Self-driving car discussion catch-all

OG, see my last post (well, the one above the GamePro graphic).

Gremlin wrote:

Something I've long suspected: self-driving cars that refuse to go above the speed limit will annoy a rather large group of people.

Just wait until people get the eventual insurance bill that charges different rates based on your 'pool' of self-driving or human-driven vehicle. I think once we get past the "is this legal and could an accident bankrupt me" stage of self-driving cars, how long before no one can financially afford the risk of driving themselves?

cheeze_pavilion wrote:
Gremlin wrote:

Something I've long suspected: self-driving cars that refuse to go above the speed limit will annoy a rather large group of people.

Just wait until people get the eventual insurance bill that charges different rates based on your 'pool' of self-driving or human-driven vehicle. I think once we get past the "is this legal and could an accident bankrupt me" stage of self-driving cars, how long before no one can financially afford the risk of driving themselves?

I was thinking just wait until they get Netflix installed in the car. People will choose to take the slowest longest route so they can binge watch Breaking Bad.

farley3k wrote:

I was thinking just wait until they get Netflix installed in the car. People will choose to take the slowest longest route so they can binge watch porn.

FTFR (Fixed That For Reality)

With dark tinted windows, of course.

I can only assume the first sex in a self driving car has already happened.

It seems the self-driving car has Cleveland as a potential destination.

Frankly, the Swedes would be balancing out their karma if they made a car that found its own damn parking at IKEA.

DanB wrote:
Demyx wrote:
Gremlin wrote:

Something I've long suspected: self-driving cars that refuse to go above the speed limit will annoy a rather large group of people.

Self-driving cars would hopefully allow us to raise or even abolish the speed limit.

Which will absolutely lead to 'pay more to get there faster' without regulatory oversight. Also speed limits are often in place to keep non-car road and sidewalk users safe. Stopping distance is stopping distance whether are computer is driving or not.

It's not that the self-driving cars would be able to drive as fast as possible on residential roads, it's that they'd theoretically be a better judge of the maximum safe speed than humans are. These days we have static speed limits regardless of traffic, amount of pedestrians, weather etc. A self driving car really shouldn't need to remain under that static speed limit if there is no one around.

Especially on highways where you are less likely to have unpredictable obstacles, you could effectively raise the speed limit.

Self driving cars are going to be an intellectual property minefield. Should car owners be allowed to customise the software running their car? If not then the current John Deere tractor licencing/owning debacle appears to be the first shot across the bows on this sort of issue. If yes then making the system safe is going to be crazy complicated.

I'm more surprised that no one is yet concerned that self-driving cars would probably be trackable.

The most legitimate concern about self driving cars, as Demyx alludes to, is the privacy issue. Though it is important to note that with increasing surveillance and license plate tracking, even that is an issue that is receding into the margins as any real expectation of privacy is at best cosmetic anyway.

Folks who think that human piloted automobiles are in any real way "safer" than automated ones have an unrealistically inflated view of their own ability and/or attentiveness. If you have ever answered a telephone call, read a text message, or adjusted the radio or environmentals in your car while moving, you immediately disqualify.

Even the issue of unseen objects like pedestrians, wildlife, or backing vehicles is a much easier fix with an automated vehicle than a human one. Our human brains are tremendously single threaded and are simply terrible at handling out of parameter inputs even when they are obvious and significant.

This famous test perfectly illustrates this point.

This is not the sort of thing a computer with 360 sensors (possibly including infrared and high frequency sonar) is likely to fall prey to.

I passed that test. :p

Spoiler:

I even said why the f*ck is there a gorilla? while I kept counting passes.

Stele wrote:

I passed that test. :p

Spoiler:

I even said why the f*ck is there a gorilla? while I kept counting passes.

I.. did not.

Spoiler:

I got an extra pass, and totally missed the gorilla. To be fair, I'm already diagnosed with ADD :)

I think my biggest concern with self driving cars is overreacting to an incident.
Do not get me wrong. I am a huge proponent for self driving cars. They can't come soon enough.

But I would love some sort of hybrid system where piloted cars can link up like train cars on the freeway and self drive for more efficient traveling (space and fuel consumption)

fangblackbone wrote:

I think my biggest concern with self driving human driven cars is over underreacting to an incident.

Yeah, not so much.

GOOGLE CARS’ 1 MILLION MILES WITH 0 ACCIDENTS UPSETS MEDIA

According to Urmson, in the last six years Google’s self driving cars have driven themselves at least a million miles. There have been twelve accidents, but some occurred when a human was driving (humans drove the cars an additional 700,000 miles). All were minor incidents with no injuries and most of these were other drivers hitting or rear-ending the Google car. This sounds like great news that we should be excited to hear. Zero accidents, injuries, or damage has been caused by er, auto-autos over the course of a million miles and a half dozen years. Beats my record.

But much of the media focuses on or attempts to fabricate fear, uncertainty and doubt. One might cynically believe the media simply lives and breathes fear because that is how you get paid in the news business, but I think there is more going on here. Watching how hard and how fast they bend over backward, how they squint at the facts in seeming desperation to find some way to be afraid of the future leads me to suspect the cultural shift to a world with machine drivers is scary because it is different, and because Americans fear change just slightly more than giant spiders that drool Ebola.

A quick review.

Self-driving cars can’t avoid accidents on California Roads, Associated Press

Google Blames Humans for Accidents Involving Its Self-Driving Cars, Time

Google’s self-driving cars have been getting in accidents in California, Business Insider

Google’s driverless cars have been involved in three car accidents, qz.com

Google’s self-driving cars are getting into accidents, MarketWatch

This is terrible journalism designed to upset you. Google’s program and the data we have are cause for great optimism, and at this point, only optimism! I make no claims this can not change, but right now, we’re golden. Here’s why.

1. The machine driver caused zero accidents. I know I said this already. It bears repeating. None. No times.

2. The “high” accident rate is new evidence about how bad humans drive. The Google cars are required to report absolutely every incident, no matter how trivial, how minor or absent the damage. People do not. The higher rate is probably a truer estimate of how much more often people cause minor accidents versus common estimates. This means machine drivers could be even better for safety and money-saving than we presently estimate.

3. No major accidents at all. The speeds were 10 mph or less in almost all cases. Again, zero injuries. Fantastic.

4. This is how airplanes have worked for years. Many can land without human assistance, and have been able to for a long time. Granted, there are fewer jets in the air than cars on the ground, but even if you controlled for congestion, there is no question that air travel and transport are almost unbelievably safe at 3 deaths per 10 billion passenger miles. Safety stats have improved as machine-assisted technologies have become more and more integrated. In 2009 there were 4.1 accidents per one million departures world-wide. In 2013, just 2.8. According to the Internation Civil Aviation Organization, 2014 was the safest year in history (pdf 2014 safety report).

5. This is beta testing. The machine drivers already have a stellar record, but are at least a couple of years away from commercial release. In that time they will surely get even better. So will their sensors, their “eyes”. Many products and technologies we rely on today started out far inferior to what we are used to. Solar panels used to be laughably inefficient and expensive. Microsoft Windows was universally detested until 3.11 (and by many still, until Win95). The first cars were horrible things; loud, dirty, breakdown-prone, expensive messes.

Why is it now we’re so averse to giving a new technology time to mature?

The only story any reputable media source should have reported is how excellent this technology, beta though it is, is performing and how it is highlighting just how badly we need it (when objectively tracked, we’re seeing how often people really cause accidents). It’s absurd for Google to be on the defensive, here. I am no shill for Google. Its failures and mis-steps are numerous and strikingly awful (Buzz, Wave, Glass, and arguably G+ to name a few). But they really deserve credit here for a fantastic research and development program that is going to radically change our lives for the better.

I had heard at some point that most of the hours/miles the google cars were logging were on a google campus controlled course. Are they properly in the wild? The only hole I can picture poking in that story would be if too high a percentage of that million miles was overly controlled.

The article linked by the story above gives a lot of interesting detail and some cool pictures of the computer visualized behavior and predictions going on in the car: https://medium.com/backchannel/the-v...

It doesn't specify a breakdown of how the driving time is divided between controlled test conditions and actual driving in the wild, but it sounds like there's pretty substantial experience of real world driving by the cars.

I'm personally quite ready for a fleet of auto-taxis to exist. My city has Car2Go, where you can pick up any one of a fleet of available cars if you can find one near you, drive it anywhere you like within the greater Seattle area and only get charged a few bucks per session. Give me a version where I can just request a car, it shows up, takes me where I want to go and can do the same to get me back (So Car2Go + Uber - Real people) and I'll be all for it.

Nissan Will Have Self-Driving Cars By 2020, CEO Says

Nissan Motor Co. will have vehicles packed with autonomous driving technology by 2020 but whether people will be able to drive them on roads is up to government regulators, Chief Executive Carlos Ghosn said Monday.

Many of the world's automakers, and companies outside the auto industry such as Google, are working on technologies that allow cars to navigate without human intervention.

Ghosn said good progress is being made toward the Japanese automaker's goal of delivering that technology by 2020.

"Our cars will be ready," he said.

Our autobodies are ready...

I tend to find that obeying traffic laws increased the chances of being rear-ended. You know, things like stopping at intersections or slowing down for emergency vehicles on the side of the highway.

farley3k wrote:

GOOGLE CARS’ 1 MILLION MILES WITH 0 ACCIDENTS UPSETS MEDIA

According to Urmson, in the last six years Google’s self driving cars have driven themselves at least a million miles. There have been twelve accidents, but some occurred when a human was driving (humans drove the cars an additional 700,000 miles). All were minor incidents with no injuries and most of these were other drivers hitting or rear-ending the Google car. This sounds like great news that we should be excited to hear. Zero accidents, injuries, or damage has been caused by er, auto-autos over the course of a million miles and a half dozen years. Beats my record.

But much of the media focuses on or attempts to fabricate fear, uncertainty and doubt. One might cynically believe the media simply lives and breathes fear because that is how you get paid in the news business, but I think there is more going on here. Watching how hard and how fast they bend over backward, how they squint at the facts in seeming desperation to find some way to be afraid of the future leads me to suspect the cultural shift to a world with machine drivers is scary because it is different, and because Americans fear change just slightly more than giant spiders that drool Ebola.

A quick review.

Self-driving cars can’t avoid accidents on California Roads, Associated Press

Google Blames Humans for Accidents Involving Its Self-Driving Cars, Time

Google’s self-driving cars have been getting in accidents in California, Business Insider

Google’s driverless cars have been involved in three car accidents, qz.com

Google’s self-driving cars are getting into accidents, MarketWatch

This is terrible journalism designed to upset you. Google’s program and the data we have are cause for great optimism, and at this point, only optimism! I make no claims this can not change, but right now, we’re golden. Here’s why.

1. The machine driver caused zero accidents. I know I said this already. It bears repeating. None. No times.

2. The “high” accident rate is new evidence about how bad humans drive. The Google cars are required to report absolutely every incident, no matter how trivial, how minor or absent the damage. People do not. The higher rate is probably a truer estimate of how much more often people cause minor accidents versus common estimates. This means machine drivers could be even better for safety and money-saving than we presently estimate.

3. No major accidents at all. The speeds were 10 mph or less in almost all cases. Again, zero injuries. Fantastic.

4. This is how airplanes have worked for years. Many can land without human assistance, and have been able to for a long time. Granted, there are fewer jets in the air than cars on the ground, but even if you controlled for congestion, there is no question that air travel and transport are almost unbelievably safe at 3 deaths per 10 billion passenger miles. Safety stats have improved as machine-assisted technologies have become more and more integrated. In 2009 there were 4.1 accidents per one million departures world-wide. In 2013, just 2.8. According to the Internation Civil Aviation Organization, 2014 was the safest year in history (pdf 2014 safety report).

5. This is beta testing. The machine drivers already have a stellar record, but are at least a couple of years away from commercial release. In that time they will surely get even better. So will their sensors, their “eyes”. Many products and technologies we rely on today started out far inferior to what we are used to. Solar panels used to be laughably inefficient and expensive. Microsoft Windows was universally detested until 3.11 (and by many still, until Win95). The first cars were horrible things; loud, dirty, breakdown-prone, expensive messes.

Why is it now we’re so averse to giving a new technology time to mature?

The only story any reputable media source should have reported is how excellent this technology, beta though it is, is performing and how it is highlighting just how badly we need it (when objectively tracked, we’re seeing how often people really cause accidents). It’s absurd for Google to be on the defensive, here. I am no shill for Google. Its failures and mis-steps are numerous and strikingly awful (Buzz, Wave, Glass, and arguably G+ to name a few). But they really deserve credit here for a fantastic research and development program that is going to radically change our lives for the better.

I suspect that the answer is far more straightforward than that.

Who sponsors American journalism?

Mantid wrote:

I tend to find that obeying traffic laws increased the chances of being rear-ended. You know, things like stopping at intersections or slowing down for emergency vehicles on the side of the highway. :?

Last time I was rear-ended, it was because I had stopped for a red light.

Gremlin wrote:
Mantid wrote:

I tend to find that obeying traffic laws increased the chances of being rear-ended. You know, things like stopping at intersections or slowing down for emergency vehicles on the side of the highway. :?

Last time I was rear-ended, it was because I had stopped for a red light.

The only time I've been rear-ended, it's because I stopped for a yellow light and the person behind me assumed I was going to blow through it.

Ugh... news.

CNN front page headline: "Self-driving cars nearly collide"

Click the article, and it's "Self-driving cars from rivals Google, Delphi in close call"

Then some drivel about people getting cut off in a lane and giving the other driver the finger. And finally, four paragraphs in:

Spokeswoman Kristen Kinley insists it wasn't truly a near miss but was instead an example about how self-driving cars effectively avoid accidents.

"Our car saw the Google car move into the same lane as our car was planning to move into, but upon detecting that the lane was no longer open it decided to terminate the move and wait until it was clear again," she said.

Stele wrote:

Ugh... news.

CNN front page headline: "Self-driving cars nearly collide"

Click the article, and it's "Self-driving cars from rivals Google, Delphi in close call"

Then some drivel about people getting cut off in a lane and giving the other driver the finger. And finally, four paragraphs in:

Spokeswoman Kristen Kinley insists it wasn't truly a near miss but was instead an example about how self-driving cars effectively avoid accidents.

"Our car saw the Google car move into the same lane as our car was planning to move into, but upon detecting that the lane was no longer open it decided to terminate the move and wait until it was clear again," she said.

Yeah. Very much a "Man briefly contemplates biting dog and thinks better of it" story.

Google’s Self-Driving Car Involved In Collision Resulting In Injuries For First Time

Sounds bad....until you read it.

According to an accident report that Google filed with the California Department of Motor Vehicles about the July 1 crash, the driverless SUV was going about 15 mph in self-driving mode behind two other cars, as they all approached an intersection with a green light. When the first car slowed to a stop so it wouldn’t block the intersection, the other car and Google’s vehicle also stopped.
About a second later, a fourth vehicle hit Google’s car from the back at about 17 mph, with the driverless car’s on-board sensors showing the other car did not brake.

So the car won't protect against other drivers being dumb....

farley3k wrote:

Google’s Self-Driving Car Involved In Collision Resulting In Injuries For First Time

Sounds bad....until you read it.

According to an accident report that Google filed with the California Department of Motor Vehicles about the July 1 crash, the driverless SUV was going about 15 mph in self-driving mode behind two other cars, as they all approached an intersection with a green light. When the first car slowed to a stop so it wouldn’t block the intersection, the other car and Google’s vehicle also stopped.
About a second later, a fourth vehicle hit Google’s car from the back at about 17 mph, with the driverless car’s on-board sensors showing the other car did not brake.

So the car won't protect against other drivers being dumb....

10-1 says that the human driver was texting.

Yet another example of why humans shouldn't drive.

Since much of the discussion in here is about Google and Oracle cars, maybe we should rename this thread to "self-driving car discussion catch-all" or something like that.

As you wish!

Paleocon wrote:

Yet another example of why humans shouldn't drive.

Exactly. Can't get these cars on the road soon enough.