Tag Archives: autonomous cars

A Happy Thought

Posted on by

There may actually be some positives to the impending arrival of self-driving cars.

Yeah, I know, that sounds odd coming from me, doesn’t it? But it’s true.

Consider the case for predictability. Many of the concerns I see about self-driving cars boil down to “How do I know what it’s going to do in Situation X?”

How do you know what any driver is going to do in Situation X? In truth, you mostly don’t. You guess, based on your experience, your familiarity with the law, and what you’ve seen of the driver’s behavior. Usually you haven’t seen much of the last. If they’re doing something blatant–speeding, weaving back and forth across the lanes, going the wrong way on a one-way street–you pay special attention to them. But for the majority of drivers, you’re guessing.

How many times have you seen someone sit at a green light because traffic is backed up from the next light, so there’s no place to go? Not very often, at least around here. The default assumption is that two or three cars will move into the intersection and still be sitting there when the light changes. Now the cars on the cross street are blocked. Presto! Instant traffic jam.

And yet, this morning I saw three different cars waiting at green lights for the traffic ahead of them to move. It was startling enough that I took special note.

My point is that over time, we’ll build up a category of experience specific to self-driving cars. We’ll assume they’ll wait at green lights instead of blocking intersections*. And we’ll make that assumption because we’ll see them do it every time. Not just on days when they’re not late for work, didn’t have a fight with a loved one, or feel like being passive-aggressive.

* Assuming they weren’t programmed by Bay Area drivers.

We’ll be able to make better predictions about what they’re going to do than we will about all those human drivers on the road.

That’s a good thing, but here’s an even better one.

Even a tiny number of self-driving cars on the road have the potential to break up traffic jams before they start.

Seriously.

Okay, I’ll admit I’m extrapolating wildly from a study I saw back in the days before journal papers went online. But I’m a science fiction writer; wild extrapolation is part of my job description.

The gist of the study was that one of the common reasons traffic jams develop is that a few drivers slow down. The drivers behind them overreact and slow further, then speed up to close the gap. Errors accumulate, and again, Presto! stop-and-go traffic.

I see this every day. There’s a curve on the freeway where most drivers slow down from 65 (or whatever faster-than-the-limit speed they were going) to 60. Outside of commute hours, it doesn’t matter. Everyone slows a bit, then resumes speed on the next straight patch. But at rush hour, that curve always turns into a parking lot.

But the study went further. The investigators found that if a small percentage* of the drivers maintain a constant speed–even if that speed is well below the limit–instead of braking and accelerating, the jam never develops.

* I want to say five percent, but I’m working with twenty-year-old memories, so that may be incorrect. I am sure it was a single digit number.

Self-driving cars, if properly programmed, aren’t going to slow down for a curve they can safely negotiate at the speed limit. More to the point, if they get proper information about traffic conditions ahead of them, they won’t get into the slower/faster/slower/faster cycle that causes jams. They’ll just slow to the maximum speed that won’t result in a collision.

Maybe that doesn’t sound like such a big deal to those of you outside the Bay Area and other commute-infested regions. But not sitting in stationary traffic on that one single stretch of freeway would trim my morning commute by ten minutes. And there are two other spots on my normal route where traffic behaves the same way.

Saving half an hour a day and however much gas the car burns idling in traffic sounds like a very good deal to me.

 

Follow the Leader

Posted on by

Can we talk about self-driving cars again? Oh, good. Thanks.

It occurred to me the other day that the public press (as opposed to the technical press) isn’t paying much attention to one particular aspect of autonomous vehicles: interoperation.

Every article I’ve seen that touches on the subject makes mention of emerging standards and the need for inter-vehicle communication, but they all seem to assume that having standards is the solution to all the potential problems.

Believe me, it ain’t. For one thing, there’s the ever-popular catchphrase “the great thing about standards is that there are so many of them”. Just because a car implements a particular standard, that doesn’t mean it implements every standard. And which version of the standard? They do evolve as the technology changes. Will a car that’s compliant with version 1.2 of the car standard for signaling a left turn recognize the intention of the oncoming truck that’s still using version 1.1 of the truck standard?

Lest you think I’m exaggerating the problem, consider the rules (not a standard, but similar in intent and function) for the noise-making apparatus in electric vehicles. (I talked about it several years ago.) That one document runs to 370 pages. Do you really think there are no errors that will require updates? Or a significant amendment to cover cars made in other countries? Or a missing subsection for retrofitting the technology to older electric cars released before the rules were finalized?

And, speaking of those 370 pages, that brings us to the second problem. Even assuming the best will in the world, no spec is ever totally unambiguous. Consider web browsers. Remember back around the turn of the century, when we had Internet Explorer, Netscape, and AOL’s customized versions of IE? All theoretically compliant with web standards, all delivering different user experiences–rendering pages slightly–or extremely–differently.

Nor do they do anything to prevent developers from introducing non-standard extensions. Do we really want some latter-day Netscape-wannabe from coming up with an automotive blink tag while their competitors over at Microsoft-like Motors are pushing their equivalent of the scrolling marquee tag?

But I digress slightly.

What started this train of thought was wondering how autonomous vehicle developers are going to handle weird, one-off situations. We know some of them are working up plans for turning control over to remote drivers (like OnStar on steroids). But how well is that going to work at 60 MPH?

Case in point: The Richmond-San Rafael has a part-time lane. For most of the day, it’s actually the shoulder on the eastbound part of the bridge. But during the afternoon rush hour, it becomes a traffic lane. There are lights to signal when it’s open to traffic–and the open hours are scheduled–but it can be taken out of service when necessary. That means developers can’t count on programming open times. Cars may or may not be able to read the signal lights. Maybe there’s a standard compliant (for some standard or other) radio signal as well.

But the critical point here is that the lane markings are, well, weird. There’s a diagonal stripe that cuts across the lane; when the lane is open, drivers are expected to ignore the line, but at other times, they’re supposed to follow it in merging into the next lane over.

How is the car supposed to know when to follow the line? (Come to think of it, how do current lane assist technologies handle that stretch of road?) How are the programmers prioritizing lane markings versus other signals?

Maybe, I thought, in ambiguous situations, the rule could be “follow the car in front of you”. That could work. Sooner or later, the chain of cars deferring to the next one forward will reach a human-driven car which can resolve the conflict. Hopefully that driver is experienced enough to get it right and neither drunk nor distracted by their cell phone.

But how are the cars going to know if the car in front of them is trustworthy–i.e. is following the same “follow the car in front of me” rule? Is your Toyota going to trust that Ford in front of it? Or will it only follow other Japanese manufactured vehicles? Maybe the standard can include a “I’m following the car in front of me” signal. But what if the signal changes in version 2.2a of the specification?

There’s a classic short story* in which cars and trucks have evolved from trains. Each manufacturer’s vehicles require a different shape of track and a different width between tracks. Some are nearly compatible, able to use a competitor’s tracks under certain special circumstances. As you might imagine, the roads are a mess, with multiple tracks on every street, except where a city has signed an exclusive deal with one manufacturer.

* Whose title and author currently escape me, darn it. If you recognize it, please let me know in the comments.

The story is an allegory on the early personal computer industry with its plethora of competing standards and almost-compatible hardware, but I can’t help wondering if we’re about to see it play out in real life on our roads.

Pay As You Go

Posted on by

The other shoe drops.

Remember last month when I pointed out that autonomous cars aren’t really intended for individual ownership?

I’ve been wondering how auto makers intended to push people away from car ownership. Prohibitive costs and regulation will only get you so far, after all. The people who can afford to buy a new car every few years–especially a luxury car–aren’t going to be bothered by the registration fee. And if you’re buying a car for the look or the name, you’re really not going to care that this year’s Lexus is an electric.

Well, today’s newspaper gave me the clue. The answer is car subscription programs. Pay a flat monthly fee, get a car, complete with maintenance, insurance, and the ability to swap it for a different model whenever you want*.

* Some restrictions apply, of course.

Some of the services are backed by auto makers–the article I saw talks about Canvas, which is a subsidiary of Ford and, logically enough, only offers Ford and Lincoln cars. Others, like Clutch Technologies, are independent to varying degrees. (Clutch offers high-end models from several manufacturers and concierge service.)

It’s not a lease. There’s no intent to own, and all of the costs except gas are included in the monthly payment. Some programs don’t even require a commitment longer than a month. That’s an attractive model to people who’ve gotten used to that approach with their cord-cutting TV service.

Try it for a couple of months. Don’t like the car? Send it back and get a different one (in some cases you’ll need to wait until the end of your billing cycle). Need something bigger for the weekend? Some services not only let you switch cars at any time, but they’ll even deliver it to your door and help you move your possessions from one to another.

Do you suppose they’ll transfer your radio presets? Wouldn’t surprise me a bit if they did. Be easy enough to have some kind of data transfer tool that downloads all your data–radio, seat position, preferred temperature–via Bluetooth. Horribly insecure, of course, but no more so than anything else in the car, so who cares?

But I digress.

It’s more expensive than buying a car, of course, even with the cost of semi-annual servicing and insurance factored in. And I really doubt that the insurance plans are as good as you could get from an independent insurance company–I’m sorry, but auto insurance is no more of a one-size-fits-all object than a spandex cat suit. Most of us might squeeze into it, but there are gonna be some bits sticking out here and there, and a few of us just plain need more.

So you’re paying for convenience and, arguably, flexibility.

And when the autonomous models start showing up in a few years, well, you’re going to feel pretty smug about your flat-rate subscription that means you don’t have to pay whatever Uber is charging by then. Never mind that Uber probably owns a good-sized chunk of the subscription company and makes the autonomous car you’re subscribed to.

The Wrong Question

Posted on by

There’s a new survey. Well, okay, there’s always a new survey about something. The one I’m talking about is from Cox Automotive, and purports to look at attitudes toward autonomous cars. More precisely, it looks at how attitudes have changed in the past two years.

Unsurprisingly, it shows that fewer people say they’d be interested in buying a fully autonomous car, one that has no option for manual control.

Cox attributes the decline to the publicity around the Uber’s testing and the death of a pedestrian earlier this year. And they’re probably correct in that assessment.

Also unsurprisingly, while three-quarters of the survey respondents say autonomous cars need real world testing, less than half would be willing to have the testing done where they live. Not quite NIMBYism, but certainly NOMS (Not On My Street).

Because people are poor at gauging risk. From what I can tell, even at their current state of development, autonomous cars are safer than manually-driven ones on a mile-for-mile basis. But self-driving vehicles are regarded with suspicion because they’re unfamiliar.

But I digress.

What I wanted to point out here is that Cox is asking the wrong question. Because makers of autonomous cars (or would-be makers) aren’t trying to sell them to the general public.

Why not? Because people aren’t buying cars as often as they used to. Twenty years ago, the average American bought a new car every three or four years. Ten years ago, it was every four or five years. Today, the average is closer to seven years.

Serving that market isn’t a sustainable model for an automaker.

Lest you think I’m guessing about that, consider the way fully self-driving cars have been pitched to consumers. The ads and opinion pieces have been heavy on the vision of your car dropping you off at work in the morning, then coming back to pick you up at the end of the day. Summoned, of course, by an app on your phone.

Where exactly is that car going after you get out?

To take your spouse to work? Maybe–but again, encouraging consumers to share a car isn’t part of automakers’ business plan.

To park? Unlikely. It’s already impossible to find a parking space near any large business, and if your car parks at a distance, it’s going to be tough for it to pick you up in a timely fashion.

Maybe it’ll go home. But do you want to pay for gas or electricity, not to mention depreciation, from an extra round trip every single day? Probably not.

No, what the manufacturers want is for you to send your autonomous car off to drive other people around. Remember, Uber is a prime mover in autonomous car development. The idea of millions of cars–that other people buy and maintain–working for them without obnoxious drivers who demand to be treated as employees must be causing enough salivation in upper management to fill a couple of Olympic-sized swimming pools.

And if you want your car to work for you while you’re doing other things, it’ll need to be up to the standards Uber (or GM, or whoever runs the service) sets. If they say it can’t be any older than last year’s model, you’re going to buy a new car every other year. And then pay for gas and maintenance out of whatever rental fee the company decides to pay you.

Or maybe they won’t bother with selling to consumers at all. Uber Motors can make the cars and sell them at a loss to Uber Rideshare. That puts UM in a great tax position. Meanwhile, UR gets tax breaks for capital investment while simultaneously writing off the cost of maintenance and depreciation. And UR also can charge an arm and a leg for ride service because fewer and fewer consumers are buying cars. Why? Not only do they not want autonomous cars, but UM’s price to consumers is several times higher than what UR pays.

And, lest you forget, there’s only so much reduction in pollution that can be attained by increasing the number of non-polluting vehicles sold. That means, sooner or later, clean air laws will mandate older, lower mileage vehicles must be removed from the road (or cost prohibitive amounts to register). Which means fewer people will be able to afford the upfront cost to buy a new car.

If public transportation isn’t an option–and it’s not for many people–UR will be waiting to collect your money.

So it really doesn’t matter whether you want to buy a self-driving car. You’re probably not going to. But you will be riding in one.

Lucky you!

Like a Two-Year-Old

Posted on by

“Whether it’s a ride, a sandwich, or a package, we use technology to give people what they want, when they want it.”

That seems to be the core of the message Uber wants to convey.

Quite a paean to entitlement, isn’t it?

With a mission like that, it’s no wonder Uber gets insulted when things don’t go their way. To be quite blunt, lately, Uber has been acting like a poorly socialized two-year-old.

In case you missed it*, Uber recently deployed part of their fleet of experimental self-driving cars in San Francisco. Unfortunately for everyone, they did it “Uber-style”.

* Though it’s hard to believe anyone could have missed it, given how loudly Uber tooted their own horn, and how many news sources joined into the noise-making.

See, under California law, autonomous vehicles need to be registered as such with the Department of Motor Vehicles. Additionally, the company testing the vehicles has to purchase a permit. For the record, as best I can tell from the newspaper stories, it’s not like the registration and permit would require Uber to break into their piggy bank. I can’t find anything that suggests the cost of registering a vehicle as an autonomous test model is any higher than that of a regular registration, and at least one story noted that the testing permit cost $150. Even if that’s per vehicle, Uber could certainly have paid it out of spare change.

However, Uber, declined to properly register their cars or to purchase the permits.

Hey, remember Google’s motto? “Don’t be evil,” wasn’t it? I begin to suspect Uber’s motto is “Don’t be good.” But I’m sure that’s just my cynical streak talking.

Anyway, when the DMV called Uber’s oversight to their attention, they declined to rectify the omissions, claiming that because the cars are not actually capable of driving without human supervision, they’re not actually “autonomous”.

While Uber and the DMV traded legal opinions and insults, several of the cars were caught on video running red lights and making dangerous turns. Which suggests that either the cars are autonomous–and desperately in need of testing and debugging–or that their human supervisors could use a couple of rounds of QA testing.

In the end, Uber displayed the priceless maturity we’ve come to expect from them. They’ve picked up their toys and stomped off to Arizona where, presumably, nobody cares if a few pedestrians or bikers get crushed beneath the wheels of progress.

This small corner of the Bay Area may have handed them a temporary defeat, but Uber’s executives and investors can relax. In the rest of the world, it’s still “Uber über alles”.

Small Bites

Posted on by

A collection of small items that don’t seem to warrant entire posts of their own.

Engadget reported last week that, as their headline put it, “Researcher finds huge security flaws in Bluetooth locks”. Briefly, he found that twelve of sixteen locks he bought at random had either no security or absolutely horrible security. That doesn’t mean, by the way, that those remaining four locks are safe, just that the researcher, Anthony Rose, didn’t immediately find problems.

Does this come as any surprise? It shouldn’t. Given how often we’ve seen Internet of Things manufacturers give no thought whatsoever to security, the surprising thing is that four of the locks weren’t trivially hackable.

Police and alarm manufacturers will tell you that it’s impossible to actually secure your house against a break in. The goal is to make it a harder target than your neighbors’ houses. Clearly, your best bet today is to buy a bunch of Bluetooth locks–and give them to all your neighbors!

Moving on.

I said that the new Ghostbusters movie wasn’t doing as well at the box office as it deserved. Apparently Sony agrees. According to Gizmodo (among many sources), the direct loss–before figuring add-on income from licensing and merchandise–could be as much as $70 million.

As a result, plans for a sequel are on hold. Instead, Sony is focusing on an animated TV show for 2018 and an animated movie for 2019.

OK, yeah, animation is potentially cheaper than live action, especially if you don’t have to pay full price for the actors. But it does rather make Ghostbusters something of a second-tier property.

And if you’re the betting sort, the smart money says neither the TV show nor the movie will feature the women who starred in this year’s film–and then, if the animation does well, it’ll be held up as further “proof” that women can’t carry a movie without male help.

Complete change of subject.

Audi is going to launch a new feature in some of its 2017 cars. Correction: IMNSHO, it’s a misfeature. They’re going to add a countdown timer on the instrument panel and heads-up display to let drivers know when red lights will turn green.

Seriously. And if Audi does it, you know everyone else will follow suit.

I don’t know how people drive where you are–or near Audi headquarters–but around here, people stretch yellow lights well beyond any rational limit. Give drivers a timer, and they’re going to accelerate as soon as it hits zero, without even looking at the traffic light, much less checking for oncoming traffic that didn’t even enter the intersection until their light was red.

The only way this could even begin to be sensible or safe would be if automakers lock out the accelerator (and horn!) until the onboard sensors confirm that the light is green, the car in front (if any) is beginning to move, and there’s no vehicle in the intersection. I regard this as highly unlikely to happen.

So, given my grumpiness in regard to new technological “advances,” you may be surprised to hear that I’m strongly in favor of this next announcement.

According to Ford CEO Mark Fields, the company is actively developing fully autonomous cars intended for ride-hailing services. They expect to have them on the market by 2021.

I’ll be blunt here: I dislike taxis and their modern would-be successors in large part because there’s no way to know whether the driver will (just to pick a few examples at random) cross solid lines changing lanes, speed, use the mirrors before changing lanes, or come to full stops at red lights and stop signs.

There’s no guarantee that an autonomous car will drive any better than any random human–and, putting on my QA hat for a moment–you can be certain that every single automaker’s self-driving car will have buggy software.

But at least autonomous cars will be more consistent. Get in a car that drives itself, and you’ll know what to expect from the driver. I find that idea soothing.

Finally, I don’t know whether to laugh or cry about this last item.

It seems that the Hacienda Mexican Restaurant chain in South Bend, Indiana thought it would be a good idea to put up billboards advertising their food as “The Best Mexican Food This Side Of The Wall.”

The signs are coming down. According to Executive Vice President Jeff Leslie, the company “didn’t expect the backlash.”

Let that sink in for a moment. This is a chain of Mexican restaurants that’s so out of touch with Hispanics, that they thought associating themselves with Trump’s Wall was a good advertising strategy.

I know the connection between an ad and the product it’s hyping is tenuous at best, but this really takes the tortilla. If the company has that big a disconnect with its roots, what are the chances that it’s food is any good at all, much less the best north of Nueva León? Small bites, indeed.