I’ll define “common” as “at least 25% of cars on the road have the capability.” I’m not including lane keeping or adaptive cruise control functions as self driving. You have to be able to tell the car where to go and it takes you. If the car needs assistance because of unforeseen conditions, that counts as self driving.
Poll coming.
I’m going 10-20 based on how long it took things like “cell phones” to become common.
I’ll say at least 20 years, and even then only in major cities and along protected routes.
People will resist letting something else control their traveling. It’s not in our nature to be satisfied with how someone (or something) else drives. If it were, there’d be a lot more commuting by bus than there is. Ask yourself how often you get impatient with what other drivers are doing when you’re out driving. Driverless cars are the same thing, except that your perceived awful driver is the one who’s in control of your vehicle.
First the technology needs to be invented right now level 4 cars are 3–5 years out. I’d guess we’re a decade out from a major production run of a level 5 car that is affordable by 25% of the population. From there we need to wait for that top 25% to replace their car which is at least 10 years. I think 20 years is the most optimistic time line so I went with more than 20 years.
The best selling car in the US in 2017 was the F150 and they sold 2.5 million. Out of 276 million cars on the road that is 0.08%. To get to 25% it would take 30 years assuming they were all still on the road. I don’t see how self driving cars are going to become the number one selling car their first year in release and realisticlly there would probably need to be at least three brands at the top of the best selling list to get to 25% of the cars in a decade so even if I’m wrong about a decade of development to get three to be developed and top the sales charts will certainly be a decade.
The convergence of urbanisation, electric cars, self-driving cars and cars-as-a-service should make it common in cities within 10-20 years. They don’t need to solve all problems everywhere, just what’s required to change transportation in cities would be a great boon.
I enjoy other people driving so I don’t have to think about that stuff. If other threads are any indication, I won’t be the only one saying that here either. Sure, there are people with control issues and they’ll do their best impression of grandpa Simpson but they’ll go the same way as manual transmission.
I’m rooting for cancer!
I could be the exception:
a) Everything I own is manual transmission because I like the control it gives me.
b) Market an affordable self-driving car and I’ll be first in line.
The ability to nap or multi-task on longer trips or to be able to basically “co-pilot” on regular drives (such as to work) has such a high appeal that I can’t even begin to quantify it. So much so that I am willing to dump all the driving habits I’ve built over my lifetime. That and you are looking at stretching my ability to “drive” way past the age and health where I could safely drive and yeah -------- I want one.
And a flying self-driving car would be even better! 
I really don’t want self driving cars on surface streets where there’s random ass bikes and pedestrians and animals darting across the road but dayum, as a fan of the road trip I’d be perfectly okay with a car I could allow to drive the looooooooooong boring stretches of interstate freeway in between interesting destinations. I’m almost always the driver and it would be nice to be able to read and look at scenery instead of concentrating on 800 miles of multi-lane boredom.
I said 5-10 year, but on reflection I think it’s more likely to be 10-20.
I think there will be a big market for them, which may accelerate (pun intended) the time needed for development. There’s a lot of Baby Boomers; as one of that group I can say that most of us like our independence, but worry about our ability to be able to drive in a few years.
Manual transmission is 2% of all vehicles sold in the US in 2018*.
That’s down to electronics and software getting better and more readily available which we can expect.
Flying cars exist; They’re called “helicopters”. Use taking off/flying/landing autopilot on a helidrone and you have a self-driving flying car. I don’t see that becoming widespread but it is possible. Right now, you could strap yourself on a scaled-up Amazon drone and have just what you ask for.
Its better worldwide which is why some of us (in the US) are looking into gray-market importations.
And I am thinking of an actual car that can fly or drive without quite the issues we found with the Pinto-planes. But I appreciate the good thoughts!
I voted 5-10 years, but it will depend on state governments. You can count on my home state of Indiana and some others to lag behind by at least 10 years. Right now, Indiana is the only state where you can’t buy cold beer to-go in a grocery store or on Sunday.
I think 10 to 20 years for them to be common on the roads. I expect to see the change starting before that, obviously…I’m guessing within 5 years it will be a completely viable option for a self driving vehicle, but from then it will take time to build up to being common. I might live to see the day when the vast majority of cars on the road are self-driving. I expect the carnage on our roads to drop radically during that period, from 10’s of thousands (nearly 50k deaths and 100’s of thousands of injuries) deaths a year to a fraction of that.
I expect to see self-driving cars in the next 10 years.
Unless they have high accident rates and that could set the technology back 25 years
The average car lasts something like ~20 years now, so unless they invent some kind of cheap and safe aftermarket attachment for self driving, these cars have to slowly enter the marketplace.
We’ve had hybrid cars for over a decade and they’re still a small fraction of cars. And a lot of people wouldn’t feel safe with a self driving car at first.
I’d wager it’ll be 2040 before 25% of cars on the road are self driving. But the first self driving cars that are available to consumers will hit in the 2020s. We already have self driving cars, but I don’t think consumers can buy them yet.
However by the 2060s or so, I think human drivers will be prohibited just because by then self driving cars will be thousands of times safer than human driven cars (self driving cars have gone from barely functional 15 years ago, to safer than human drivers now). I’m sure by the 2050s or so, when human drivers only make up a tiny fraction of drivers they will make up virtually all the accidents. So at that point legislation to protect other drivers will start prohibiting human drivers.
I said “10 to 20 years,” based on the OP’s definition of “common,” and it will probably be the higher end of that range.
Although the average age of vehicles on the road today indicate a slow replacement regimen, in many cases the availability of self-driving cars will accelerate that. After all, those that can afford them will probably be early adopters. I can see city centers restricting manual cars, thus incentivizing self-driving, much like congestion charges that we’re starting to see. And like HOV lanes, dedicated self-driving lanes (or entire roads) will further incentivize the turnover.
On the other hand, we don’t have the political will to increase our state fuel taxes to fix America’s most embarrassing roads, so it’s easily foreseeable that we’ll not do anything of the sort because it would be labelled as regressive. I suppose a new “cash for clunkers” might help.
I voted over 20 because of the amount of time it takes to turn over the whole automotive fleet.
I’m guessing we’re about 10 years out from a true self driving car being commercially available. Within 5 years after that, I’d expect self drive to be as common as anti-lock brakes, on new cars only of course. Then it’s a matter of displacing the fleet which will take at least a decade longer.