Self-Driving Electric Cars Will Dominate Roads by 2030
spectrum.ieee.org
spectrum.ieee.org
That's localized death on the order of thousands at most, not global death of millions like the parent post was suggesting. Still obviously horrifying, but doesn't really compare in terms of the threat.
A sufficiently hypothetically scary driverless car attack might manage to do more deaths (though it'd be really hard, even for a state actor). But I think it wouldn't really get to the level of "more capable of causing terror." If you could cause let's say 15,000 deaths by a well-coordinated one-day bombing attack, and 80,000 deaths by a well-coordinated one-day driverless car attack, do those two things actually differentiate themselves in terms of your nation's agenda? I'd think they'd cause more-or-less identical amounts of terror, and the only purpose of such attacks are terror.
Ultimately, I don't think it's important. Nobody who's resourced well enough to pull this off is interested in doing it.
It is extremely clear to me that today both desktop computers and IoT devices do not require a well-resourced attacker to compromise (can we agree on this?).
The point where we disagree is that I see little evidence that driverless car companies are investing in defense in depth that is at a level significantly higher than either IoT devices or desktop operating systems. Indeed, there is such a race to be first-to-market that the companies are incentivized to not spend resources on it.
People who write these reports seem to be ignorant of the diversity of the automobile owning public. Does no one live in suburbia or have kids? The idea of sharing a car just doesn't make sense. it's part home / workshop / storage area. Maybe innovation solves this. Your rental minivan will dump its pod of crap at your house when it drops you off and picks it up when you call it.
Yes, if you live in a dense area this stuff makes great sense. But how do you deal with these use cases:
1) Follow behind your kid on halloween 2) Casually drive anywhere and pull over to watch the view. "Alexa - pull over at that little turn off." maybe?
Ok, so this is in no way a criticism of you or your kids, but having kids doesn't have to mean that (e.g., would you let your kid leave her crayons on the subway? in the public library?). Sure, kids anywhere could occasionally mean kid vomit, but that's what damages fees are for, right?
Lots of people, my sister included, have kids and don't even own cars, today, right now.
My car has in it:
1. My daughter's carseat.
2. About three kleenex-box-sized volume worth of her books and toys.
3. Wet wipes and diapers and changing equipment.
4. A booster seat.
5. A box of food and eating stuff (bibs, etc.)
Those all live in my car permanently. I don't want to take them all out and put them all back every time I go to the store.
Maybe the design of cars radically changes. Maybe the design of spaces in which we live changes to accommodate a more car free/car subscription but not ownership lifestyle. Parents are a large market and people will make money by solving these problems, but the first step before that is autonomy.
1. Pay a bit more for a car you own (I mean, this isn't THAT expensive. A good used car costs maybe $8,000 and can last for five years).
or
2. Fundamentally restructuring your life to a pre-modern approach of childcare
Which one would you do?
From an individual perspective though, few people will need to make a big tradeoff decision all at once. It'll start by just trying out the service, using it when convenient. A few years later, when the question of fixing or replacing a car comes up, then people will ask themselves whether they really need a car.
A few things that I think you are assuming that will not change are:
1) High availability of free/cheap parking at all of your destinations including your home.
2) high availability of autonomous cars actually for sale to end users. Due to the odd liability of an autonomous car (is the owner responsible for a crash? how do you insure against that) I anticipate most/all L5 autonomy to be subscription based, not available for purchase.
3) Prohibitively expensive congestion/registration/etc charges for owner occupied cars in order to reduce the number of cars on the road
4) car ownership remaining legal (unlikely to change, I admit)
There is a lot that can happen. You will probably own a car for the rest of your life, your children probably will never own a car.
My family have two cars. We have two kids. Somehow we manage to do all stuff you are moaning about using a bag. I think your argument boils down to how much effort are you willing to do to save some cash. For you storing a few books in the car is worth $500 a year to you. For most of us we'd take that $500 and buy a bag.
I'm sure if we're all sharing cars you'll be able to order one with a child seat. And why wouldn't they have complimentary wet wipes and all the other stuff? Books, iPad, whatever.
Yeah good thing we pushed on with the technology and ignored them.
If the software is tested then what? We trust the people who test it. Send the self driving car through a driving test.
Sure level 5 is hard... but most people don't appreciate that. They assume it's tested and it works most of the time. Most people will jump in the self driving car and whip out a phone and tweet and facebook and enjoy the free time.
> compared to the internal-combustion standard, self-driving EVs will represent, “a 90% decrease in finance costs, an 80% decrease in maintenance costs, a 90% decrease in insurance costs and a 70% decrease in fuel costs.”
> Car subscription services will increase utilization of a vehicle from 16,000-32,000 kilometers per year to 160,000 kilometers or more. And that reduces the number of years the car is in service (and thus interest paid on any financing plan) as well as increases the amount of use wrung out of it per year.
These seem like wildly optimistic projections based on an assumption that a single self-driving car can take the place of 5-10 conventional vehicles. This could possibly work if commuters don't all need to use those cars at the same time (maybe a wider prevalence of non 9-5 work schedules) or that a substantial portion of autonomous vehicles will be high-occupancy minivans and the vast majority of commuters are willing to carpool with random strangers. I don't expect either of those things to happen.
I do think self-driving cars will be successful and widely used, but it's hard to imagine them having a 95% share of passenger miles when there's plenty of manually-driven cars still on the road.
> The savings represent some $5600 per consumer per year or $1 trillion in additional disposable income in the U.S. alone by 2030.
According to the first thing I found on Google[1], average consumer spending on transportation was about $9503 per year in 2015 (which is a lot higher than I would have guessed. Apparently I'm an outlier). I think it's unrealistic to assume that lower vehicle maintenance and insurance costs are going to be passed on directly to consumers rather than shareholders. I know that's how efficient markets are supposed to work in general when you have multiple competitors, but I'll believe it when I see it.
Level 5 autonomous driving means driving better than most humans so it automatically means safer.The problem is that they basically have to perform better than humans in all situations. As soon as someone dies from a crash that a human could have prevented -- like the Tesla that plowed into a semi because it was white and it couldn't distinguish it from the sky -- it's basically game over for consumer trust.
If all we care about is safety then total automation is going to be fighting a losing battle against a machine assisted human drivers for a long time.
Shouldn't the company who develop the Self-Driving tech be liable for any accidents/damages its tech might cause?
If so, great deal for consumers, but I doubt the developers of the Self-Driving Tech would agree pay for the insurance for everyone
If the software developer is liable, then the software developers will just include that extra cost upfront when you purchase the self driving software/vehicle, or more likely as a monthly subscription to continue using their software, just like insurance!
If the customer is liable, then customers will just purchase separate insurance like we already do.
At the end of the day it's pretty much the exact same system. My 2 cents is we should just keep our current system and let people buy insurance from a third party instead of having to go through the software developer.
If your car is 'at fault' in an accident and you weren't driving do your rates go up? What does that even mean?
It's a really, really uniform pool, especially if you also have geographic risk information.
In fact, your rate could go up because other people with your software are getting into accidents? Sounds crazy right? But it makes complete sense. Eventually it'd encourage you to switch to a safer self driving car system, which is good.
It's possible that a car-share would set up a system where you can swap out the battery of the car at a stop and continue driving; similar to stopping at a gas station.
But long distance travel is getting better all the time in EVs, by 2030, it may not be a big deal to find a charger and get a quick charge every so often.
Why not have a car drive to pick you up for your commute maybe with someone else heading your way - then it goes about and picks up people who are heading to the shops, have doctor's appointments, are heading to the airport.
You might even part-own that car - you get priority on your travelling hours - and the lease company does a daily valet before the car comes to pick you up.
These are just two examples from my own life but I am sure there are other reasons to own a private car.
There is some kind of happy-ish medium there, but it's another chip away at the claims of the omnipresence of rides-for-hire.
But owning your own car means:
1. You can leave your stuff in it and only grab your stuff when it's convenient.
2. You always have access to your car and are never stuck during high demand times.
3. You can go to places that aren't in dense population areas without either paying a premium or waiting a long time for a ride to come pick you up.
4. Potentially prestige or personal style things.
It's a luxury (but so is owning a car today). If the price of the luxury is low enough, lots of people will pay it.
How many times am I stuck waiting for a car to arrive, and how many times do I need to make an unnecessary trip back home to unload my gear before getting back in yet another car?
Certainly there are times when I get into the car for one thing. But you know what? I walk outside and I start the car and drive off. I don't have to schedule anything or wait for anything.
More specifically, the pick up/dropoff points are always just slightly out of the way, I'd have no problem with car sharing if improved locality was a thing. Which would be easily solvable with self-driving cars, coming right up to my curb.
Seriously? Fucking walk.
Personal: My mom has CRPS and walking around causes a great increase in pain for her(painkillers, even morphine level have almost 0 effect), even being in a car can be very difficult for her, but a vehicle she is able to better position herself than in a wheelchair. She can't drive because of it, but an autonomous car would make it so she could potentially go places (or follow the grand children around at Halloween). Both my parents are in their 60s and 70s and are super excited about level 5 self-driving cars.
Random other example: Parent is on crutches from injury, insurance didn't cover wheelchair, or perhaps live in area with poor wheelchair accessibility(bad/no sidewalks). But crutches are too slow to keep up with the kids, they could easily just sit in a self driving car and keep up. But not having to focus on driving and hitting other kids night, can focus on watching their kids and hanging out the window to interact with them.
Sure these aren't necessarily applicable to ALL people, but you asked "such as?" to "Um, there are reasons other than laziness."
I gave you at least 2 specifics such as. There are plenty of laziness reasons, but doesn't mean there aren't potential valid reasons besides laziness.
I also generalized the "Various forms of disabilities, injuries, other conditions that walking isn't great for." in the beginning. And there are literally tens of millions of Americans that have at least 1 disability (approximately 20 percent of the US population). And lots of other countries like Japan has an aging population which are likely to have various mobility issues. Not including injuries, or potentially other issues. So I would say that is fairly generalizable.
* it's cold and kids need a place to warm up
* mixed-ability group of kids want to go together,
tired kids need a place to take breaks but keep up with the groupDriving is a magical experience. We don't think about it at all, we just do it. The issue is that a level 5 car will still need traditional input devices for things that are harder to describe with voice than just, you know, doing it. And because you still need traditional input controls you probably will still need insurance.
I think there should be a law that these cars are no bigger than a Fiat 500 (a little bigger than a bumper car).
Tesla market cap now outsizes Fiat. Tesla should buy Fiat immediately.