Slashdot Mirror
How Google's Autonomous Vehicles Work
An anonymous reader writes "IEEE reports that Google's autonomous cars have logged more than 190,000 miles driving in all kinds of traffic, and the company is also testing a fleet of self-driving golf carts on its campus. In a recent talk, Sebastian Thrun and Chris Urmson of Google gave details of the project and showed videos of the robot cars driving themselves and even doing some stunts. The goal is that the technology will help reduce congestion, fuel waste, and accidents."
We desperately need this tech. Aging populations drive less well after some kinds of age-related degradations, where automatic driving mechanisms could maintain mobility for affected people. Cars could also take kids to school, and park themselves in the parking lot, making the kids much less vulnerable to child molesters and bullies, as well as freeing Mom and Dad from transportation duties.
For everyone else, automation in driving will do a better job than people that are blabbing on cell phones, eating, drinking, fiddling with the radio, changing CD's, etc. etc. If drivers don't normally do these things, then automatic driving will enable them to do so, as well as a lot of other things that might be productive, like office work or simply enjoyable things like reading the newspaper or surfing the internet.
I'm all for autonomous vehicles but I think before it goes forward (ha ha) shouldn't there be some sort of legal framework in place? I mean the first fatal accident that can be even possibly attributed to an autonomous vehicle could very well kill the industry for a while at least. (I'm reminded of how the fatal crash of one of Buckminster Fuller's super efficient teardrop shaped 3 wheeled vehicles killed that concept).
How about for all "certified" (through rigorous federal testing) vehicles, there be "no-fault" collision insurance (or limits on damages). Unfortunately I'm neither a transportation expert nor lawyer so I'm just guessing.
You mean the "Speed up and turn on the windscreen wipers" decision? :D
My sister opened a computer store in Hawaii. She sells C shells by the seashore.
Like:
Can it, like I do, notice that the baseball rolling down a driveway may be followed by a child who is currently invisible behind a parked SUV?
Can it, like I do, notice that the driver *behind* me is distracted by her cell phone, has started late at the last three lights, so I should give myself more than average room between me and the car in front of me, so in case it stops suddenly, SHE won't have to stop as suddenly and will be less likely to rear-end me?
Can it, like I do, notice that even though the road has been clear of ice and snow, the next curve up ahead is deeply shaded and is likely to be slick?
Can it, like I do, notice that the baby deer is one one side of the road and his mother on the other, and even though he isn't charging across, it looks like he's about to do so, so I better slow down? Because this happened to me not one week ago, and it DID charge across.
Can it react to the highly dynamic and unpredictable world in ways that require human intelligence? And before you claim it's going to be a better driver than I am, note that I got my license in 1973, and have YET to have any sort of accident. Because I can provide human intelligence to the task of driving, and unless you want to claim the machine is as smart as a person, I am not so sure I believe I want to leave these choices up to it. I'll gladly fly on a computer flown aircraft, because that's a very, very different kind of task. No children appear in the sky from behind parked vehicles, and the pilot stands by the entire time to take over if something doesn't go right. Car drivers won't be that alert: they'll just doze off or play with their phones and not even glance out the window the whole time.
If automatic drivers initially reduce congestion (due to smoother driving) that effect will soon be diminished by increasing amounts of traffic.
Judging by my friends, most of them seem to prefer driving their own car if it will take them no more than 1.5 times as long as using the subway. If they had automatic drivers, they would probably accept even longer delays, so they might take their car instead of the subway even in rush hour when it would take more than 3 times as long.
This will also reduce incentives to move closer to your place of work or taking other measures to reduce your dependence on cars.
Don't get me wrong. I'm all for this technology. But make sure to combine with substantially increased gasoline taxes or you will be spending a large part of your life trapped inside a car.
compared to the logistics of driving, those systems are simplistic, only requiring a few inputs of simplistic data sampled at low rates, and one or maybe two outcomes based on that data (deploy airbag/antilock brakes etc).. even the cars that have auto-max braking concern me because they may anticipate when I most definitely do not want it to. computers are faster than humans, no question, but they are not as aware. hell even the electronic throttles in modern cars overanticipate, often getting the wrong idea..they also lag. I had the 'pleasure' of driving a 2011 with electronic everything, including throttle. it's a crapshoot whether the car will do nothing for a second, or rocket out from under me after a light change. in situations like this, consistency is safer than some lame heuristic programmed into a microcontroller somewhere. Keep it simple stupid is key here. stupid simple doesn't break or create unexpected behavior either.
then there's the messy issues of politicians and businessmen inserting themselves into the programming once the machine is placed between you and your car. until you can guarantee a society that will respect my freedoms and liberty, stay away from my steering wheel and shifter.
the impressive math to prove that airline software operates correctly, obviating the familiar testing found in most software development efforts, is also not popularly recognized.
Is it a rule, that there's an exception to every rule?
Seriously. And it's more than enough reason to get it legal on the streets.
Just have a way to lock out the manual override until you can pass the built in breathalyzer. It can transmit encrypted times and duration of computer control to local police vehicles in addition to the fact that you locked yourself out so they won't bother pulling you over.
Of course you will probably have to pre train it to park. Or have a parking assist mode that you can engage even while intoxicated that won't let you get into an accident. And of course a button to pull over just in case you need to throw up.
No more drunk driving or risking the fine and you have your car to get to work in the morning. And no $50 one way taxi ride or having trouble getting a taxi at last call. Besides most people would probably prefer to nap on their way home anyway.
This Google experiment proves that you can build a self driving car who can drive safely for thousands of miles on actual public roads. Yes, there are some additional conditions which are not practical (like the driving the path manually before to record a precise map), but it has been done.
Most drivers are not very much 'observing', I just read somewhere that the Blackberry outage caused vehicle accidents to drop by 20-40% in some Gulf states. Observing drivers, Ha !
Just compare how many accidents are caused by trees falling onto the street compared to texting.
The biggest problems to solve are the cultural, legal and liability issues.
Markus
Your car very likely does not cost $0.10 to operate. $60 for 600 miles is 35 mpg, which is certainly doable.. but that doesn't consider any maintenance, insurance, etc.
It seems unnecessary to buy a 2nd RT ticket if you're purchasing a transfer (the whole point of a transfer is to avoid charging a double fare).. but since you neglected to identify the system you were trying to use, your post is kind of worthless.
the kid/granny in front of you are not at fault. if you hit them you are at fault.
someone who is behind you *must* be at a safe distance to stop even if you perform emergency braking. if they hit you, they are at fault.
fill the rest yourself. in short you are just plain wrong. and dangerous.
Looking for people to chat about multicopters, coding, music. skype: gtsiros
It's actually very good idea. You may send a package to your wife in automatic car instead of driving there.
I'm sure there's probably some law against sending explosive devices in automatic vehicles. :)
On the other hand, I'm sure terrorists would love the idea...well, except for missing out on the 72 virgins.
Strat
Progressivism (aka US 'Liberalism'): Ideas so good they need a police/surveillance-state to enforce.
Part of how their system works is to have a fairly extensive database of what the local roads look like. Not just GPS and a roadmap, mind you, but actual 3D laser scans of a particular area taken from the ground. The first database was generated with humans behind the wheel of these cars; the autonomous vehicles used that information later for their own driving. That may seem like a big investment - building a detailed 3D map of an area just for a handful of vehicles. But like the web-crawler database that underpins Google's search engine, the same dataset, well distributed, can be used by near infinite clients. And once you have some number of these vehicles out there, they are continuously updating and adding to that database. Think of it like google street view, but where the streets are visible from any car-accessible vantage, and you can get synthesized live views from nearly any vantage. The images of buildings are no longer just images, but rather complete 3D models.
Google's stated motives are vaguely altruistic: reduce traffic and fuel consumption, make the roadways safer and more efficient. That's great and all, but hardly generates direct revenue for Google. But I think that having a very detailed, 3D, realtime map of most of the roads of the US would be a very valuable asset. How you monetize it is, right now anyway, somewhat speculative. On the other hand, I think Google, with tens of billions of dollar cash-on-hand, can sink a lot into this just on speculation without risk or immediate justification. There was a time not too long ago that companies would do that kind of thing: spend money on speculative research without immediate plans for payout. AT&T Bell Labs panned out as a pretty good benefit for the world.
There are, of course, a whole lot of 1984-esque aspects to this project that need to be addressed, too, so I wouldn't say this project is all upside.
right now I put my life in the hands of the other people on the street. the ones already playing with their phones or yelling at their kids in the back (commute is past 3 schools, 1 a HS)
If the drivers aren't watching the road now, at least put in a computer to watch for them. Hell, just have it apply the brakes if something gets in front of the car and it'd be an improvement
I had the 'pleasure' of driving a 2011 with electronic everything, including throttle. it's a crapshoot whether the car will do nothing for a second, or rocket out from under me after a light change.
The behavior you're complaining about is probably caused by the transmission rather than the drive-by-wire components. I've driven many automatics of varying vintages, all of which exhibit that behavior. My manual with electronic throttle, OTOH, always behaves exactly as I expect it to. This actually confirms your base point, though - with a manual transmission, I'm able to select the gear I want for the conditions, rather than having the computer guess for me.
No, congestion is caused by having too many cars for the the amount of road. If you are on an 8 lane highway, a there is some dumbass swerving back and forth across the lanes, and you and him are the only people on the road, there will be no congestion. Conversely, if you are driving on a two lane highway that is at full capacity. Adding hundreds of extra cars will cause congestion even if everyone driving perfectly.
Yes, people doing dumb things can be the final straw, but the idea that roads can take an infinate amount of traffic as long as people don't do dumb things is a myth.
Congestion is not caused by too many cars on the road, it is caused by what any individual car is doing. Look at the Indy wreck at Las Vegas. While the cars were all traveling around in the pack, things were tight, but not congested in the traffic sense of slowed traffic. It wasn't until one car did something that caused another car to shift that problems tragically began.
Traffic follows fluid dynamics, just like water through a pipe. Yes theoretically a highway or a pipe can only suppor so much, but for highways we don't reach that. For instance, a 100 mile straight highway with no exits can support how many cars? Assuming that there are no unexpected events, the cars can be quite close together and still move like a block.
However, in the real world is where we drive and theory falls apart. It is precisely the person cutting across lanes that is going to cause congestion as the lead cars will need to brake and the cars behind them and the cars behind them, etc., etc. That is why, unless most cars are the google cars, there will still be congestion. The google car can't eliminate that, it can only adjust for it.
How will this affect the automotive insurance industry? If I have a car that does all of my driving then I should never have an at-fault collision. So it only makes sense to me that my rates should be a fraction of what they are now. If the insurance industry sees this as a potential threat to their profits then I think they will lobby hard to keep it from advancing.
They mentioned it in this article. Watch the movie. The car always works off the laser 3d mapper, the GPS just assists as GPS isn't even accurate enough to drive by.
APK likes to ask for responses to the same things over and over. Maybe he just likes the responses?
Sorry, no.
Let's assume your car is older and paid for, like mine, so we'll ignore car payments, as that greatly increases your cost per mile.
Even with the car itself effectively free, you have to pay for 1) tires, 2) maintenance, 3) repairs, 4) insurance, and of course 5) fuel. Tires is really part of maintenance, but it's such a huge item that I gave it its own category. For most decent cars, a set of tires probably costs around $300; for 20,000 miles of service, that comes out to 1.5 cents/mile. But it goes up from there; if your car has lower-profile, larger tires, or you drive a truck or SUV, a set will probably cost more like $5-600. If you drive a BMW, or anything with run-flat tires, expect to pay quite a bit more, probably $1200/set. So you're already looking at another 3 cents/mile or so just for tire wear.
Then you need to account for maintenance costs: changing the oil, belts, spark plugs, etc. These things are generally mileage-based, so if you drive your car 4x as much, you'll have to do them 4x as often.
Then you need to account for repair costs. Eventually, stuff wears out or breaks. If your car is older and paid off like mine, this is a much bigger factor. If your car is new and in-warranty, it's not yet a factor, but then you have to account for hefty car payments (and the fact that your warranty is mileage-limited).
If you're like me and can do your own repairs and maintenance, this cost isn't so bad, but even so expect to pay $100 or more every time something breaks, just for parts. If you're not like me and you have to go to a mechanic, expect to pay double for the parts, and another $70/hour for labor, so things like changing the timing belt (regular maintenance on many cars) come out to $500-750, which is a large chunk of money to spend at one time for most families. What if your starter craps out like mine recently did? While I got a remanufactured one on Ebay for $70 and drive our other car while waiting for it to arrive, that's probably a $250-300 service fee for normal people who need their car repaired right away to get to work. Again, a big surprise expense for someone on a budget. Of course, if you do do your own work, then you have to account for your time too; you could be spending that time doing something else besides fixing your car. (But taking it to a mechanic doesn't eliminate the time factor either, unless your mechanic comes to your house in the evenings to fix your car; sitting at the auto service center waiting for your car to be repaired is wasted time just as if you had done the work yourself.) Driving your car less will reduce the amount of time you spend fixing your car or getting it fixed.
Finally, you totally ignored insurance costs. Insurance cost varies greatly depending on how old your car is (how much it costs to replace), and also how much you drive. If you drive 30,000 miles per year, you'll pay a lot more than if you only drive 5,000 miles per year. You could always lie and say you only drive 5k when you really drive 50k, but if the insurance company finds out they can deny your claims and maybe prosecute you for insurance fraud. So whatever you pay for insurance in a year, you can subtract whatever it'd be for "recreational use only" (ask your agent for a quote), then divide that by your average number of miles driven per year, and find your insurance cost per mile.