The end of the human car driver is near ?

[Mihai Lihatchi]

Since I took my driver's license in 1999 I expected to be in the last decade of human driving.
Right now here is the status :
http://www.virtualworldlets.net/Resources/Hosted/Resource.php?Name=SelfDrivingAI2010
We are not there yet .
Why we need less human drivers ? There are thousands of reasons but here are a few
1) Driving is a very error-prone repetitive task ..with deadly consequences. The number of deaths due to car accidents is about 40.000 in US only.
Let's not even mention the wounded and disabled ...
2) Huge ineffectiveness in traffic. My opinion is that AI cars can be able to coordinate better without traffic jams and less damage to
car components.
3) Lost time : I used to do about 2h/day in my car listening to the radio .. now I am spending only 1h 20 biking. Most of the time I was very upset
and angry. Looking for a parking space and walking to work sometimes took 30 minutes.
4) Last but not least : Some people really are unable to drive , they are too old , too drunk or too young. Few of them are unable to bike or walk (the elder) . Some people are
too scared to drive (mostly women .. but I won't generalize) and can't use the car. They are unable to go shopping or to leave their area and are disadvantaged when looking for
better solutions to their problems because the roads are made for cars and the distances are sometimes huge.

Don't get me wrong , I used to be a car lover , I could name every BMW , Audi or VW car made in the last 20 years together with engine characteristics , top speed , 0-100 km/h .. it just seems so .. 20th century right now ..
I used to love driving fast too .. but learned better. Not even the simplest solution of limiting the top speed in cars was applied.
In my country you are supposed to drive with 130 km/h on the highway , 90 km/h on national roads and 50 km/h in residential areas. Too often cars speed sometimes twice as fast on these roads. It's true some people are ticketed but the rest are unharmed. Why can't we at least limit the top speed in vehicles ?
Right now I am a bike commuter (almost year-round) and I am a lot more relaxed . I still drive but only for long distance trips or on holidays. Still I would love to not loose my 4h driving but instead reading a book and still getting there.
I am looking forward to just telling my car : Take me to Paris and get some sleep on the way.
As a side-note .. what happened to the Skycar project ? (http://en.wikipedia.org/wiki/Moller_Skycar_M400) guess it never took off.

[Pat Farrell]

Mihai Lihatchi wrote:Why we need less human drivers ? There are thousands of reasons but here are a few
1) Driving is a very error-prone repetitive task ..with deadly consequences. The number of deaths due to car accidents is about 40.000 in US only. Let's not even mention the wounded and disabled ...

I agree that its inevitable that computers take over much of the effort of driving. But I think it will happen slowly over the next 50 years or so.

Your death rate number for the US looks about right, but you are making a logic error claiming that a computer driven car would avoid all of these. First, if the control software is running Windows, you have to add in the number of accidents caused by Blue Screen Of Death. More importantly, some are caused by external conditions, weather, ice, etc. and some are mechanical. Perhaps the computer would be smarter, and slow down or even park when the weather is really bad, but would the block-headed human stay parked?

Before any automated driving can be accepted, we first have to have massive accidents caused by hackers/crackers pumping malware into the car software. Then we will have decades of law suits. Then the engineers can harden the software. They it can begin to be accepted.

In the meantime, there are trains.

[Mihai Lihatchi]

Well I know about trains , I also know about public transportation. It does not replace the privacy and comfort of your own cars.
I once returned to Bucharest during a very dense fog in the middle of the night. The plane(a Boeing 737) had to land with about 240-250 km/h ... you could not see more than 10 m ahead of your.
As much as I respect the pilots I do not think they would have managed without the ILS.
During the 15 km road trip to my home the taxi driver could barely reach 50 km/h (and even he claimed it was dangerous weather for driving).
I just hope that people will realize sooner rather than later that driving is not a sport and that is more like work then like fun and a low paying sweatshop type of work ...

Windows ? Who uses Windows ? Most of the companies have *.nix servers for their critical stuff.
I am on Linux (Ubuntu) most of the times I am not playing.. let's remember that unlike most enterprise software (CRUD stuff) we can have very low rate defect software ( I remember code at NASA having something like 1 bug /10000 line of code) if we are willing to invest time ( i.e. money) and effort into it.
What I don't see is what is the technical bottleneck preventing this to happen ( I can see plenty of psychological and legal issues) and I wonder what can we software engineers do about this.

[Campbell Ritchie]

Pat Farrell wrote: . . . if the control software is running Windows, . . .

You can't count them. The limits of the 32-bit binary integer are too small.

. . . there are trains.

Some of which ae driven by computer already, in this country.

[Pat Farrell]

Campbell Ritchie wrote: . . . there are trains....Some of which ae driven by computer already, in this country.

Even in the US? Are you sure? I have not kept up, but I know that the DC Metro was run by PLC logic ladders at least for the first 30 years. Part of why there are seats that face the center of the car (rather than fore or aft) is to hold the PLC relays. They are/were definitely not solid state.

A lot of process control guys do not have confidence in these new fangled computer things.

I've love to see a list of which train systems are actually computer controlled, rather than PLC logic driven. The BART system was designed/built just a few years after DC's Metro, I wonder if its PLC too.

[Ernest Friedman-Hill]

My experience suggests that the number of human car drivers is actually quite small already. Australopithecenes seem to dominate in my local area, followed closely by parasitic Blackberries, generally still clinging to the heads of their zombified hosts.

[Mihai Lihatchi]

I did not mean to say that but I know that behind the wheel people change their personalities a lot ... and not for the best. And like it or not a computer is not likely to take a phone call while driving at 100 km/h in heavy traffic. Some people will ... some won't , some oare overconfident in their driving abilities some are too scared.

Regarding trains I know for sure that in France at the Charles de Gaule Airport they have A.I. controlled trains between the two terminals.
Didn't a human error caused the train crash which occurred recently in Washington ? This year if I remember correctly ?
The bottom line for me :
More cycling == less blood pressure . But today it rained maybe I see a velomobile in my future.

[Greg Charles]

I heard of a study some years ago, where they simulated automatic driving on freeways, and returned manual control on the offramps for surface street driving. That seems like something we could plausibly see in our lifetimes, more than automatic driving everywhere you want to go. Unfortunately, during the study, a gruesome percentage of people crashed immediately when they were force to take over driving. I don't have a cite for that study though.

[Mihai Lihatchi]

Well guess what :
http://techcrunch.com/2010/10/09/google-car/

Google is on to the race and it looks like in 8 years (say a decade) we will have a working AI driven car.
Of course they wouldn't have done it without Stanford's AI Lab and it's not clear to me whether it's Stanford's Audi TT (well VW's actually) or Google's own Audi .
Anyway I am hapy they are working on this and hope I will not have to renew my driving license (expires in 2018 ).

[Chris Baron]

DARPA,
the inventors of the internet, had two prize competitions:

the Grand Challenge in 2005 where robotic vehicles had to drive through a desert.

and the Urban Challenge in 2007 where the vehicles had to interact completely autonomously in traffic in an urban setting and comply to the traffic laws.

DARPA is an agency of the US Defense Dept and they have an interest in pushing the development of robotic vehicles.
And because Stanford won in 2005 i guess the Google car is the civil usage of something that has been developed for them.

But i also believe that the development in direction to the self-driving car will be a slow process. And it will be introduced step by step as drive assistants like ABS, EPS, etc.
While it may sound fascinating to us techies, for most people it's a terrifying idea to give away control completely. And the car makers know that.

[Peter Rooke]

I don't want one! When I do I'll take the train or bus.. I much perfer the motorbike

[Hussein Baghdadi]

I hope not, driving a car you like is like nothing else.

[Allan A Peak]

I don't think I trust AI to handle driving. I know humans aren't that great either (I have a banged up bumper to remind me of that), but driving just seems too unpredictable for computers to handle.