Slashdot Mirror
Oxford Tests Self-Driving Cars
halls-of-valhalla writes "Using advances in 3D laser mapping technology, Oxford University has developed a car that is able to drive itself along familiar routes. This new self-driving automobile uses lasers and small cameras to memorize everyday trips such as the morning commute. This car is not dependant on GPS because this car is able to tell where it is by recognizing its surroundings. The intent is for this car to be capable of taking over the drive when on routes that it has traveled before. While being driven, the car is capable of developing a 3D model of its environment and learning routes. When driving a particular journey a second time, an iPad on the dashboard informs the driver that it is capable of taking over and finishing the drive. The driver can then touch the screen and the car shifts to 'auto drive' mode. The driver can reclaim control of the car at any time by simply tapping the brakes."
on your car, you have "brakes". if the brakes break, then you have big problems.
Kindly consult the Oxford English Dictionary.
there are 3 kinds of people:
* those who can count
* those who can't
For a prototype it's OK but I wouldn't bet my life on anything running on consumer-grade devices.
... a shadowy flight into the dangerous world of a man who does not exist...
"breaks" for vehicle brakes keeps coming up here. Is this an Americanism?
But this is a story about Oxford FFS, the cultural heart of the English language, UK version.
http://en.wikipedia.org/wiki/Oxford_English_Dictionary
Agreed. Once one company has done it all others should stop.
BTW, Good thing Opera replaced its own engine with WebKit too.
none
Jeeze, I have enough problems with cruise-control being enabled when it should be disabled, or vice versa. Now it will steer for me too? No, wait, it's not! Gaaahhhh! "Sorry!" "Asssss-hhhooollllleeee...."
Except for serving the public good.
Don't have one to check.
What if I need to accelerate to avoid a hazard? How does "braking", which I assume the poster meant but does not understand, help me regain control?
I've heard that "cruise control" systems allow temporary acceleration, then fall back to the desired speed, while braking puts them into a sort of standby mode which can be resumed. Maybe they should consider something similar.
You'd think it would be obvious to the folks at Oxford: if you're building 3D maps, and storage is getting relatively cheap, why not just build 3D maps of whole regions so the car knows its way around? Then the human can pick any route, rather than having to teach the computer.
at least in my opinion.
And we are all so much better off for having read your very worthwhile, informative, and thoughtful opinion. Thank you for taking the time to share such a well-argued gem with us.
So an iPad is telling me it knows the way and is able to guide me through Oxford traffic safeley? Erm, not thank you.
I like how tapping the brakes gives control back, it's like complete cruise control. Wonder what happens when it snows 3 ft (guess what part of the country i live in) or a different situation where the landscape changes. Even after the roads are clear there are still walls of snow that would block any sensors abilities. Maybe we need something up on the phone poles or in the center lane embedded that tells cars where they should go. Seems more reliable that way, but I guess that would be more public in nature than the scope of this project. But anyway you look at it, driving is way too important a responsibility to let any adult that wants to/can afford to do it whenever they want, in whatever condition they happen to be in. Hope this project bears fruit.
Once one company has done it all others should stop.
Well, not necessarily. There could be lot of interesting (read creative) ideas one could have missed. And indirectly, it creates a healthy compitition everybody benefits from.
If that was not the case, we would have had only one type of car, only one type of plane, only one type of phone..and the list may go on.
hilarious
So, what happens if, say, a building gets demolished, or a set of trees are cut, or it snows? If the software is looking for specific topography, which it learned from previous trips, isn't it as likely as humans to get lost when things change?
According to a TV report I saw on this, the point of the Oxford technology is that it's supposed to be much simpler and cheaper than existing implementations, with the development version costing only £5000, and projected price of a commercial version of just £100.
It would be nice if such the article mentioned the existence of comparable tech, such as Google's self-driving cars, and perhaps did some comparisons, but unfortunately being a science and technology journalist these days means copying and pasting press releases, so the journo in question probably actually does have such little interest in technology that he hasn't head of the Google initiative. Sad.
Oh no... it's the future.
can it change lanes? can it route around road blocks? can it stop for red lights with out getting messed up?
P.S. according to this article, Google's self-driving car is probably considerably more expensive at around $250000 (£160000), although G don't release figures.
Oh no... it's the future.
It's news because it's a different approach.
Theoretically this mixed with a service that has surveyed the entire road network could be amazing. It would remove the need to have travelled the route previously.
Although I do wonder how it copes with changes in road layout/diversions etc.
"I see you are driving down Laurel Ave and I can drive you to your destination. Are you heading to:
- Bosco's Liquor Store (1.73 mi)?
- The Bouncing Pasty Gentleman's Parlor (2.64 mi)?
- The Purple Nurple Tobacco Accessory Shop (1.25 mi)?"
".... Siri, change profile to 'Mom'."
"Okay. Changing user profile settings to 'Mom', please wait"
"Not all who wander are lost" -- JRR Tolkien
All that time and engineering effort. All that programming expertise. And yet.... it still drives on the wrong side of the road.
"He took a duck in the face at 250 knots." -- William Gibson, Pattern Recognition
To 1 and 3: almost certainly. To 2: probably not unless it has already "learned" the alternative route.On the other hand, one difference between computers and humans is that you can copy the "learning" from one computer in a way you cannot copy from one brain to another. So it would not strike me as unreasonable for a net-connected car to download the images of a detour route within a few seconds of recognising a roadblock.
Consciousness is an illusion caused by an excess of self consciousness.
As Kruschev told Nixon, it is a waste to have two teams of designers for washing machines.
Consciousness is an illusion caused by an excess of self consciousness.
Where is the "break" pedal?
*whoosh*
which is totally what she said
Yours is broken, otherwise it would read like this working one
It can probably change lanes and handle red lights, otherwise it wouldn't be very usable.
As for road blocks it's no longer "routes that it has traveled before" and one would have to assume that it needs to give back control to the driver for him/her to drive around the road block back to familiar territory.
Yeah, but this is being developed in Oxford, and whilst we have things that look like roads around here, they're not actually designed to be used by cars. All a self driving car around here needs to do is say: "I'm sorry, I can't do that Dave. Have you considered using your bicycle? Be sure to avoid being hit by a bus!".
Who writes these summaries?
Hey, I've been down this road 200 times, and you've went three different ways! Want me to drive the rest of the way? All I did was RTFS, but that's a problem with bad summaries...unless that's what the article implies. How does it know which of the three ways you want to go? That sort of thing could kill this innovative idea if it's indeed the case.
I have been a captive in America my entire life. Everybody and everything uses customary units instead of metric.
Google's self-driving car is probably considerably more expensive at around $250000
That is the cost of an engineering prototype. The cost of massed produced cars would be far lower. I talked to a Google engineer that was demoing one of their cars at the San Jose Fairgrounds. He pointed out a bulky optical rotary encoder on each wheel, about the size of a soda can, and said they cost over $2000 each. He said they were going to soon replace them with a magnetic hall-effect encoder the size of a penny. Cost: $3 each.
It's news because it's a different approach.
Except that it's not. Other self-driving cars, including Google's, already do route learning and mapping. The difference is that Google integrates that with stored maps, and and lots of other sensory input, including GPS. So these Oxford researchers are not doing something new, they are just doing less. In safety critical systems, removing redundancy is usually not a good idea.
One way to make their system more useful would be to upload learned routes to a server, so they can be auto-downloaded to other vehicles. Then your car could self-drive even on roads you haven't driven on before, as long as someone else has driven them.
While replacing their expensive encorder certainly helps, Google has a long way to go to bring down their pricing. In particular, the LIDAR unit on the top is probably dominating the price. The model in question costs around $75,000 and as far as I can tell, Google isn't getting rid of it anytime soon.
Of note: I expect that the LIDAR unit in the Oxford car is also dominating the price, and expected price decrease in the future would be achieved by going camera-only.
100 quid? Sounds a bit on the low side: we're talking several camera's, lasers + sensors, a control unit, and actuators for pedals & steering. Even £1000 seems too low... compare this to what you can expect to pay for factory installed cruise control.
If construction was anything like programming, an incorrectly fitted lock would bring down the entire building...
...till the first foggy day.
Rome taught me patience and assiduous application to detail. Virtues which temper the boldness of great, general views.
the LIDAR unit on the top is probably dominating the price. The model in question costs around $75,000
How many LIDAR units are sold every year? Maybe a few thousand? 60 million cars are manufactured each year. That kind of volume can lead to huge price decreases.
expected price decrease in the future would be achieved by going camera-only.
Cameras don't deal well with rain, snow, and fog.
No, it is not misspelled: initially they put the iPad on the hood
Hood? That's the leaky fabric bit you get on top of convertibles. Perhaps you mean "bonnet"?
Silly Americans naming the metal flap that covers a car engine after a type of headwear... oh, wait... :-)
Meanwhile, if they're going to test these things in Oxford I hope that they're fitting the car with an industrial strength bike-catcher and an AI that can cope with one-way systems designed by M.C. Escher.
In a survey of 100 programmers, 111111 thought that duck-typing was a good idea.
So these Oxford researchers are not doing something new, they are just doing less.
Doing less is a new approach. A sensible one, particularly in robotics. For example see the Roomba, vs the Electolux Trilobyte. The Trilobite mapped the whole room before designing an efficient cleaning route. The Roomba just wanders randomly, with some simple heuristics for occasionally following walls and occasionally changing direction. Result: The cheap Roomba approach is successful in the market, and the expensive Trilobite is a failure.
Here for example you mention GPS. That's of limited use, as the accuracy is in terms of meters. Far too course for self driving. And it can disappear completely in cities. And all it would do is narrow down the initial search space to identify the current location.
One way to make their system more useful would be to upload learned routes to a server, so they can be auto-downloaded to other vehicles. Then your car could self-drive even on roads you haven't driven on before, as long as someone else has driven them.
I suggest you RTFA, then you won't spend time describing something they already have slated for the future.
"This new self-driving automobile uses lasers..."
expected price decrease in the future would be achieved by going camera-only.
Cameras don't deal well with rain, snow, and fog.
Infra-red cameras cope fairly well, better than the human eye sometimes.
No colour or religion ever stopped the bullet from a gun
Cameras don't deal well with rain, snow, and fog.
Neither do your eyes, as they really aren't anything more than cameras.
Cameras don't deal well with rain, snow, and fog.
Neither do your eyes, as they really aren't anything more than cameras.
Yes, but the goal of self-driving cars is to improve on human drivers, not just replace them.
It's going to have to be more than 100 pounds, but you're not going to need the actuators by the time we really have self-driving cars, because the majority will have electric power steering, and modern ABS can already activate without the driver pressing the brake, which is used for TC and ESP.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
don't drive past The Mirror Store.
improve on human drivers, not just replace them
Replacing them would be a huge improvement.
Apple Corp. (AAPL) announced that with its wide array of personal electronics, it is now testing self-driving people. Basically it's people walking around cities on autopilot. The test is to see if other people can avoid running them over or colliding with them on sidewalks, etc., while they're busy playing with their toys instead of looking where they're going.
So far, mixed results.
No. Just making it non human is already an advantage in some respects. I would gladly replace myself with a self-driving car if it was as good as myself. I might even be willing to pay double for a car with that feature. I mean leather seats, climate control, wood paneling interior, crazy powerful engines, are features I wouldn't pay an extra dime for. Luxury for luxury's sake is stupid, imho. As is speed for speed's sake. But give me a car that drives itself, I 'll buy one right now for twice the price that my existing car is.
Well.. maybe. Or Maybe not. But Definitely not sort of.
the google cars have been.
Cameras don't deal well with rain, snow, and fog.
Neither do your eyes, as they really aren't anything more than cameras.
A lot of that is fixed by the information processing equipment attached to them, but that's something you really can't expect to emulate in silicon in near future - the lidar would be cheaper than that.
Ezekiel 23:20
Not really, if there's something in the optical path...
So it would not strike me as unreasonable for a net-connected car to download the images of a detour route within a few seconds of recognising a roadblock.
Especially if the workers have a small beacon that broadcasts at, say, 5 GHz the map of the area that shows how to drive around the work area. All cars in vicinity receive that and can act upon these instructions if they are signed and the chain of trust is good enough.
That would be better than what we have now - a mass of cars trying to get by the work area, and nobody can reliably know what path is safe until you see the cones and have to change the lane. If there is a sign "<< RIGHT LANE CLOSED <<" it still doesn't tell you how many lanes are closed, and where are the transition zones, and whether there is a worker with a STOP sign. The broadcast could easily contain all that, tied to the GPS and to local markers.
That's of limited use, as the accuracy is in terms of meters. Far too course for self driving.
You're in the wrong thread. The thread on homophones and illiteratism is just above this one.
Cameras don't deal well with rain, snow, and fog.
LIDAR doesn't deal well with other LIDAR units in the same area.
sub f{($f)=@_;print"$f(q{$f});";}f(q{sub f{($f)=@_;print"$f(q{$f});";}f});
Obviously nothing but an iPad would do as a user interface. Considering the tech they've come up with, you'd think that was the trivial bit and they could have saved quite a lot of dosh by creating their own interface. Or are they trying to be trendy (sorry "nextified")?
Would you like some beef with your horse-burger?
Birds are not dinosaur descendants;birds are dinosaurs, for all useful meanings of "birds", "are" and "dinosaurs"
The new Oxford automatic-driving car should be called the "Comma".
That would be awesome.