Autonomous Cars Aren't As Smart as They're Cracked Up To Be (computerworld.com)

Gill Pratt, executive technical adviser at Toyota, offers a note of caution, even as more car companies start putting AI elements into their cars. Speaking in Tokyo at the announcement of a Silicon Valley AI research center that Toyota is to open in early 2016, Pratt pointed out the big shortcoming in an AI system as applied to automobile: Autonomous cars might look great in controlled tests or on pristine highways, "but soon fail when faced with tasks that human drivers find simple." From the article: Drivers, for example, can pretty much get behind the wheel of a car and drive it wherever it may be, he said. Autonomous vehicles use GPS and laser imaging sensors to figure out where they are by matching data against a complex map that goes beyond simple roads and includes details down to lane markings. The cars rely on all that data to drive, so they quickly hit problems in areas that haven't been mapped in advance. ... A truly intelligent self-driving car needs artificial intelligence that can figure out where it is even if it has no map or GPS, and manage to navigate highways and follow routes even if there are diversions or changing in lane markings, he said. I regularly drive a stretch of road that's just a few miles long, but between construction, accidents, poor marking, bicycles, and heavy traffic I'd be nervous about letting an AI system navigate. In what real-world driving scenarios would you most want humans to take over?

Comments

That's OK, I only care about bar crawls

[drinkypoo]

As long as the car can drive me home after the last bar in the line, I'm happy.

Re: That's OK, I only care about bar crawls

[t1oracle]

Cheaper? An AI doesn't need a salary, or a medallion, and it's insurance will be cheaper. If taxis were cheap, most of us wouldn't have cars.

Re: That's OK, I only care about bar crawls

[mysidia]

As customer of a Taxi, you don't have to pay any of those things, only a miniscule portion of them.

The AI is more expensive, because all new cars are super-expensive, and you have to add Research and Development costs, "brand premium", And "coolness premium" the manufacturers will charge b/c the thing can drive itself.

If instead of buying a $12,000 used car that meets all your needs, you spend $60,000 on a brand new car-that-can-drive itself and lasts 10 years, plus a $50 monthly service fee for the cloud maps service, then you're paying approximately $5000 extra a year for self-driving capabilities.

That would buy you 333 $15 taxi rides.

Anyways, based on that, unless you spend more than $5000 a year on the Taxi, then it just isn't a worthwhile economic proposition.

Also, the self-driving cars are probably going to be introduced at about the $120,000 price point, not the $60,000 price point.

Then you'll also spend an extra $10,000 in vehicle loan interest per year to get the self-driving feature.

So just have the cars drive where it is easy

[LetterRip]

You don't have to have the car drive everywhere, 95% of the places you drive will probably have all of the factors needed for the car to navigate easily. Just don't have the car drive in areas where it can readily get in trouble.

You don't start teens off in ambiguous hard to drive conditions, but rather low traffic side streets or empty parking lots, etc.

We don't need self driving cars that are perfect from the start, merely good enough to drive us most places most of the time, and do not have accidents in the areas that are suitable for it to drive.

Re:So just have the cars drive where it is easy

[Anonymous Coward]

Don't know where you live, but here they change lanes around, close lanes, and reroute traffic on a nearly daily basis. Its been non-stop doing that for over 10 years straight. If auto driving cars depend on GPS that means I can't use it to just go to work even, not to mention going somewhere else.

What happens when they close part of the city for a parade and a lot of streets are closed? Your car just stops and waits the 4 hours for it to pass before continuing? Sure you could take over, but if your drunk and you do that you committed a DUI, the reason you bought a self driving car to start.

Re:So just have the cars drive where it is easy

[TWX]

Don't know where you live, but here they change lanes around, close lanes, and reroute traffic on a nearly daily basis. Its been non-stop doing that for over 10 years straight. If auto driving cars depend on GPS that means I can't use it to just go to work even, not to mention going somewhere else.

What happens when they close part of the city for a parade and a lot of streets are closed? Your car just stops and waits the 4 hours for it to pass before continuing? Sure you could take over, but if your drunk and you do that you committed a DUI, the reason you bought a self driving car to start.

I expect that in addition to cars being able to self-determine routes and find barriers there will need to be intelligent barriers that the cars can detect and follow the instructions of. These kinds of barriers would be used by construction crews, emergency responders, and perhaps even as a function of the four-way hazards when a car is stopped on the side of the road. Call it a more precise means for the autonomous car to determine what it should do or what the expectation is in a complex situation.

Just as an example, in long-term highway construction projects it's not uncommon to take a two-lane-single-direction stretch of Interstate and to route both directions on it, one going the natural way, the other driving what would normally be opposed, while the other two-lane stretch is being worked on. In cases like this there needs to be a way for the construction barriers themselves to notify the vehicles both that something has overridden the expected behavior, and that this particular path is the override. The car will in-turn have to account for this deviation in the path and to know that it's not actually trying to go the wrong-way even though its default programming would say that it is, and it would have to understand that while one lane is now no longer the wrong, way, the other lane still is the wrong way and to not try to use it.

Other construction-related examples include the ability to follow a pilot car and the ability to pay attention to flag-men. The flag-men method is a variation of the one-lane bridge in many cases with the addition of a very spontaneous control (ie, the switch from slow to stop and stop to slow comes without warning from the flag-man himself, so the vehicle must pay attention to the flow of traffic in addition to somehow figuring out the sign or receiving a signal from the sign), and the nature of pilot cars means that there has to be some means for cars to be subordinate to other vehicles, which leads into the next example...

...emergency responders. Cars will need to respect things like fire trucks blocking the road, or police cars blocking the road, or tow-trucks blocking the road, or any other sort of obstruction that will be present for awhile and indicates that it isn't safe to be within a certain area. Cars may also have to react to barriers placed by these responders, and it may make sense for those barriers to have some kind of component that lets them more intelligently broadcast so that the cars don't have to figure out what they are visually. Obviously if the police are attempting to close a stretch of road due to an accident investigation they want to keep cars out of that area so that the evidence is not disturbed. If firefighters are working on a structure fire they need to keep cars out of the immediate staging area and from driving down the road that the firemen may be crossing regularly without notice. They also need to keep a wide berth when a tow truck driver is working with a disabled vehicle, wherever that vehicle is disabled and whatever is wrong (ie, difference between an overturned vehicle on the highway, a stalled vehicle on the highway, and a stalled vehicle on the median or shoulder). These are all complex situations that happen all of the time, and cars need to be able to handle them.

I think the first application for autonomous cars will be open-highway dr

Two camps

[transporter_ii]

There seem to be two camps of people. Those that think we will be living on mars and have fully autonomous cars in a couple of years, and those that actually look into it and see how hard it is going to be. For some reason, the media seems to prefer the first one. Reality prefers the second one.

Re:That's nothing

[belthize]

I don't see the issue.

For many scenarios where a human will have to fave that decision the autonomous car never will because it would have chosen option C, avoid situation long before it became an issue. Take the Oklahoma parade a few weeks ago, an autonomous car would not have had to decide should I kill the drunk driver or plow into the parade.

But lets pretend that the car has to decide and chooses poorly and wipes out 10 people in a crowd including little kids. How is that a problem if across the set of all autonomous cars 100 such occurrences were avoided. Choices should be made on relative merits, X is better than Y. Not X is perfect or X is imperfect.

Are you trolling or just boring?

[drinkypoo]

People keep saying this, but the truth is that the car is going to [be programmed to] follow the law. That means it's going to approach intersections at safe speeds, and it's going to avoid hitting pedestrians in crosswalks but will simply murderize them even if there's ten of them in your lane, and a cancer-ridden octagenarian driving a yugo in the other lane — even if the car has enough sensors to smell cancer, it's still going to run right into those pedestrians like you've gone bowling rather than deviate from the marked lane. It's going to make a good-faith best effort to stop. But remember, it's not going to go around a blind curve at a speed at which it can't stop if there's an obstacle. It's simply going to decelerate for the curve, and then accelerate again on the other side. If someone is in the road, it won't hit them, because it's not driving for fun. It's driving to minimize risk.

Toyota getting left behind

[monkeyxpress]

Translation: Toyota is woefully behind in autonomous car development, and rather worried about it.

The FUD begins.

Re:Toyota getting left behind

[RR]

Forget autonomous car development, Toyota is woefully behind in computer-controlled car development. Random relevant article: Toyota's killer firmware: Bad design and its consequences.

Re:"It has to be perfect before it'll work"

[Gaygirlie]

Frakly this is BS... I drive a large portion of my day for work (not a trucker, IT guy going to clients.) I run into "diversions or chaning in lane markings" and have to stop and think about what to do at times too! Why should an AI have to understand the intentions of a road worker/civil engineer better than we do before it can be accepted as intelligent?

I'm not seeing anyone in TFA saying it'd have to be better at it than us, just that it'd have to be able to do it at all would be a good start. As things stand autonomous cars are not anywhere near of being capable of doing that on their own.

I will have no clue where I am and will have to basically start driving in one direction (which these cars can do) until I figure out where I am.

No, they can't. That's the whole point here: as long as they rely on GPS and very detailed mappings for navigation they won't be able to do that -- they need to know where they are to be able to start driving at all. The author wasn't saying the car should be able to magically instantly know where it is even when no mappings or GPS was available, just that the car should still be able to try and figure it out -- quite possibly doing the exact thing you suggested and trying to find a roadsign or two. The issue here is that these cars won't know even how to get off the god damn parking lot without GPS and mappings, let alone going out and figuring their own surroundings on their own without some very extensive AI.

Answer

[Anonymous Coward]

So nobody is answering the last question in the post - about when would you want to take over. I say never. The whole point of a self driving car is so that I don't have to watch the road, don't have to pay attention, and can sleep, read, or work while the device drives me like a taxi would. If it can't do that, it is less than worthless. Less than worthless because why would I want to pay for a fancy AI in my car if I have to keep my hands free, near the wheel, and my eyes on the road? I do that NOW. If I have to pay attention and be ready to take over then the thing is just a toy and they can keep it. Once it can actually drive me without using my eyes, ears, hands, and brain as a backup then I might want one.

Re:That's nothing

[Dutch Gun]

decide between plowing into a crowd of people to protect the driver, and smashing into a tree to protect the crowd of people

People hold up ridiculous scenarios like this as some sort of hypothetical metric, but how well would a human do with an insane choice like this, presumably with only a split second to make the decision? Not very well, I'd imagine. Don't put AI up against ridiculous situations. Put them up against realistic obstacles, which we might actually have a chance of seeing in our lifetimes. Road construction. Temporary obstacles with police directing traffic. Blizzards. Temporarily flooded road. Parking lots or garages.

There's also this false dicotomy presented, wherein some people seem to think that unless an AI can can handle ALL situations possible, it can't possibly work. I'll tell you what will happen in many situations. The AI will come to a controlled stop and tell the human "Hey, I don't know what's happening. Take over the wheel, please." That seems perfectly reasonable for crazy scenarios that only rarely occur.

The answer to what would likely happen, by the way, is that the AI in the car would have long ago started braking, so as to avoid the problem in the first place.