Self Driving Car thread

[dh87]

ChrisG[/url]":s07z0mnw]
Actually, I think much of this thread is pure fantasy. Self-driving cars aren't going to start appearing, on-roads, in significant numbers, for many decades yet.

From NHK:

Nissan to test self-drive car on public roads

...

Nissan is hoping to put the self-driving car on sale in 2020.

6 years might be a bit optimistic, but "decades" is very likely wrong.

I think that a huge benefit of s-d cars will be their fuel economy. There's no reason for an s-d car to go from 0 to 60 at any appreciable rate. In fact, if I'm reading the newspaper or taking a nap, I'd prefer 0 to 60 to be as gentle as possible. Hence, engines can all be low-power hybrids or electrics. That's how all the carmakers are planning to meet the 2025 standard of 55mpg fleet average.

 

[dh87]

ChrisG[/url]":1dphmk6b]

dh87[/url]":1dphmk6b]

ChrisG[/url]":1dphmk6b]
Actually, I think much of this thread is pure fantasy. Self-driving cars aren't going to start appearing, on-roads, in significant numbers, for many decades yet.

From NHK:

Nissan to test self-drive car on public roads

...

Nissan is hoping to put the self-driving car on sale in 2020.

6 years might be a bit optimistic, but "decades" is very likely wrong.

Keyword : "test".

You can focus on "test," and I can focus on "on public roads," and we'll revisit this is 2022 to see whether Nissan's forecast or yours is more accurate.

Edit: It's not just that Nissan is working on this; all the carmakers are working on this, along with several tech companies. That competition will rapidly push progress. The companies have invested in this because they think that the difficulties will be surmounted and any company without a self-driven car will be lost.

 

[dh87]

Alamout[/url]":1r5ie3qx]

You can focus on "test," and I can focus on "on public roads," and we'll revisit this is 2022 to see whether Nissan's forecast or yours is more accurate.

The use of the phrase "significant numbers" in ChrisG's statement is something of an escape hatch, but it's also important--if Nissan (and every other company) release self-driving cars in 2020, they won't be a large fraction of the cars on the road for a long time after that. Cars on the road right now are 11.4 years old, on average. It will take a long time to for autonomous vehicles to cycle into the mix.

My view is that if s-d cars meet 1/3 of realistic expectations of what they can do, everyone will rush to buy one. The replacement cycle won't really play into it, unless you're thinking about complete replacement of the cars.

 

[dh87]

Alamout[/url]":24k2u372]

My view is that if s-d cars meet 1/3 of realistic expectations of what they can do, everyone will rush to buy one.

Most people can't buy a new car regardless of how nice it is. Cars are really expensive! That's why turnover is so slow--because lots of people buy used cars and keep them for as long as possible. It's not like 12-year-old cars are really nice, but they're all over the road anyway.

There are a lot of people who drive older cars because a newer car is just a newer version of the same thing. A new car represents an evolution of the old car. I think that self-driving will be different; it's such a big deal that it will be worth replacing an otherwise serviceable car to get self-driving. It's true that some people can't afford to replace their older cars, and they'll be left out.

The replacement cycle won't really play into it, unless you're thinking about complete replacement of the cars.

What other replacement cycle is there? I'm not talking about new-car-buyers buying new cars every few years. I'm talking about how long it takes for a current-year model to get off the road, and it takes 15-20 years. As cars get even more reliable, the cycle can last even longer.

We were discussing how long it takes for a large number of cars on the road to be self-driving. Unless used cars don't count as cars, that means you have to consider the entire cycle.

The replacement cycle doesn't influence the first adopters. They can just buy a new car if the benefits are greater than the costs. That way, I expect that there will be a "significant number" of self-driviing cars in a hurry.

 

[dh87]

Alamout[/url]":2n8sqkcu]

dh87[/url]":2n8sqkcu]There are a lot of people who drive older cars because a newer car is just a newer version of the same thing. A new car represents an evolution of the old car. I think that self-driving will be different; it's such a big deal that it will be worth replacing an otherwise serviceable car to get self-driving. It's true that some people can't afford to replace their older cars, and they'll be left out.

...

The replacement cycle doesn't influence the first adopters. They can just buy a new car if the benefits are greater than the costs. That way, I expect that there will be a "significant number" of self-driviing cars in a hurry.

Okay. I guess we'll just have to agree to disagree about fairly well-established consumer behaviors.

OK. Just mark this page for study in 2022. I was going to end this with a sarcastic icon of some sort, but then I realized that I've been reading Ars for longer than the 8 years (±) between now and 2022, and you've been posting for 13 years. So, we'll check back.

 

[dh87]

HappyBunny[/url]":2ix2rzzp]

Frennzy[/url]":2ix2rzzp]
The major obstacles posited in this thread are simply those...obstacles...engineering challenges to be overcome. There is nothing impossible about the idea. It won't happen overnight, but few worthwhile things do. You can expect an iterative approach, with consumers and governments lagging behind achievements, but ultimately adopting them.

While there are certainly big engineering challenges, I actually think the political side is the bigger obstacle. Self-driving cars are legal now in a few states, but only with someone still behind the wheel. And who knows what will happen once they start getting in accidents and causing injuries/deaths. They may well be much, much safer than human drivers, but that doesn't make the legal problems go away.

I'm not sure that I see this "legal problem" so clearly. At present, I carry insurance against accidents. If I were involved in an accident, it's very unlikely that I would be charged with a crime as long as I wasn't violating the law. Instead, my insurance company might have to make a payment if an accident were judged my fault. If s-d cars are safer, then my insurance premium is lower, and the insurance company pays in case of accidents. If my insurance company wants to sue the manufacturer because of a defect, they can. However, all the insurance company is really interested in is that the premiums cover the payouts. If an s-d car violates the law, then that's likely the fault of the manufacturer, unless I have specifically instructed the car to do something illegal.

 

[dh87]

Peldor[/url]":1ygqdbza]

If an s-d car violates the law, then that's likely the fault of the manufacturer

Then you do see the problem very clearly. Currently the car manufactures have a very low chance of being found legally at fault in a car crash. It's almost always a driver error. When the car is the driver, now it's a design flaw.

The companies will either carry insurance or self-insure. There seems very little chance that a company or an individual engineer could be found criminally liable. For insurance purposes, the only thing that matters is the statistics. I agree with some earlier posters who think that many individuals will not want to give up driving, ceding decisions to the car, even if it's statistically safer.

 

[dh87]

From my very limited reading, the 2014 Acura RLX seems to be among the most technologically advanced cars. This video (which I think was made by an Acura dealer) seems to show the car executing a lot of self-driving. The RLX is an available car, for $65,000 with all the tech toys. If this video is not exaggerating, it's hard for me to see that fully self-driving cars are that far in the future.

 

[dh87]

The brand new Mercedes S550 has an array of sensors and sufficient computer power to know what to do with the input. The whole thing is described in The New York Times. Unfortunately, the key thing missing from article is an assessment of whether any or all this technology works in real life. The car is reported to have a variety of sensors, IR cameras, and visible-light cameras to enable it to identify objects, for example, to differentiate a pedestrian from a statue. The car gets unhappy if there's no input from the driver, but I really want to know if this stuff works as advertised. Can someone buy one and test it?

 

[dh87]

wco81[/url]":3qa7fech]

...

Driverless cars need robust long-range lidars that can work in any weather conditions and perform flawlessly at highway speeds, said Mario Brumm, co-founder of Ibeo Automotive Systems GmbH, a European lidar maker.

Because people perform flawlessly at highway speeds, regardless of weather conditions. If that's going to be the standard, then all the work on s-d cars is in vain. Especially during the first 10 years of s-d cars, when there's mixed population, there will be accidents, and some of them will be the fault of the s-d car.

 

[dh87]

At least for short stretches (longer if their software allowed), the Mercedes E and the Acura RLX (any maybe more) can drive themselves. I doubt they're using $10,000 worth of sensors to do it, but I can't find what they're doing immediately.

 

[dh87]

Chuckstar[/url]":37oj06xg]

dh87[/url]":37oj06xg]At least for short stretches (longer if their software allowed), the Mercedes E and the Acura RLX (any maybe more) can drive themselves. I doubt they're using $10,000 worth of sensors to do it, but I can't find what they're doing immediately.

Those cars use stereo cameras in the bumper. I believe the problem with relying on stereo cameras, though, is that you really need depth perception out past their effective range. I'm not positive about the Google car, but the earlier car it's derived from (that won the DARPA challenge) had stereo cameras to build a map of the close-in 3D environment, and LIDAR for the farther away environment. I forget exactly where the cut-off was, though.

Mercedes says short-, medium- and long-range radar as well, shown in the Mercedes videos here. I'm not sure that the linked article is the full current system. Acura also uses some type of radar for braking, but the lane-keeping-assist appears to be stereo cameras.

 

[dh87]

Megalodon[/url]":hay57kur]Thanks for the detail on time-of-flight cameras.

Even a few khz seems wildly excessive. This would be sampling stuff every few millimeters even for obstacles at highway speed. Much lower rates, even like tens of hz, seem plausible. At highway speeds, sampling a stationary obstacle at 30 hz would be updates before the distance changes by a meter, and still probably beating attentive human reaction time by an order of magnitude.

I wonder if these calculation still underestimate the problem with interference with sensors that transmit (as opposed to cameras). In a difficult situation, for example a deer in traffic lanes with several cars in both directions, sampling rates might have to increase to match the complexity of the situation. If all cars increase their sampling rates, there could be problems. On the other hand, if there are many s-d cars, they can talk to each other as well.

 

[dh87]

Chuckstar[/url]"6eotw91]
You only need the sampling rates to be so high. We're talking max speeds in the hundreds mph. You still don't move very far between samples, at the kind of sample rates under discussion. Furthermore, you're still dealing with a situation where most of the time the sensor is not broadcasting. Finally, at those sample rates you also have a situation where individual anomalous frames won't matter, because you take another sample so soon after.

Ever been somewhere that a lot of people are taking flash pictures at the same time? How many pictures were ruined by those other flashes? Essentially none, right? Because the odds that someone else's flash goes off exactly when your camera's shutter is open is still tiny. Same thing here.

I tend to think that the rate will be sufficient, but I wonder about one sensor collision leading to more as all sensors hurry to replace the failed reading.

I hope that the camera flash analogy isn't right, because camera shutters are open for considerably longer than flash durations.

 

[dh87]

I don't understand the discussion of a system that doesn't have to recognize this or that kind of hazard. A system that actually drives on the road has to be able to distinguish a cardboard box that the last car ran over from a couple of broken wooden boards with nails in them. Stopping on the highway in the middle of traffic for a cardboard box (or a paper bag or a palm frond) is extremely unsafe.