I wrote about in 2001 is whether such software and data will be FOSS or proprietary? As I wrote there: "We are about to see the emergence of companies licensing that publicly funded software and selling modified versions of such software as proprietary products. There will eventually be hundreds or thousands of paid automotive software engineers working on such software no matter how it is funded, because there will be great value in having such self-driving vehicles given the result of America's horrendous urban planning policies leaving the car as generally the most efficient means of transport in the suburb. The question is, will the results of the work be open for inspection and contribution by the public? Essentially, will those engineers and their employers be "owners" of the software, or will they instead be "stewards" of a larger free and open community development process?"'

Sorry! There are no comments related to the filter you selected.

can it get me home from the bar? (5, Funny)

thebeastofbaystreet (3805781) | about 2 months ago | (#47791317)

I will only buy a Google pod or whatever they're going to call it when it can safely and legally get me home from a night of alcoholic excess.

Re:can it get me home from the bar? (1)

Anonymous Coward | about 2 months ago | (#47791411)

I think I'll just call a cab instead. Unless you're at the bar every single night getting sloshed (in which case you have other problems), I'm pretty sure my way will be cheaper for a long, long time to come.

Re:can it get me home from the bar? (5, Funny)

Anonymous Coward | about 2 months ago | (#47791439)

Unless you're at the bar every single night getting sloshed (in which case you have other problems)

Maybe he's British.

Re:can it get me home from the bar? (5, Funny)

Anonymous Coward | about 2 months ago | (#47791515)

Wouldn't that be covered by the "in which case you have other problems" part?

Re:can it get me home from the bar? (1)

davester666 (731373) | about 2 months ago | (#47791887)

No he's Irish. That's why they are going for independence. The Brit's can't do it every day, and the Irish don't want to hang out with non-hard core people.

Re:can it get me home from the bar? (1)

dylan_- (1661) | about 2 months ago | (#47792041)

You're a bit confused. In fact, this is you [youtube.com]

Re:can it get me home from the bar? (1)

mark-t (151149) | about 2 months ago | (#47791443)

So you won't, in other words.

I mean sure... they'll eventually get to that point, but the number of years of statistics they are going to have to collect on millions of driverless cars driving, collectively travelling billions or even trillions of miles to determine exactly how much safer they are than cars that utilize human drivers is going to be enough that most people alive today probably won't see it happen.

Re:can it get me home from the bar? (1)

Noah Haders (3621429) | about 2 months ago | (#47791555)

the whole conversation about google cars shows how selfish people are. the entire conversation about safety is really about safety of the occupant. it's clear from the summary that they are dangerous to pedestrians. how do they handle bicyclists? google cars should never be allowed on the road.

Re:can it get me home from the bar? (1)

stormpunk (515019) | about 2 months ago | (#47791655)

They handle them fine, detecting when you use hand signals to indicate intentions, assuming you're one of the few people that remember those exist.

Re:can it get me home from the bar? (1)

Noah Haders (3621429) | about 2 months ago | (#47791701)

i want to see proof that they handle them fine. and what about hand signals like go around, fuck off, stop, slow down, what are you doing, I'm walking here. there are a dozen ways to read ped body language than these semaphore official hand signals.

Re:can it get me home from the bar? (5, Funny)

stormpunk (515019) | about 2 months ago | (#47791881)

Ok, you got me. I'm part of the conspiracy. Google cars have not only hit but killed nearly 80% of people that are unlucky enough to wander into the 10 foot radius kill zone. Nobody wanted to speak out against Google because we were afraid they'd divulge our search histories and revoke access to youtube.

Re:can it get me home from the bar? (2, Insightful)

Noah Haders (3621429) | about 2 months ago | (#47791951)

I'm not saying you're part of the conspiracy. I do think a lot of the excitement for google cars comes from the "privileged white driver" mindset in which there are no pedestrians, no bikes, no transit. Nothing but people like them in their single occupancy vehicles.

Re:can it get me home from the bar? (3, Informative)

stormpunk (515019) | about 2 months ago | (#47791963)

https://www.youtube.com/watch?... [youtube.com]
There are some other videos but this one was the neatest one I think. About 1:30 in it shows a cyclist.

Re:can it get me home from the bar? (3, Insightful)

geekoid (135745) | about 2 months ago | (#47791729)

You'r post tells everyone everything about you: You are a pretentious hater.

" google cars should never be allowed on the road."
Not: " until they can reliable detect bicycles, driver-less cars shouldn't be sold to the public.

Have you contact Mercedes to tell them they need to stop selling there cars that can automatically follow the car in front of them? do you rally against self parking cars?

I'm sure you ancestors railed against fire.

Re:can it get me home from the bar? (0)

Anonymous Coward | about 2 months ago | (#47791753)

you'r ???? YOU'R???? YOU'R???

What the hell is wrong with your brain wiring?

Re:can it get me home from the bar? (2)

Noah Haders (3621429) | about 2 months ago | (#47791937)

Have you contact Mercedes to tell them they need to stop selling there cars that can automatically follow the car in front of them? do you rally against self parking cars?

Car following happens at freeway speeds. There are no peds or bikes there. Self parking happens at 2 mph. In both cases there is a driver behind the wheel. Methinks y'our being disingenuous.

Re:can it get me home from the bar? (-1, Flamebait)

nospam007 (722110) | about 2 months ago | (#47791859)

"how do they handle bicyclists? google cars should never be allowed on the road."

Human driven cars killed 726 bicyclists in 2012, Google cars none.

http://www.pedbikeinfo.org/dat... [pedbikeinfo.org]

Re:can it get me home from the bar? (-1)

Noah Haders (3621429) | about 2 months ago | (#47791943)

Also, google cars drove 0 miles in open traffic. Troll harder next time.

Re:can it get me home from the bar? (3, Informative)

Cyberax (705495) | about 2 months ago | (#47792061)

Google cars handle bicyclists and pedestrians just fine. They even understand the bicycle-style hand 'turn signals'.

Re:can it get me home from the bar? (1)

ShanghaiBill (739463) | about 2 months ago | (#47791599)

... to determine exactly how much safer they are than cars that utilize human drivers

They don't have to determine exactly how much safer they are. They just have to determine that they are safer. Also, safety is just one of the benefits of SDCs. Other benefits include better road utilization, since they can drive much closer together. So highways won't need as many lanes. There will be economic benefits as fewer people need to own a car, since driverless taxis will be much cheaper, resulting in fewer cars but also fewer and smaller parking lots.

Re:can it get me home from the bar? (1)

mark-t (151149) | about 2 months ago | (#47791657)

True... but they have to demonstrate that they are safer by a margin that is large enough to be statistically significant, which is why how much safer they would be actually matters.

Re:can it get me home from the bar? (2)

geekoid (135745) | about 2 months ago | (#47791737)

Why can't the be just as safe?

Re:can it get me home from the bar? (1)

mark-t (151149) | about 2 months ago | (#47791931)

If that's all it could offer, then what would be the point?

It probably can. (4, Insightful)

khasim (1285) | about 2 months ago | (#47791473)

Judging by how badly TFA was written.

If a new stop light appeared overnight, for example, the car wouldn't know to obey it.

Got it. So the cars cannot handle changes in traffic markers.

Google's cars can detect and respond to stop signs that aren't on its map, a feature that was introduced to deal with temporary signs used at construction sites.

So they cannot deal with new stop LIGHTS but they can deal with new stop SIGNS. WTF?

But in a complex situation like at an unmapped four-way stop the car might fall back to slow, extra cautious driving to avoid making a mistake.

And it would be "unmapped" for the first attempt. Right? Because the cars should be sending back data on road conditions and such to HQ. Right?

Maps have so far been prepared for only a few thousand miles of roadway, but achieving Google's vision will require maintaining a constantly updating map of the nation's millions of miles of roads and driveways.

And the car needs the map to drive, right?

Google's cars have safely driven more than 700,000 miles.

So they just drove over the same "few thousand miles of roadway" again and again and again and again? Until they got to 700,000 miles?

The car's sensors can't tell if a road obstacle is a rock or a crumpled piece of paper, so the car will try to drive around either.

As it should. Because you don't know if that piece of paper is covering a rock or a pothole or whatever.

For example, John Leonard, an MIT expert on autonomous driving, says he wonders about scenarios that may be beyond the capabilities of current sensors, such as making a left turn into a high-speed stream of oncoming traffic.

Isn't that one of the easier problems? The car waits until it detects a gap of X size where X is dependent upon the speed of oncoming vehicles and the distance it needs to cross PLUS a pre-set "safety margin".

Re:It probably can. (1)

Anonymous Coward | about 2 months ago | (#47791715)

All the data gets sent back to the Google Plex for processing and updates back to the cars, and the best part is they can generate ads using this informtion. Mobile billboards in your car. The police will no longer need to attach a GPS to your car, they can simply ask Google.

It probably can. (0)

Anonymous Coward | about 2 months ago | (#47791901)

"Isn't that one of the easier problems? The car waits until it detects a gap of X size where X is dependent upon the speed of oncoming vehicles and the distance it needs to cross PLUS a pre-set "safety margin"."

You've just excluded a huge part of the world. Where I live the right of way is not a right but something you have to fight to take. Google car would be waiting for years trying to make the simplest turn or merge.

Damn! (3, Funny)

nospam007 (722110) | about 2 months ago | (#47791325)

The Google car has to be shown how to get to the garage, on your property, behind your house.

But seriously, if they'd known the way already, some people would have a heart attack.

GoogleCar: Please select the destination:

A. Before the garage where you cook your meth?
B. Before the garage where you distill your moonshine?
C. Before the garage where you grow your weed?

It's almost... (0)

Anonymous Coward | about 2 months ago | (#47791333)

It's almost as if it's a research model and not intended for mass-market use. It's the car equivalent of an early alpha, get a grip. If you RTFA, they're not even selling it yet.

It's almost... (0)

Anonymous Coward | about 2 months ago | (#47791363)

No, homez, this isn't anywhere near "early alpha" analogy. This is like saying you're well on your way to producing a written a web server, when in fact what you've built is something which can deliver a single web page to a single client at once, and requires editing of configuration files to deliver another page.

The Not-Google-But-They-Slapped-Their-Brand-Name-On-It Self Driving Car is more a self-driving tram (where the rails are clear road markers) with basic collision avoidance. I will not be looking forward to when they're licensed to run in the UK. Although to be honest if I saw one I'd probably step out in front of it, because the earlier this dangerous project is aborted the better.

Re:It's almost... (2)

MisterSquid (231834) | about 2 months ago | (#47791387)

No, homez, this isn't anywhere near "early alpha" analogy. This is like saying you're well on your way to producing a written a web server, when in fact what you've built is something which can deliver a single web page to a single client at once, and requires editing of configuration files to deliver another page.

I'm having a hard time understanding comparisons to web servers and a trams. Could you use a car analogy instead?

Re: It's almost... (0)

Anonymous Coward | about 2 months ago | (#47791403)

I want to know how many football fields it can travel without a collision.

Re:It's almost... (1)

mysidia (191772) | about 2 months ago | (#47791685)

can deliver a single web page to a single client at once, and requires editing of configuration files to deliver another page.

Kind of like the early Gopher servers?

This is why you need MANY girls (-1)

Anonymous Coward | about 2 months ago | (#47791337)

So at any time at least a few are available to take your drunken ass where ever you need to be.

Re:This is why you need MANY girls (0)

Anonymous Coward | about 2 months ago | (#47791371)

Where I live girls get drunk, too.

Re: This is why you need MANY girls (2)

Anonymous Coward | about 2 months ago | (#47791415)

They are the best kind of girls

Re:This is why you need MANY girls (1)

Noah Haders (3621429) | about 2 months ago | (#47791639)

just take uber

columns of pixels? wrong. (2, Informative)

Anonymous Coward | about 2 months ago | (#47791345)

I just played with one of these at the California Academy of Sciences, and waving at it was one of the things I did to see whether the visual representation of the lidar's output was real. It had no problem detecting that I was waving, or the movement of individuals in the crowd around me.

Re:columns of pixels? wrong. (1)

guruevi (827432) | about 2 months ago | (#47791503)

But you waving vs. a cop waving is a big difference. Otherwise it would be a very effective DoS.

Re:columns of pixels? wrong. (1, Troll)

Noah Haders (3621429) | about 2 months ago | (#47791605)

i am seriously concerned about how the google car treats bicycles. can it determine a bicyclist's intention from his hand signals? how about from where it looks on the road and if he's looking over his shoulder? google cars should be banned from shared access roads. dedicated right-of-ways and private property only.

Re:columns of pixels? wrong. (2, Insightful)

geekoid (135745) | about 2 months ago | (#47791759)

Well, you start to turn, the next millisecond the car detects something is moving in front of it and slows. Far faster then any current driver.

You are just another object. The fact that you are on a bike getting in everyone's way is irrelevant.

Re:columns of pixels? wrong. (0)

Noah Haders (3621429) | about 2 months ago | (#47791889)

Except I'm moving in a particular direction and with a particular trajectory, which a human driver can interpret. Can google car?

Re:columns of pixels? wrong. (3, Insightful)

LordLimecat (1103839) | about 2 months ago | (#47792009)

Uh, yes.

That sort of thing is trivial for computers as its basically a simple physics question; whats not trivial is predicting behavior. The point is that a GoogleCar probably wouldnt need to predict behavior in the same sort of way.

People are acting like a googlecar needs to have the exact same senses and responses as a human driver, which is not true; it doesnt have the same limitations (field of view, ~200ms minimum reaction speed, distractions, imperfect data from car) so it can operate differently.

For instance, a person driving a car on an icy winter night has all sorts of unknowns to deal with, between limitied vision, glare from ice / oncoming traffic, not knowing how slippery the roads are, etc. An automated car will have much better vision, a better sense of how well the tires are gripping, and wont be affected by glare. Saying "how will the car know if theres snow in the forecast" is completely missing the point.

Baby steps (5, Insightful)

Anonymous Coward | about 2 months ago | (#47791365)

I have worked 20 years for a major auto OEM. Every time this site runs a Google car article (and there are too many) I cringe.

The first autonomous vehicles will only operate on controlled access expressways, and upon exiting there will be areas where the driver will have to take over or the vehicles will stop.

It will be decades before these vehicles can handle real life situations. You will need AI that can improvise as well as a human. Good luck with that.

Re:Baby steps (1)

MrL0G1C (867445) | about 2 months ago | (#47791509)

With a car 'improvise' is generally stop or swerve or avoid, it's not like these cars have to win at chess.

Google painted an overly rosy picture before, I think the summary has gone the other way.

I hope they do get these cars working to a sale-able level, they could change the face of our cities, driveways could become gardens again, the cars could park themselves in secured parking (underground), act as autonomous taxis etc. residential roads could narrow down to 1 lane + 1 cycle lane.

It's a shame the lack of creativity amongst urban planners and architects... (I live in Britain, this doesn't apply to all countries.)

Re:Baby steps (1)

number17 (952777) | about 2 months ago | (#47791783)

With a car 'improvise' is generally stop or swerve or avoid, it's not like these cars have to win at chess.

Stop or swerve to avoid doesn't resolve driving on snow covered roads. All they've dealt with so far is directional driving. Nobody has mentioned anything to do with traction. So far other companies technology involves winter tires and transmission modes to prevent slipping when accelerating. This this car requires an AI than can improvise. It has to know that the hill you are about to go up is snow covered which means you have to gun it and not stop, otherwise you will get stuck. Its the type of things they don't teach you on the driving test.

Re:Baby steps (1)

LordLimecat (1103839) | about 2 months ago | (#47792027)

Stop or swerve to avoid doesn't resolve driving on snow covered roads
The car will know way better than you ever could how well the car is gripping at any particular moment.

People keep missing that the cars dont need to "know" things or "improvise"; they will have way better data than the human driver in most circumstances and far better reaction times. "Improvise" is somewhat of an oxymoron / bad usage anyways; computers dont "improvise", they follow a structured set of rules, and will always do so until we create a strong AI (which will never happen IMO). The thing is, if you come up with a good ruleset, theres no need to improvise.

Re:Baby steps (1)

Anonymous Coward | about 2 months ago | (#47792077)

With a car 'improvise' is generally stop or swerve or avoid, it's not like these cars have to win at chess.

Stop or swerve to avoid doesn't resolve driving on snow covered roads. All they've dealt with so far is directional driving. Nobody has mentioned anything to do with traction. So far other companies technology involves winter tires and transmission modes to prevent slipping when accelerating. This this car requires an AI than can improvise. It has to know that the hill you are about to go up is snow covered which means you have to gun it and not stop, otherwise you will get stuck. Its the type of things they don't teach you on the driving test.

The traction control of many cars can't even handle snow and slush that well without freaking out and locking down the throttle at the slightest bit of wheel-spin. Last winter I had to drive with it turned off much of the time to avoid getting stuck at intersections when it was snowing. I heard so many complaints from other people around the office about the same issue.

Re:Baby steps (5, Insightful)

Solandri (704621) | about 2 months ago | (#47791549)

Put another way, if autonomous cars started off working on 0% of roads and you want them to eventually work on 100% of roads, well somewhere in between you have to pass through 1%, 5%, 10%, 25%, 50%, 75%, and 90%. It's rather disingenuous to criticize them for not getting all the way to 100% in one fell swoop. I'm shopping for a new car right now, and the new autonomous-like features like adaptive cruise control, lane change assist, and parking assist are really nice (haven't gotten to play with lane departure warning or assist yet). By themselves, no they don't make a 100% autonomous car. But each gets you a small fraction of the way there.

It will be decades before these vehicles can handle real life situations. You will need AI that can improvise as well as a human. Good luck with that.

I see that problem mostly being attacked from the opposite direction. With cars getting radar and proximity sensors, and being able to electronically communicate their intent with each other before actually moving, you reduce the need for the AI to improvise. If an autonomous car wants to pull in front of your car, the two car AIs will communicate it with each other and work out a plan to make it happen before changing lanes. No improvisation required. Sure you might get the stray deer hopping through traffic that requires a human to take control and improvise. But the vast majority of improvisation situations can be eliminated before they ever happen with better communication. That is after all the whole idea behind brake lights and turn signals - to allow you to communicate your intent to the drivers behind/beside you so they don't have to improvise in response to your sudden moves.

Re:Baby steps (0)

Mr. Slippery (47854) | about 2 months ago | (#47791845)

It's rather disingenuous to criticize them for not getting all the way to 100% in one fell swoop.

No it's not, not when they themselves are talking about getting to 100% in one swell foop [nytimes.com] , about building cars with no steering or brake controls.

Re:Baby steps (2, Insightful)

jeffb (2.718) (1189693) | about 2 months ago | (#47791823)

It will be decades before these vehicles can handle real life situations. You will need AI that can improvise as well as a human. Good luck with that.

I'm sure that there will always be a few situations where a skilled human driver will make better decisions, and produce better outcomes, than standard automation.

I'm equally sure that there will be exponentially more situations where standard automation will make better decisions, and produce better outcomes, than average (or even well above-average) human drivers.

I'm sorry, but "there will always be situations where a human performs better than AI" sounds an awful lot like "I won't wear a seat belt because it might trap me in a burning car". It's not wrong, but it is foolish, and it's a poor decision.

Re:Baby steps (0)

Anonymous Coward | about 2 months ago | (#47791983)

I am not trusting my life to a half-baked algorithm which can't handle a snowfall, ice, or heavy rain, or open manhole cover. I will take my own chances.

As far as the "gaping pothole" goes... (0)

Anonymous Coward | about 2 months ago | (#47791369)

I know a LOT of people that do drive over them.

So, I see no problem with that.

"nearly nothing about parking"??? There are PEOPLE that can't park - just look at any shopping parking lot and you will find a lot of vehicles that aren't parked between the lines... taking up two or more parking spaces. And yet, some of the advertisements for cars are now for "self parking" ability so that the driver doesn't have to.

No problem there either.

As for the rest... DARPA has already given contests (which have been won) about driving without a road map.

No problem there either.

Re:As far as the "gaping pothole" goes... (1)

jklovanc (1603149) | about 2 months ago | (#47791771)

I know a LOT of people that do drive over them.

Those people are called idiots. Designs should not imitate idiots

"nearly nothing about parking"??? There are PEOPLE that can't park - just look at any shopping parking lot and you will find a lot of vehicles that aren't parked between the lines... taking up two or more parking spaces. And yet, some of the advertisements for cars are now for "self parking" ability so that the driver doesn't have to.

Again, idiots. Parking like an idiot is still wrong. If every car did it there would be big problems. Also parking assist is designed for parallel parking between two vehicles. It is a well defined problem with a vehicle in front, a vehicle behind and curb on one side. An open parking lot on the other hand much more difficult. Unless the car can decipher the markings on the pavement it has no clue where to park. Sure a car could find a spot between to other cars but what if that "spot" is actually lane between isles? What if that spot is actuall the open space between two vehicles parked in handicap spots? What if there are few cars in the lot. Can it tell the difference between a regular spot and a restricted (handicap, loading zone, parent with kids, etc) spot?

No problem there either.

As for the rest... DARPA has already given contests (which have been won) about driving without a road map.

References? If you are talking about the DARPA Grand Challenges [wikipedia.org] noe of the winning technology was even close to commercially viable. They were proof of concept at best.

Stop being so impatient.... (5, Insightful)

seanvaandering (604658) | about 2 months ago | (#47791375)

The technology is in it's infancy stages. Why the media keeps hounding Google on all these issues seems immature. I don't see any other competing company attempting to do the same thing, and if there is, they are definitely staying clear of the media spotlight.

I see Google making some great progress in this area, but give it time people - they will work out the kinks, but it won't be done in year.. lets realistically say that maybe in 5-10 years from now we might fathom the idea that the car is safe enough for whatever weather and situations we can throw at it.

Re: Stop being so impatient.... (0)

Anonymous Coward | about 2 months ago | (#47791435)

Google is a marketing whore. If I was CEO of a company that was in this space (and I might just be one) then I'd stay very clear of the press and build a great team of engineers that can make it happen in complete secrecy. I'd only release that knowledge once we had something that actually could be useful.

Re: Stop being so impatient.... (1)

Noah Haders (3621429) | about 2 months ago | (#47791625)

yeah but this whole car effort is one big marketing blitz. they're not in the car business. they won't make a profit from the cars. they'll play with it right now because it feeds their image, but they'll drop it as soon as it becomes inconvenient.

Re: Stop being so impatient.... (0)

Anonymous Coward | about 2 months ago | (#47791697)

they're not in the car business.

For now, yes. But they haven't been in browser business as well.

they won't make a profit from the cars.

Why are you so sure about that? Google is putting quite a lot of effort into robotics, what exactly is stopping them from becoming a leader in this field?

Re:Stop being so impatient.... (2, Insightful)

jklovanc (1603149) | about 2 months ago | (#47791557)

Why the media keeps hounding Google on all these issues seems immature.

It is to counter Google's skewed data that make it look like autonomous cars are just around the corner. For example, why come out with a vehicle that has no steering wheel if it is not viable for another 5-10 years (by your estimate)? Do you ever see a Google press release mention any of these limitations? All you hear from Google is a rising tally of miles driven and the fact that there have been no accidents. The fact that the miles are driven on carefully selected, heavily scanned roads under optimal conditions never seems to make it into the reports. Driving down the same roads thousands of times is not progress.

Re:Stop being so impatient.... (5, Insightful)

geekoid (135745) | about 2 months ago | (#47791817)

"It is to counter Google's skewed data that make it look like autonomous cars are just around the corner."
Google has never said that. And this guy doesn't have all the data, nor does he know whats in development.

"why come out with a vehicle that has no steering wheel if it is not viable for another 5-10 years (by your estimate)?"
The same reason worlds fair showed tech that will be coming out in 5-10 years. Its' fun, it's cool. It also show they are thinking long term and not quarterly. It also shows a company spending money on RnD.
I consider all of that a good thing.

"Do you ever see a Google press release mention any of these limitations?"
Yes.
http://googleblog.blogspot.com... [blogspot.com]

" All you hear from Google is a rising tally of miles driven and the fact that there have been no accidents. "
Which is pretty important.

"The fact that the miles are driven on carefully selected, heavily scanned roads under optimal conditions never seems to make it into the reports."
That is the smart way to start, but they are moving past that.

" Driving down the same roads thousands of times is not progress."
Of course it is. Same roads, different traffic. The same rods can have 10's of thousands of changing variables at any given time.
The team members are using them. A team member took one from Google campus to Tahoe on a trip.

Do you lay awake at night just trying to think of ways to hate cool new things?

Re:Stop being so impatient.... (1)

jklovanc (1603149) | about 2 months ago | (#47791955)

he same reason worlds fair showed tech that will be coming out in 5-10 years.

I think your 5-10 year estimate is ludicrously optimistic. Until the vehicle can classify what a person is doing on the side of the road it is not a viable solution. That person could be a statue, a child who could dart into the road, an person standing safely on the side, a police officer pulling the car over, etc. That kind of AI is much further down the road. By the way, that is one reason for the pre-scanning. The vehicle scanner can not tell the difference between a mailbox and a person let alone predict identify what that person is doing.

"Do you ever see a Google press release mention any of these limitations?"
Yes.
http://googleblog.blogspot.com... [googleblog.blogspot.com] [blogspot.com]

Notice that they never mention pre-scanning roads.

That is the smart way to start, but they are moving past that.

References please. I have never heard of any test under adverse conditions.

A team member took one from Google campus to Tahoe on a trip.

After the route was pre-scanned.

Do you lay awake at night just trying to think of ways to hate cool new things?

No, I just hate hype.

Stop being so impatient.... (0)

Anonymous Coward | about 2 months ago | (#47791577)

Google's autonomous car technology is in its infancy. There have been other company's doing this for many many years with more advanced code / refined. The company I used to work for was able to detect negative obstacles (holes) water and humans. Their technology also was able to operate without a predefined map unlike Google's car.

Re:Stop being so impatient.... (-1)

Anonymous Coward | about 2 months ago | (#47791683)

"technology is in it's infancy stages"

So is your grasp of the apostrophe.

Re:Stop being so impatient.... (1)

TubeSteak (669689) | about 2 months ago | (#47791735)

The technology is in it's infancy stages. Why the media keeps hounding Google on all these issues seems immature.

Google successfully lobbied California for new laws regarding autonomous cars and they keep putting out press releases.
Google put themselves in the spotlight.

only if intricate preparations have been made beforehand, with the carâ(TM)s exact route, including driveways, extensively mapped. Data from multiple passes by a special sensor vehicle must later be pored over, meter by meter, by both computers and humans.

That's more prep than a rally driver gets before he barrels down a 1-lane dirt road at highway speeds.
That's certainly not what Google has been selling the public and State governments.

Re:Stop being so impatient.... (0)

Anonymous Coward | about 2 months ago | (#47791751)

If you don't see a competing company, then you're not looking hard enough:

http://www.ultraglobalprt.com/ [ultraglobalprt.com]

Self-driving car research is also pretty big in Germany, which is in part why Sebastian Thrun is a lot less of an impressive personnage there.

Re:Stop being so impatient.... (1)

geekoid (135745) | about 2 months ago | (#47791763)

Google is big.
Google general does good things
So people who make money from hating need to find something, anything to get clicks from google hate.

Needs more infrastructure (1)

spire3661 (1038968) | about 2 months ago | (#47791379)

To get truly self-driving cars there needs to be a lot more tech in place. We need to develop a Roadmaster computer system that 'controls' all the cars by feeding them road parameters. WE need vehicle to vehicle(V2V) comms as well. The idea that self-driving cars are going to happen without beacons and roadmaster computers and V2V is silly. All the preparations that go into the google car to make it work needs to be baked into the road system to make it work en masse. We have long way to go, and google's car is a tiny baby step. In no way should be it considered ANYTHING but a pure research platform.

Re:Needs more infrastructure (1)

kheldan (1460303) | about 2 months ago | (#47791575)

more infrastructure

Sure. Just like broadband Internet, right? Because everyone in Kansas has broadband now, right? Oh, wait, that's right, they don't! There are suburbs I've lived in where there was nothing but dialup..

The fact of the matter is: The entire highway infrastructure will have to be completely overhauled, at enormous expense, before 'self driving cars' could be a reality, or even have a chance at being prevalent. Even then there will always be roads and places where there is no such infrastructure, and people will have to operate the vehicle themselves to get where they want to go. Also, what do you do about situations where you're not even sure where you're going? At the very least, the puzzle of real, full-on human-level artificial intelligence has to be solved, I think, before anything like this can be practical. There are too many variables, you can't program an 'expert system' for all of them, something will come up it doesn't know how to handle.

I thought of something else that will be a problem with 'technology' like this, with regards to criminal activity. I've made the argument before that hacking a system like this will be as easy as hacking any other computer system, and criminals will use it to steal vehicles, or to carjack/kidnap people. Imagine this scenario: All you'd need are two or three 'attacking' vehicles, to surround and restrict an autonomous vehicle in such a way that is has no choice but to slow down, in order to avoid what it perceives as an imminent collision. From there it becomes a trivial matter to continue slowing down, and forcing the vehicle to pull over and come to a complete stop. From there extracting the passenger(s) becomes exceedingly trivial, and voila, you've either kidnapped the passenger, stolen the vehicle, or both. There's no way they're going to code an 'evasive maneuvers' routine into the vehicle, or even come up with a reliable way to have it detect that it's being gamed like this. Of course someone will say 'Nobody is going to do this!' or 'That'll never happen!', or 'All cars will be autonomous so it won't be possible', but that's as complete a fantasy as saying (in some people's dystopian future) 'guns are illegal so where would anyone get a gun?'. At best I think we'll have an advanced 'autopilot' system, as an option, of course, but people will still be driving their own cars, at least for the next, say, 50 years or so.

Re:Needs more infrastructure (1)

geekoid (135745) | about 2 months ago | (#47791855)

" entire highway infrastructure will have to be completely overhauled, at enormous expense, before 'self driving cars' could be a reality"
since they are using them everyday, and taking them on trip in CA, on normal roads, I don't think you are correct.
Of course are road infrastructure could use a few smart changes anyways.

It's trivial to kidnapped some in a car today.

"There's no way they're going to code an 'evasive maneuvers'"
Oh, I see. You think you would be able to do some Die Hard esque driving to get away from kidnappers.

Re:Needs more infrastructure (0)

Anonymous Coward | about 2 months ago | (#47791863)

You're thinking way too much. Why bother with blocking the car in, when you can just stand at the side of the road with a construction worker's 'STOP' sign? The car will stop, and you just take the passengers at your leisure.

Re:Needs more infrastructure (1)

Noah Haders (3621429) | about 2 months ago | (#47791637)

there's a lot of people on the road other than cars. pedestrians, bicyclists. are you going to get a roadmaster computer to tell them what to do as well? some pedestrians are elderly. some are drunk. some are children. i don't know where you're from but in america pedestrians and bicyclists have the right of way.

If get drunk... (1)

Lisias (447563) | about 2 months ago | (#47791399)

If I get drunk, and my Google car drives me home and overrun a red light or something like that. Who the cops should blame? Who will get the ticket?

Re:If get drunk... (1)

itzly (3699663) | about 2 months ago | (#47791501)

Simple. The owner of the car will get the ticket.

Re:If get drunk... (0)

Anonymous Coward | about 2 months ago | (#47791813)

And a DUI for not overriding the car in time to prevent running the red light.

Re:If get drunk... (0)

Anonymous Coward | about 2 months ago | (#47791877)

and the DUI

Still useful (1)

Kohath (38547) | about 2 months ago | (#47791417)

Even if it can only self-drive on a routes I've already driven manually, it would still be extremely useful.

Agreed; incremental versions can be useful (1)

Paul Fernhout (109597) | about 2 months ago | (#47791839)

I agree (article submitter here). I submitted the article mostly not to complain about lack of progress but because the article covered a lot of interesting details about how the Google technology worked in discussing the limits of the current system. I have little doubt such systems will continue to rapidly improve.

I was involved briefly on a project for self-driving cars in the late 1980s at Princeton involving neural network ideas for image processing, and I suggested we could just train the cars to drive specific routes. However, that suggestion was scoffed at (and I did not try hard to push it). My argument was that most driving is stuff like daily commutes or runs to well known stores, and so pretty much the car could drive exactly the same way every time, seeing the exact same sights. That might make it feasible to train the neural networks from just a few video recordings of drives over the same stretch of roadway. Granted, lighting conditions, weather, other cars, pedestrians, and possible lane changes make that harder -- but is seemed like a good place to start, rather than try to create a car driving system that could drive in arbitrary new circumstances where it has never seen the road before. Solar panels have succeeded much the same way -- the early ones were niche (like in calculators or satellites), but sales drove more R&D that lead to better and cheaper panels in more and more applications. A self-driving car that could only drive me from home to a few local towns and back on fixed routes (safely, while, say, I surfed the web) would still be tremendously valuable to me. Think of how many people commute the same routes every day for years and could use that commuting time more productively in other ways via the internet. If people with an hour commute could use that time to answer email, then maybe they could work one hour less in the office? Also, a car that just knew how to park itself in a standard location and come back to pick you up in front of some building you work at or apartment you live in would be very useful in cities.

Another idea I had several years ago is that we could have an open source software effort to drive cars in various simulated racing games like "Gran Turismo" or other free play driving games like "Driver: San Francisco" or various off-road sims. That would be a inexpensive and safe challenge for college students. Those driving simulators go to great lengths to make realistic looking images (including things like dust clouds and vehicle dynamics), and they continue to improve. You just feed the first-person video generated the game into the car-driving visual processing algorithms, and you have the software control the game via USB outputs. As the software gets better, then you can fuzz up the image more and more by adding more white noise to it, or whatever other distortions you wanted (like bug white blotches over parts of the image) to challenge the algorithms. Or you could introduce delays and noise in how commands for steering were processed. Such an approach makes writing such software feasible for the average software developer without a special car. Granted, the software would have to focus on processing 2D images instead of 3D laser ranging data. Even Google has talked about testing their software in simulations regarding certification. Ideally, the simulations used for testing would be open source too, like Rigs of Rods (or even more realistic) and if so, things like 3D ranging data could probably be extracted too: http://www.rigsofrods.com/ [rigsofrods.com]

Hype (1)

LookIntoTheFuture (3480731) | about 2 months ago | (#47791477)

Google often leaves the impression that, as a Google executive once wrote, the cars can “drive anywhere a car can legally drive.”

Wow, that's just a big ol' lie isn't it?

Re:Hype (1)

jklovanc (1603149) | about 2 months ago | (#47791587)

Technically speaking they can. They just have to wait for the road to be surveyed before hand.

Re:Hype (1)

geekoid (135745) | about 2 months ago | (#47791909)

What he wrote was:
"Our self-driving cars have now traveled nearly 200,000 miles on public highways in California and Nevada, 100 percent safely. They have driven from San Francisco to Los Angeles and around Lake Tahoe, and have even descended crooked Lombard Street in San Francisco. They drive anywhere a car can legally drive."

I like how you left out the fact that, clearly, they didn't need special roads.

Glass half-empty much? (1)

wonkey_monkey (2592601) | about 2 months ago | (#47791493)

Bitch, whine, moan. Autonomous cars are a work in progress. Didn't anyone think they weren't?

Would you buy a self-driving car that couldn't drive itself in 99 percent of the country?

Of course not. But then, no-one's selling them yet, because they're STILL DEVELOPING THEM.

Re:Glass half-empty much? (0)

Anonymous Coward | about 2 months ago | (#47791841)

Of course not. But then, no-one's selling them yet, because they're STILL DEVELOPING THEM.

But they're hyping them like they'll be available to buy next year. That's what pisses off those who understand how hard the problem really is.

A Pile Of Money (1)

Jim Sadler (3430529) | about 2 months ago | (#47791607)

In areas with fast growth detailed maps of every inch of the roads will require an accumulation of data almost hourly. For a specific map to be updated hour by hour may offer a billing opportunity for a major data service. Assume your car drives you to work at 8 am.. When it is time to return home the computer will need to know if any alterations or conditions have occurred during the day. Since most of us repeat patterns in daily driving those routes may update themselves while the vehicle is at rest. But any variation in routes would call for some sort of advanced route request so that one could get in the car and be transported without delay. Many sales organisations have craved a new street or new dwelling directory for decades and it has never happened yet. I live in a region in which new roads and homes crop up daily and it often takes maps months before those new streets appear. One can get a list of new building permits from the city or county but there is no map to guide one and the new places may be months from completion. The sales community would love daily mapping of their region.

I actually don't see a problem (5, Insightful)

whistlingtony (691548) | about 2 months ago | (#47791611)

I'm going to map my drive to work, by driving it a few dozen times. Then the car can take over. I don't care if it's no good in parking garages or my own driveway. I'll spend 3 minutes driving from my house, let the car take over, let the car do the boring freeway driving, and it can alert me when I'm 3 minutes from work. Then I'll take over and get into the parking garage and park my car.

Are we really whining because a brand new technology can't do EVERYTHING for us? Because it only takes care of MOST of the drudgery?

Re:I actually don't see a problem (1)

mysidia (191772) | about 2 months ago | (#47791769)

I'm going to map my drive to work, by driving it a few dozen times. Then the car can take over.

Why not provide a mechanism so that when you drive over the road, data is sent to Google. Once enough people have driven the road, the data gets shared with all the vehicles, so it's as if they all have the latest map of the road.

Also, there could be some mechanism designed for the DOTs, Police and construction workers, to "FLAG" a position and broadcast based on GPS coordinates as "X event now", OR "X event next Friday from 12pm to 4pm", "Emergency", "Danger STOP/SLOW", "Lane Closure", "Lane Detour", "Flagman", or "Change Made", "Signal Removed", "New Signal Added", "Stop Added", "Intersection Added", and specify a number miles radius, so the first approaching vehicles to encounter will automatically approach with extra caution, until they have collected sufficient new data

Re:I actually don't see a problem (1)

jklovanc (1603149) | about 2 months ago | (#47791801)

Why not provide a mechanism so that when you drive over the road, data is sent to Google.

One word; bandwidth. Do you have any idea the amount of data there is in a road scan? That data would also need to be analyzed by a person to filter out problems.

Re:I actually don't see a problem (1)

mysidia (191772) | about 2 months ago | (#47791871)

One word; bandwidth. Do you have any idea the amount of data there is in a road scan?

One word:WiFi, or Google WiFi [gigaom.com] and Background incremental data uploads and downloads with Automated analysis.

Re:I actually don't see a problem (1)

jklovanc (1603149) | about 2 months ago | (#47791973)

When you are talking about a couple hundred vehicle trying to transfer gigabits of information WiFi won't cut it. Also, most roads are out of WiFi connectivity anyway.

Re:I actually don't see a problem (1)

mysidia (191772) | about 2 months ago | (#47792049)

Also, most roads are out of WiFi connectivity anyway.

I think you're totally out of touch with what the technology is capable of. After you have logged the trip, you will be within WiFi range at your destination or at other points in time.

Also, it's up to the vehicle to distill that information, so they aren't necessarily uploading Gigabytes.

Even if they were, it's quite doable.

It's also quite possible analysis of the data may be distributed, and different vehicles could share that data with each other, and once many vehicles have the data, they would start seeding the same Torrent.

Re:I actually don't see a problem (1)

jklovanc (1603149) | about 2 months ago | (#47791791)

Issue is not just the scanning but that the scans have to be gone over by a person to decipher them and make them usable by the vehicle. Also, what happens when something changes?

Mod parent up. (1)

khasim (1285) | about 2 months ago | (#47791829)

I'm going to map my drive to work, by driving it a few dozen times.

And that is if you are the ONLY person with a robot car on that road. Which may be correct for the initial roll-out. But this is a great example of the "network effect". If 100 people in your state own robot cars then a LOT of your state will be continuously mapped / re-mapped / re-re-mapped / etc.

Are we really whining because a brand new technology can't do EVERYTHING for us? Because it only takes care of MOST of the drudgery?

There is space to be filled and page hits to be collected. Demanding instant perfection for every edge-case is a good way of doing both.

Google has logged over 700,000 miles in those vehicles. Without a single robot-controlled accident.

There might be problems in certain weather conditions. Or in certain other conditions. Or whatever. In which case the driver should take over.

And since it is software, eventually those problems should be solved.

call me when they're ready (0)

Anonymous Coward | about 2 months ago | (#47791629)

and I'll 3D print my self-driving car at home.

Can't drive in 99 percent of the country. (5, Funny)

Snufu (1049644) | about 2 months ago | (#47791633)

"Would you buy a self-driving car that couldn't drive itself in 99 percent of the country? Or that knew nearly nothing about parking, couldn't be taken out in snow or heavy rain, and would drive straight over a gaping pothole?"

Google is just trying to accurately mimic the real driving population.

been wondering many similar things (1)

cascadingstylesheet (140919) | about 2 months ago | (#47791635)

How's it going to park in a dirt lot? Recognize which spots are likely to be too muddy? How's it going to park behind the barn?

Don't some of the breathless stories talk about whether or not these cars will need steering wheels? What's going to happen when the steering wheel-less models meet something they can't deal with? Blue screen of death?

Re:been wondering many similar things (1)

geekoid (135745) | about 2 months ago | (#47792081)

look at another parked car and park next to it?
How do you know where to park?

Extensively mapped? (0)

Anonymous Coward | about 2 months ago | (#47791641)

Google often leaves the impression that, as a Google executive once wrote, the cars can 'drive anywhere a car can legally drive.' However, that's true only if intricate preparations have been made beforehand, with the car's exact route, including driveways, extensively mapped.

Sounds like a bunch of FUD.

I can go to my provincial (Canada) government website and download the entire traffic map, with exact distances (vector drawn to scale) and including every entry to every driveway, be that city or rural. Where do you think Google get's its map data?

Avoiding potholes, sinkholes, obstructions, and other cars is part of the algorithm. If you mean the car will not swerve madly into the other lane, potentially creating a dangerous situation but drive over a hole in the road, well, good!! If it also does not drive over barricades, that's good too.

So yes, bunch of FUD. If you need car to park itself, there is already technology out there TODAY on the roads.

https://www.youtube.com/watch?... [youtube.com]

It's 3 years old already.

You want a car that detects if road surface is more slippery and adjusts traction control and/or warns you? Out there already.

If a car can drive itself 90% of the way, and it requires me to park on a parking lot, what's the big deal? I don't want to drive during the tedious part of the route, where most of the fatal crashes happen because of DUIs or distracted or speeding/dangerous drivers.

Re:Extensively mapped? (2)

jklovanc (1603149) | about 2 months ago | (#47791891)

I can go to my provincial (Canada) government website and download the entire traffic map,

I have worked in GIS road mapping for years and government geocoded maps do not have driveways marked. Google has access to that type of data yet they still need to pre-scan roads. Also the mapping for the autonomous car is not just about the road. It also maps things close to the road. For example, that blob of pixels may be a person or it may be a mailbox. They are treated differently by the vehicle.

If you mean the car will not swerve madly into the other lane,...

No we mean moving a couple of feet within the lane so the sidewall does not get ripped out causing an accident.

If you need car to park itself, there is already technology out there TODAY on the roads.

Sure it works for parallel parking but that is only one kind of parking. That algorithm does not work for angle parking or side by side parking.

If a car can drive itself 90% of the way,

Google is not even close to that yet they make it seem they are.

Right (0)

Anonymous Coward | about 2 months ago | (#47791665)

So, we are just as far from self-driving cars as we were 50 years ago.

No worries (1)

CODiNE (27417) | about 2 months ago | (#47791935)

They'll have a nice big BETA on the sides and the public will be very understanding of little bugs here and there.

Of course manufacturers will need a little bit of time to integrate their value-added enhancements so you may want to wait for the Nexus cars for trouble-free firmware updates. Or if your model can't be updated simply get a new car every 2 years.

Re:No worries (0)

Anonymous Coward | about 2 months ago | (#47791965)

Don't forget that you'll get into the car in the morning to go to work and find it's in the middle of a firmware update and will have to reboot a few times before it will drive anywhere...

self driving car better able to avoid pothole (2)

aepervius (535155) | about 2 months ago | (#47792019)

Look this is a question of cost. Self driving car must be able to detect the end of road or a hole or whatever so if google did not already install a radar or similar to check the road status ahead, it would be incredibly dangerous. And a radar or similar detection device would be far better at detecting :
1) pothole
2) slowdown speedbump
3) whatever the state of the road
better than human say, at night. Or in the fog. Or distracted.
Load More Comments
Slashdot Login

Need an Account?

Forgot your password?