This Ride Will be Different: Completely Driverless Has Arrived

-edited

Have the car to yourself. Completely driverless commercial cars are now in operation.

Waymo just emailed its customers: "Completely Driverless Cars On the Way"

Here is the email, confirmed by Waymo.

Waymo Is Letting People Choose Rides in Completely Driverless Cars

Car and Driver reports Waymo Is Letting People Choose Rides in Completely Driverless Cars

Waymo uses two different self-driving vehicles in Phoenix: Chrysler Pacifica minivans and Jaguar I-Pace SUVs. Speaking at the Frankfurt auto show last month, Waymo CEO John Krafcik said the company has been testing fully driverless vehicles since 2017, but only on a small scale. "We've begun to responsibly ramp up our driverless offerings to riders in the metro Phoenix area," he said.

When a rider is matched with one of these fully autonomous vehicles without a human behind the wheel in the Waymo app, they will get a special notification that the robot taxi is on the way and will be presented with some information about the technology and what to expect. If a rider isn't comfortable without someone sitting in the driver's seat, they may cancel their "driverless" ride by contacting rider support and will then be placed in a vehicle with a Waymo-trained driver.

Waymo says the driverless rides will be available for users "soon" but would not offer any more details about when that would be. Waymo has also announced it has signed an agreement with Renault and Nissan "to explore mobility services for passengers and deliveries in France and Japan," Krafcik said.

Calendar Check

A quick check of my calendar shows this is 2019, not 2030, not even 2024.

Nonetheless, I fully expect the usual barrage of comments about balloons, theft, dogs, and 78-year-old men on roller skates putting a quick halt to this.

Others will no doubt piss and moan about Phoenix having no snow.

My response is "This is 2019". Things are way ahead of my allegedly super-optimistic schedule.

What About Trucking?

City traffic is logistically much more difficult that point-to-point interstate truck hub driving.

Within two years of Department of Transportation approval, driverless trucking will be common if not the outright majority on interstates.

Within three years of approval, human driven long-haul trucking on interstate highways will about cease.

Expect approval within two years.

In case you missed it, please note my August 18 article UPS Quietly Using Self-Driving Trucks For Months

What About Drones?

Overall Picture

Millions of driving jobs will soon (2-4 years) vanish, mostly interstate trucking. Taxis and other delivery services will follow, perhaps much sooner that I expect.

Mike "Mish" Shedlock

Comments (42)
No. 1-19
Quenda
Quenda

I'm a bit pessimistic on driverless cars. However long haul truck driving makes perfect sense for automation, as do slow frequently stopping vehicles such as garbage collection trucks.

Tony Bennett
Tony Bennett

"Waymo says the driverless rides will be available for users "soon" but would not offer any more details about when that would be."

...

Waymo = Tesla

A lot of bark ... not so much on the bite.

...

"Millions of driving jobs will soon (2-4 years) vanish, "

...

Waayyy optimistic (imo). Not only do I think technology not "there" yet ... but the imminent recession (doozy of) will put the brakes on capex bigly for the companies pushing this.

Sechel
Sechel

I agree this stuff is coming. I see driverless coming to trucking fleets before they come to the consumer market myself. Regulators in cities like New York and Los Angeles will be very slow to permit these vehicles and one mis-hap will cause a huge backlash(ironic when you consider the reckless driving some green cabs and ubers engage in in cities like New York

abend237-04
abend237-04

It will be the insurance companies that take the driver out, and it will happen in long haul trucking first.

njbr
njbr

Hmmm--liability much?

Uber and Lyft have the shield of their wealthy, insured drivers to provide the first line of defense in any lawsuit arising out of accidents.

Waymo, no way....

davidyjack
davidyjack

Heavy job losses extremely likely in the transportation sector between 2023 and 2032. Job losses extremely likely to hit 80 to 95% by 2032. These driverless cars are coming relatively soon. Many workers will have a hard time retraining.

Stuki
Stuki

There already are self driving trucks out there, operating in mines. Has been for quite a while.

And that's what holds the key to interpreting all these announcements: IF you can constrain and control a theater of operation sufficiently tightly, you can make it possible for robots to operate more efficiently than humans there. Exactly how tightly you have to be able to control it, is evolving. As sensors, processing and basic leaning algorithms improve, richer and more complex theaters, which were off limits before, now become possible. And simultaneously, as first hand experience grows, you'll get more confidence. Both wrt putting a bot into a given environment and, and this is important, discerning which environment is good candidates for robot drivers to begin with.

Eventually, someone will attempt extending the operating theater of drivebots, with commercial as opposed to pure pie-in-the-sky research intent: From closed mines to the most predictable corners of the public roads transportation network. More likely that not, to long haul trucking on the most predictable, hence least risky, stretches of freeway. When that happens, especially as soon as more than one group is doing it profitably and hence competition ensues, that will be a big step forward.

Then, the race will be on to, slowly as the risk of real injury is great once your flinging multiton objects around at freeway speeds, open up more stretches of roadway.

Takeaway being, there is a much bigger step from "autonomous vehicles being viable somewhere" to "them being viable anywhere," than it is from where we are today to the former. They're already viable in some specific mining sectors. That hurdle is already met. But that hurdle is also only the tiniest of baby steps of the way to the latter, which after all does include drivebots conducting themselves better than humans as drivers of police cruisers and others engaged in running gunbattles through rush hour Manhattan.

Exactly how quickly the level of complexity increases as you move away from the most predictable of environments and into richer ones, is the central question of concern for those bent on a fully general deployment of drivebots into infrastructures fully shared with humans. Noone has even the remotest idea how to estimate that yet.

But so far, in EVERY other attempt at AI: Early developments has proven easy, promising AND very impressive. Making everyone optimistic. Then, they have ALL, fairly quickly thereafter, hit a complexity wall so steep, almost any progress from the quick, "easy" early stages ends up being completely stymied. Until eventually all those involved get sufficiently frustrated, that they slink back to do something more rewarding with their Nobel laureate sized brains.

So, predicting when autonomous vehicles are viable, doesn't really mean much, unless you specify the environments in which you posit viability. Do you mean in certain, tightly selected mining operations? Do you mean as pure research without even a concern about whether a certain scenario can be self funding? Do you mean commercially viable operation on 10% the total mileage of US freeways? Or do you mean running gunbattles through Manhattan? The difference in years from the first to the last, could be 20 years. Or it could be 20,000 years. Or more. Noone knows, and the history of previous flirtiations with AI booms, certainly don't provide much grounds for optimism.

Herkie
Herkie

Human driven vehicles and computer driven vehicles do not mix. I hate the term driverless by the way, they do have drivers, computers in the cloud and onboard, so we will see how well these things work, and look to the day when humans are not needed for anything at all.

I know that I will never ride in a vehicle (at any speed) that is operated without a human in control, if that means heading off to someplace where this level of technology and infrastructure is not going to be implemented in my remaining years then so be it. I will retain control of my own life and driving no matter where that has to be done.

Arnstein
Arnstein

Wrong, wrong, wrong. Each car will have a "human safety driver behind the wheel". That is a direct quote from https://techcrunch.com/2019/10/09/waymo-to-customers-completely-driverless-waymo-cars-are-on-the-way/ .

The state-of-the-art for autonomous vehicles has not changed in years: research projects involving new technology and one or more humans on-board.

themonosynaptic
themonosynaptic

There is a concept in engineering called "stiction" - it is an extra force that a body needs to get moving - it is always (if my memory serves me - this was a long time ago) greater than the friction an already moving body has to overcome.

Once a force is applied that overcomes the stiction, the same force is an accelerator as it is greater then the consequent friction.

I expect self-driving cars (I agree with an earlier post, "driverless" isn't the best term) to have to overcome "stiction", but then accelerate faster than expected.

Mish has been ahead of the curve on this subject, and I think he is on the optimistic, but also more realistic side of the argument than a lot of the people I see on this board.

Most of my friends (engineers, profs, etc.) are on the Mish side of this.

William Janes
William Janes

As a nation , we should be encouraging this by helping convert our interstates to accommodate this new technology. The increases in safety and productivity will be astounding: less congestions, faster speeds,(speed limits in the 90's) and more productive use of interstates. We can race ahead of the Chinese by not making wasteful and inefficient expenditures on high speed trains that don't take you to where you want to go.

Jojo
Jojo

Tesla’s Autopilot Could Save the Lives of Millions, But It Will Kill Some People First The complicated ethics of Elon Musk’s grand autonomous vehicle experiment. By Zachary Mider October 9, 2019

jivefive99
jivefive99

#1 real job in America is "driver." Millions and millions of unemployed people. Before we start the big driverless celebration, better get A Yang in place as well as his Universal Basic Income ...

Realist
Realist

Driverless vehicles are coming. Like most technological advances, they are inevitable.

Their advantages will far outweigh their disadvantages. Books have already been written explaining hundreds of advantages.

The sooner they are the vast majority of all vehicles, the better. The transition period when human drivers and self driving vehicles share the road will be difficult at first.

The only disagreement I have with Mish is in the time frame. It will be at least two decades (though probably closer to 3 decades) before the vast majority of all vehicles are driverless and the accompanying benefits are substantial.

Part of the reason for the 20-30 year estimate is production capability.

There are currently over one billion vehicles being driven by humans in the world today. Annual production of vehicles is 70 million. So at a minimum, it will take 14 years to replace the present vehicles on the road.

I believe that the majority of self driving vehicles will be fully electric (though they do not have to be). There are currently 5 million fully electric vehicles on the road, and production is currently 2 million per year. Though this will rise quickly, it still has a long way to go to reach 70 million.

At present, there is not enough materials (such as lithium) in the world to produce the number of batteries needed to manufacture 1 billion vehicles.

The nations that can dominate development, production, and implementation of these vehicles will be the biggest winners.

JMOD46
JMOD46

Mish apparently thinks driverless cars are a technological problem to be solved in order for those vehicles to be ubiquitous. If he thinks that, he is wrong. The State (and the insurance industry) will determine whether these vehicles displace conventional transportation with cultural shifts a distant second.

If the free market determines the outcome, my money is on the status quo as far as personal transportation is concerned.

JG1170
JG1170

A computer can be programmed to drive off of a cliff, and it will execute the command perfectly. My Uber driver, conversely, cannot be convinced to drive off a cliff, no matter how large the tip. So I'll stick with option #2 and just pray I never run into a suicidal driver.

Realist
Realist

Statistics say 94% of auto accidents are caused by human error. If you take the human out of the drivers seat, accident rates will drop dramatically, as will auto insurance rates, health care costs, road rage, etc

Sologretto
Sologretto

Just did a bit of research. It looks like there are only about 9 million driving positions that would be threatened by self-driving vehicles. That's less than 10% of workforce.

How might things change though?

How many truckers end up becoming local last-mile teamster/delivery persons?

How many truck-stops go out of business and/or transition from human service focus to automated vehicle refill/cleaning service focus?

How many bus routes expand and speed up operations by utilizing multi-hop implementations with a mix of buses making less stops and vans/cars providing the regional transportation to bus stops? and how much would this reduce the need to own a vehicle in a much wider range of consumers?

How many service stations go out of business as less people go to convenience stores as part of gas fill-ups and simply send their vehicle to get gas by itself?

How many delivery drivers lose positions to new curbside delivery options where customers walk to the vehicle to get their delivery?

How many less people choose to own cars and instead join a time/share vehicle service or use one shot fleets like waymo/lyft (uber's doomed due to cash burn).

Alternately how much MORE transportation would happen due to conveniences?

If I could work, play and sleep during a 3-6 hour commute would I be willing to commute those distances if it were affordable? What about road trips? If my vehicle could drive to my next destination at night while I slept how much more travel would I do? (hint... lots)

Housing... cost of housing may come down if people are willing to commute further and this makes extreme housing prices in urban areas come down, however what about singles and childless couples who alternately choose to invest in a self-driving RV/Van? This would allow a very efficient nomadic lifestyle where access to shopping, camping, parking and conveniences could be easily accessed. Go to sleep after full day working and playing in city. Vehicle drives to national park for free camping for a couple days before automatically night driving back to city for next day(s) of work?

I sell product on the road. A self driving vehicle would be a Godsend. Work on computer between client visits as vehicle drives. Have the vehicle drop me at appointment (no parking concerns), call it back after appointment finishes... wash, rinse, repeat. Sleep in vehicle and have it deliver me to my nearest gym for exercise/morning-ablutions/showering and so on the next morning.

These trips have an average of 6 hours a day of driving focus for me. Getting that time back would be valuable to me... but unfortunately paying a driver and their per-diem would eat a large fraction of my personal earnings or else I'd just do that..