How does an autonomous car decide which parking space I want?

You are suggesting that gravity can help get you where you are going but the slippery weather is otherwise not a problem for you. There are no uphill segments between your house and your destination that might cause a problem? Networked self-driving cars will know more about the travel conditions over the whole route than you are likely to know when you leave your house. You are betting that you can make it down that two mile stretch and then betting that the roads after that are good enough for you to traverse.

Essentially every person in the article I linked to was in the same position you were. They thought they would be safe on their route, or at least they were willing to take the risk. Instead, 67 of them were in a chain reaction pileup, dozens more were in another pileup, and hundreds more were in other collisions. Three are now dead.

Did you know half of people killed in car collisions in the U.S. die in single-car crashes? Some, I’m sure, are suicides. Some were passengers. Many are just people who overestimated their abilities. People are terrible judges of their own limitations. Please stay safe while you operate at the limits of your abilities.

I drive okay in adverse conditions too. I made it safely to my destination in that same storm Saturday that I linked to with no slipping or close calls. On Sunday, I was rear-ended by an inattentive driver while stopped on a mostly clear road heading into a mall. The biggest road safety problem is drivers. I wasn’t injured and my car wasn’t meaningfully damaged but I’ll bet I wouldn’t have had even that inconvenience if the car behind me had been self driven. If I had been a pedestrian, i don’t know how badly I would’ve been injured.

Finally, if ascending the unplowed road is a problem for you, what are you doing for the return trip in your car? Are you gambling that the plows will make it up that two miles before you have to? If so, you are still travelling at the mercy and limits of other vehicles (i.e., snow plows) whether you like it or not. In my opinion, that’s not really very different than relying on autonomous cars who might sometimes tell you that you can’t make it to where you want to go when you want to go.

You are certainly right about this. Self-driving cars can probably use the roads more efficiently. They can drive closer together and coordinate their movements. They won’t cause as many traffic-clogging crashes. Unfortunately, all this benefit will be more than offset by the fact that if people’s commutes are 20% faster, they may move 25% farther away and keep their commute time the same. If they can work from their car, maybe they will move an hour further away from work and just catch up on emails during their commutes. People might switch from public transit to cheap self-driving taxis. Add to that all the empty cars carrying packages, cars driving to distant, cheaper parking, and cars carrying non-drivers (children, the blind, the elderly). More traffic is a distinct possibility. I certainly don’t see it getting better.

Correct, because once I don’t need gravity, I’m to a plowed highway. Sure it may be snow packed and icy. It is six months out of the year. That’s what it’s like when you live at 11,200 feet 1/4 mile from the continental divide. :shrug:

No because that part (the continental divide) gets plowed

Not worried about the rest of the route, and there is only one route to work or town.

And for 24 years traveling every day I have won that bet. So has my Wife.

I guarantee they are not in the same position I am.

Once in 24 years on the return trip the last part of my road was not plowed and impassable. I simply parked my SUV, walked the last bit, got my plow truck and took care of it myself. Often though, I have to park on the ‘road’ at the base of my drive way and plow out before I can actually get up my drive to my house. Getting out from my house in the morning has only been a problem 3 times in 24 years. Stayed home from work to dig out.

Will the autonomous car balk at a foot of snow that throws an occasional blizzard on your windshield? Then it’s no good for me in my situation. Will it refuse to smash through a berm thrown up by a plow or snow drift because it looks like a wall? Then it’s not going to work for me. Yes I’ve been stuck in snow. I’ve a winch for that. Still made it where I was going. Sort of a pain, yes, but not really a big deal.

As it is both my wife and I have to turn VDC off to get up our driveway if there is more than about 4 inches of snow. Yep, people can be smarter than the car. We get 30 feet of snow a year.

I think it highly unlikely that an autonomous car will be developed to handle the things that I face. It may be able to be done, but conditions such as I face are rare. Wouldn’t be economically feasible.

I don’t understand why people always make a connection between self-driving cars and car sharing. Yes, there are advantages to car-sharing, but all of those advantages already exist now, with human-driven cars, and to nearly the same degree they would exist with self-driving sharecars. People now have, for the most part, decided that the advantages of sharecars are not worth the drawbacks, and I can’t see why that would change in the era of self-driving cars.

Agreed. But somehow somebody always conflates the two things.

Emphasis mine.

Chronos. As LSL said, self driving and ride sharing gets mixed up. I do it, I know. I do take issue with what you said in that earlier post. I hate absolute statements like that.

There are plenty of people, like myself, wife and friends that can face conditions that I very much doubt a self driving car will ever be able to handle. The best it will do is say no.

I’m a GIS programmer myself (I deal with spatial stuff all the time). I understand a bit about the different algorithms that go into systems.

Two weekend ago I was asked by a new neighbor for pointers after I plowed him out. He was having problems plowing (he has a CJ chained up on all four wheels). I gave him some tips, but remember my words to him after plowing his drive. “It’s different every time”.

Sharing just seems to be a natural consequence of fully autonomous cars.

I have a car that can drive itself. With the right software, that car can make money for me by being an Uber “driver”, during times when I don’t need it.

Many people would make a similar choice, which would make Uber like vehicles highly available. Available enough that maybe I don’t even need my own car, I can pay a fraction of the cost of a car to get enough transportation for my needs.

The difference between self-driving car sharing and car sharing is that I don’t have to get from my house to the shared car, I don’t have to drop off the car somewhere that’s not my house, I don’t have to keep the car for the whole day. It’s being able to hail a cab any time you want from anywhere you want, to go anywhere you want for a much lower cost than owning your own car.

As long as you don’t want to go someplace at rush hour when they’re all already in use.

I assume the law of supply and demand will take care of that.

“If I can make that much money lending out my car at rush hour, I’ll travel earlier or later.”

“If it costs me that much to rent a shared car at rush hour, I’ll travel earlier or later.”

Right. And that’s cool. You want to share your autonomous car, that’s fine.

As long as people don’t say that an autonomous car, one you own, and don’t share will always drive better and make better choices in any conditions.

I have the option of saying “I have about a 75% chance of getting up my driveway”. I have to consider how important it is to get up my drive. The ONLY one involved here is ME. Do I want to plow? Or would I rather not, and risk getting stuck and have to pull the car out? The autonomous car will no doubt say no. VDC already does.

That’s my option, and is my objection to people saying that an autonomous car will always drive better. It may not be a good idea for a person that has little experience driving on snow on ice to continue. I’ve been doing it for 24 years. I’ve been driving for 41 with never any accident of any type. I would not be very happy if a car told me that it would not continue when it’s a simple, daily experience for me.

But the high volume self-driving rush hour traffic will move much more smoothly, efficiently and quickly and avoid the accident tie-ups.

I suspect it’s more likely to be insurance costs that is the driving factor (pun intended). Once the actuaries do the math and work out the actual higher costs for human driven cars that will be passed onto the consumers in state mandated insurance. Meanwhile insurance on autonomous vehicles will be much lower. But then I suspect they’ll never stop making human driven cars in the US, they’ll just be limited to performance and luxury models aimed at people that can afford the massively increased insurance premium for choosing to drive yourself. (And of course they’ll also have autonomous modes, for usage optionally as needed).

Meanwhile, outside the western world, a huge percentage of the world relies on cheap simple 100cc or less scooters as their main transport. I really can’t see those getting replaced with autonomous vehicles of any kind in the foreseeable future, not without massive government interventions. They’re already using road space and fuel much more efficiently than cars, so some of the key arguments for autonomous vehicles don’t really apply.

I could see it happening, if the cost of hailing an automatic cab decreases below the per-ride cost of that cheap scooter.

Revolutionary changes are rarely ideal for everyone, though. The fact that your current lifestyle won’t be viable if/when that happens might or might not be of concern to the 99.9% of people who live in less challenging environments.

All the discussion about snowy drives and highway lanes is pretty irrelevant at present. Autonomous cars are really being designed for city centres where they currently offer the most benefit. It is reasonable to assume that they will be fully electric as well so as to reduce pollution.

London is already considering banning or pricing diesels out of the City Centre - no doubt others will follow and petrol will not be far behind. Self drive is a no brainer when the average speed is less than 9 mph and parking can cost £250 a month.

After reading the comments about what people will do, left unsupervised in a public SDC, I have no doubt that CCTV will be standard as it already is in London Cabs. Surfaces will all be wipe-clean too.

We already have roads where bicycles, modpeds, and other slow-moving vehicles are not allowed. Motorcycles would fall in the same category as human-piloted automobiles; where they are allowed, they are allowed and SDC’s will be alert to handle them. I imagine a day where SDC’s communicate with each other, someone violating “SDC only” rules will be noted and reported to 911 by the autopilots (complete with pictures) almost immediately. Autopilot already avoids known obstacles. Wandering pedestrians and erratic vehicles won’t be a new phenomenon. Getting a ticket for being one might be.

I expect expressways will have “Autopilot only” lanes; what’s the advantage? SDC’s can travel faster and denser, allow merge and unmerge faster, faster reaction times to incidents, advance warning from other cars about obstacles, etc. (and don’t slow to gawk at accidents in the other lane) Then, whole expressways will become SDC-only. Then expect the same creep on main streets…

Coremelt has a good point - at a certain time, insurance will probably determine whether human control will be an option. OTOH, insurance is priced according to risk and expected costs. With fewer stupid drivers around you, the risk of being hit by an uninured stupid driver goes down (and with faster reaction times, perhaps damage will be less). So unless health costs become astronomical (what are the odds of that?) or liability awards (driving your own car instead of letting the autopilot do it is negligence?) someone will sell reasonable insurance to cover the chance you damage your vehicle or hit someone.

(Watch for the first case around 2030 where the person who was injured says “I assumed I could do that, because I expected the car to be smart enough to automatically avoid me.”)

I was discussing this with my hubby, and he pointed out that store Web sites never remember my favorite store in the chain. If I go to Sears.ca or CanadianTire.ca or BestBuy.ca, you can be sure that it will give me delivery and stock details about a store near Toronto, and I will have to tell it where I live again so that it will know which store I’m interested in, again.

So it’s 2025, I tell my autonomous car that I want to go to Best Buy, I take a nap, and I wake up in a Toronto suburb! :slight_smile:

It’s important in your self-driving car not to turn off cookies.

It’s hard to see that there would be enough profit motive to bother, but maybe. These things can go 20 km on one litre of petrol, are mechanically simple and parts and labour are cheap in the countries I’m talking about. They cost almost nothing to run.

I can predict a future where “self driving tourism” becomes a thing, where people go to countries where self driving is still allowed in order to rent a car or motorbike and enjoy driving it themselves.

That’s a cute, witty comment. :slight_smile:
But it actually raises a very important issue: how robust will the software be, and who will set the standards.
Because if, say, one brand of car allows cookies, and another company does not—there are going to be big big,problems.
Will all cars recognize each other all the time? Will Apple cars know how to deal with Windows cars?
Will individual drivers set their own cookies and preferences?

Here’s an example, :
There was a funny-but-not funny story on Fark .com about an ambulance which failed to save a patient because it didn’t arrive on time. The reason: the ambulance driver went to the location by strictly following the route displayed on GPS(as required by the regulations of his job.) But somebody on a previous shift had reset the GPS program, and accidently changed the mode–to “scenic route”, instead of “fastest route”.

nm