Slashdot Mirror
GM Claims Advanced Cruise Control By 2008
pavelvp writes "Economist is reporting that General Motors is working on the prototype of the self-driving car. From the article, "The car uses updated technology combined with several existing innovations and, according to the manufacturer, could be in production by 2008. But, while the technology takes some of the boring bits out of driving, it falls far short of an automatic taxi service and, anyway, various legal, technical and social barriers to its introduction remain.""
Can I push the gas pedal down hard enough to triumph over the computer system? I would really hate to see a computer have more control than the human. We've seen plenty of movies, where that happened.
relinquish all control of our cars once we enter the freeway
Although your numbers are probably a joke, the actual theory merits discussion. Still, it won't happen for quite a while. People enjoy driving far too much. Movies like I, Robot and others involving "manual overrides" are actually fairly close to the truth. You think all the owners who buy sports cars, whether it's a base-model ford mustang or a top-end ferrari, are going to be satisfied "turning over all control" of their car and just reading the newspaper or watching a movie on the way to work? The only way this highway works is if every single car participates, otherwise the stupid human will be bumping into the super-fast computer-driven cars all day. Good luck getting buy-in from 100% of drivers. Not during my lifetime.
It seems as if the ultimate (admittedly, far off) goal of such developments is to establish a virtually automated mass transit system using today's road networks. When cars automatically take the correct routes, make correct lane changes, and communicate with each other to ensure safe distances and traffic queues the joy of driving, for those who value it, will be essentially lost. I can't help but wonder if the joy of driving will be entirely lost in 20-50 years.
Business Voyeur
or computer only lanes/sections ala car-pooling?
they can have a much higher density, along with less braking idiocy, or the "look at that crash" which stops traffic for miles. the govt can make that an incentive to have auto-cars.
The first rule of USENET is you do not talk about USENET.
Also, gone will be the 'sporty' designs in cars, especially the dash. There are already designs for cars with 'couplings', so they work like train carriages in built-up areas. Your car will resemble a small train compartment more than a Ferrari.
On the other hand, at least it will rid the world of this lot.Huh? ABS was first implemented by BMW and airbags are definately a US invention from the 1950s, so you are probably wrong on your remaining point too... Ooops...
Oh well, what the hell...
It would not be unreasonable to make it legally required to use such a system on some highways. You don't like it, take the side streets.
Nice thought experiment. In reality? Good luck getting your congressmen and senators to vote for this. First the auto-lobby will scream bloody-murder because it would require extra systems in every car, which raises their costs. They sure as heck aren't going to eat into their own profits, so that means the price increases are passed along to the consumers, who want to know why their honda accords now cost $35,000 for a feature they don't want anyhow.
This doesn't even take into consideration the ACLU who will claim this is a movement to make cars inaccessible to poor people, as well as the far right who doesn't want no stinkin' computer driving when they can do it perfectly well themselves.
In summary, you have a system that is popular with about 18% of the population. Yep, I bet that one races through the legislature.
From the article:
For example, self-steering cars are currently illegal in most European countries. Carmakers want the law changed to allow them, but they are also keen not to be held legally responsible for any accidents which result. Drafting legislation which would make it attractive for carmakers to introduce the technology, but still allow some recourse for those hurt if something goes wrong, could prove tricky.
I see, they want to build self-steering cars but take no responsibility for what they have built going wrong.
I think alerting systems are a great idea - if you think I'm shifting lanes or following too close or not noticing a braking driver in front of me, by all means help me out.
But don't take away control based on incomplete sensor input. The distance that it is safe to follow another car depends on conditions. What if you reach a sharp bend and that other car can corner a lot better than you? What if you stop paying attention to the road ahead because the car is following another, and that other car fails to notice traffic stopping too quickly ahead? That car hitting anything else probably means you're going to hit it.
The consider the whole business of automatically following lanes as the GM car does. Has no-one out there seen roads after construction is through that have remains of old lines? What happens when a car starts AUTOMATICALLY correcting your steering to follow the old lines instead of the new?
To give auto makers a free Get-Out-Of-Jail card because they have the hubris to think they can build something that really drives better than a human is madness. Let them try but they must pay for failure so they are properly diligent as to issues that may arise.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
I hear concern that it MIGHT happen is why they stopped calling it "cruise control" and started calling it the more accurate "(automatic) speed control".
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
Without quoted statistics this reply is of course conjecture, as is yours.
I believe very few people enjoy the act of day to day driving, sports cars are a minute proportion of road traffic. What I and many others enjoy is the comfort and convenience of using a private car to travel. Auto-drive cars can increase these benefits by the following:
- Removing requirement for continuous, dedicated, concious control. Instead, read slashdot or watch I Robot as you travel to work.
- Removing necessity for private vehicle ownership, instead rent use of a class of vehicle - no need to find parking, drive straight to your destination, get out and the car will route/drive itself to a holding area or pickup.
- Safer my faith in computer control is greater then my faith in millions of bored, distracted humans. Provided the system is built up over time, slowly, a few features integrated to a trusted (proven) platform at each revision, as cars today are develop.
- Faster, with many vehicles coordinating with one another, it should be possible to increase thoughput and aggregate speed. As you say these will be mitigated by human and other uncooperative (incompatible) drivers. But it shouldn't be all or nothing.
Auto-drive cars won't appear overnight - enabling features will accrue over years - Sat Nav, cruise control, rangefinding sensors, lane drift alarms, drive by wire, braking assistance, ubiquitous wireless communication etc.
Thats my optimistic view, anyway.
Regards
Alex
Maybe, but the thing I'm really interested in is not can the software be bug-free, but can it be more reliable than humans?
The article says:
The funny thing is that I feel a lot less safe because other people, people I have no control over, are in control of their vehicle. The sad fact is that so many people are so mind-numbingly stupid behind the wheel of a car that I would much rather trust software that's slightly buggy and causes a few fatal crashes a year than humans who are outright stupid that cause almost 40,000 fatal crashes a year. And yes, if that means giving up control of my car so that other people have to give up control of theirs, I'm okay with that.
I think you would be more lucky if you asked to have dedicated lanes set aside for the Robot cars.
Some lanes for the advanced cars, the rest for older cars and people who want to drive manually.
Everyone wins.
Remember folks, slashdot doesn't have a -1 "disagree" moderation!
Yep, and GM is also pioneering hydrogen powered vehicles...
... while the Japanese are (again) dominating with hybrids.
Where's your nearest Hydrogen pumping station ?
Hydrogen is a pipe dream, and a dangerous one at that -- extracting hydrogen from water (destroying the water in the process) is a recipe for the end of life on Earth. The returned water is, as I understand it, less than what is net removed.
If GM continues to live in the Surreal World the price that will be paid will be the ultimate demise of The General. It's not as unimaginable as you might think. The analysts say the clock is already ticking...
Regards, Lex
"Chrysler is OWNED BY Mercedes."
You are, of course, correct. GM and Daimler-Chrysler are fierce competitors; perhaps even more so now that Chrysler/Dodge designers have access to the Mercedes parts bin. Today you can find Distronic in Mercedes-branded cars; perhaps in a few years it will be in Chrysler-branded minivans. All that GM can do for the time being is take diversionary tactics by talking about what they'll have several years down the road.
Sitting in my day care, the art is decopainted.
If a truly self-driving car ever does come around, it would not only revolutionize our lives in terms of commuting, but also in the movement of goods. Why even have stores when you could let customers select what they want online and then immediately dispatch a cartload of goods to their house? If this comes around the time of some sort of renewable electric car, we may find ourselves living in ever more sprawling low-density cities. What trouble is it to live 50 miles outside of town when you can get into town in no time while reading the paper in your own private vehicle?
Why don't you RTFA before posting. The cruise control designed by GM can actually steer.
Not to mention it also mentions how some pieces of the proposed system (cruise control including braking control, using radar/laser distancing controls, alarms on lane drift) occur in some cars, but not all of them together.
In other words, had he RTFA, he'd just pretty much be moderated -1, redundant.
Not sure how we got on the topic of hybrids, but while we are, there was a fantastic article if one of the automobile monthly magazines.
The author was talking about the fact that a hybrid needs to be driven for something like 66,000 miles per year to realize a savings of fuel costs versus price of the car.
He also went in to the eventual disposal of the giant battery/batteries in these cars. Not a pretty picture.
And perhaps most interesting was this. While the Toyota Prious (sp) is a hot seller, the other hybrids are not. The Toyota is a hot seller because it LOOKS like a hybrid an appeals to people wanting to make a statement. The Honda Accord hybrid looks like a regular Honda Accord so they are not selling nearly as well.
Anyways, I did a ton of paraphrasing, but the author made some very interesting points. Don't expect hybrids in their current form to ever take off.
You're right that it's not as easy as it looks (not by any stretch), but the FAA charts are incredibly useful. Yes, when looking at highways and rivers it can be tricky to figure out just where you are relative to the highway or river, but the charts have bridges and large structures on them, as well as power lines, towers, and anything else that is identifiable from the air. I'm a guy who loves maps, and FAA sectionals are some of my favorites.
Flying by dead reckoning is not easy, but it is the fundamental skill of any pilot -- any jackass can read an altimeter, for example, but it takes a special skill and lots of practice to accurately determine altitude and speed by looking out the window (let alone being able to accurately measure wind direction and speed by looking out the window and at the airspeed indicator, or knowing the RPM by the sound of the engine).
Before I did my first cross country-solo, I went up with my instructor and we did about an hour with me using the hood -- attitude and flight control by instrument. Part of this invloved assuming various headings using VOR, but I didn;t know where the VORs were. The instructor made sure that we were somewhere we hadn't been before, then had me remove the hood, figure out where we were by dead reckoning, and take us home.
I was completely lost at first -- nothing but farms and roads. But look around, study your maps. Ah ha, there are hills to the west, a highway running N-S, and the highway crosses a small river just south of a town that has a water tower next to a radio tower. Bingo.
GPS is great, as are all the other navigational aids. But knowing what to do when they fail is important, and knowing where you are and how to get where you want to go with just a map and compass gives you the warm and fuzzies.
Now, where did I leave my sextant?
For private pilots
;)
;) They cost a bit to join (you don't need to join to get the magazine) but you get a bunch of discounts, swag, etc, so it is worth it in the long run.
USA - 2-5k, although it can be a bit less.
Depends how good you are at it, the cost of the aircraft rental, instructor rental, etc.
You need 40 hours of flight time (30 with instructor, 10 solo IIRC) in order to get the license, although you might want to take more.
Plan on spending at least $20 an hour for the instructor and $35 for the aircraft. At least.
Canada, 7-15k-ish. I think $40/$35 an hour and up and 40 hours are required.
None of the above includes ground school - you usually have to pay for that seperately, although some flight schools will bundle it into a package.
Depending on the school, most will let you double up with another student in the cockpit and split the instructor cost. Or you can find a flying club that owns a plane and joins in. Or if you have a rich uncle with a plane and mechanic, that works well too
Oh, last I checked AOPA was sending out a free half year sub to its Flight Training magazine. 1-800-USA-AOPA and beg
In the USA, if you're under 21, consider the Civil Air Patrol. They will help you with paying for the classes - or pay for everything in some cases - and the "volunteer work" (i.e. stuff you do so they send you to flight school) looks great on resumes, college aps etc. Free / cheap to join - like $25 a year for CAP and you might need to buy a uniform and some small stuff.
They will often help with ground school and while they are supposed to teach stuff, the squadron I went to was pretty disorganized and nothing really got taught. Still, show up and they would send you to a camp for the summer (parachuting, gliding, flying, survival, etc)
In Canada, you have the Royal Canadian Air Cadets, which is also free / cheap and they basically give you a uniform for free (you're supposed to return it, but lets just say a certain cold weather gortex parka got lost along the way).
The squadron where I was at also paid for the medical certificate (including exam, which cost $200 or so, a nice chunk of change in a country where a doctors visit is free) and the ground school was taught free of charge by good instructors. Books, etc, also free. If you wanted to get a full scholarship, you had to apply, pass the written exam with good marks (not terribly difficult, mind you) and write an essay or something. Usually one or two cadets went every summer to get their pilots license and one or two went to get their glider license. Other camps included survival, leadership, music, physical fitness (for the nutcases who like to run 4 marathons in a row), first aid, etc, etc.
Don't let the military structure, etc, scare you - it isn't like you'll be press ganged into shooting drug runners or anything like that, CAP does a lot of SAR work and RCAC doesn't really do anything except for going flying / gliding / camping / shooting every month or so in addition to the weekly meetings.
I don't think CAP Cadets can take part in SAR missions directly, but can do a ride-along type thing, which is pretty cool as well.
But do realize that if you're going to get them to pay for everything, you're going to have to suck up a fair bit in order to get a scholarship and do a lot of grunt work like showing up for meetings, controlling parking at airshows and basically showing them that you're disciplined and dedicated.
From my experience, the RCAC had a lot more attendance, etc, than CAP - the RCAC squadron had usually around 100 people show up weekly (I think 140 was the most one day) when I was in this small town up in Canada (200,000 people), while the CAP squadron that covered all of Portland, Oregon and the surrounding cities (3.5 million+) had maybe 25 on a good day, including all the people whose job it was to be there.
That might of have had something to do with the way it was run, but a small squadron can work to your advant
1q2w3e4r5t6y7u8i9o0pqawsedrftgthyjukilo;p'azsxdcf