6 ms·

How Many Miles of Driving to Demonstrate Autonomous Vehicle Reliability? (2016)

21 points by quickfox 8y ago

mLuby 8y ago
>Self-driving vehicles must be driven for hundreds of millions of miles to demonstrate reliability, which would take 10-100 years. So we can A) accept higher fatalities for convenience, B) demand much better performance from self-driving vehicles than we expect of humans, or C) wait.
- dwighttk 8y ago
  What’s the difference between B and C?
  - manicdee 8y ago
    The difference is in how you measure "better than humans." Which also means looking closely at how we measure how "good" a human driver is. It's the same argument as "what is AI" which inevitably devolves to "must be demonstrably smarter than humans in every way before we consider it to actually be intelligent."
    - gehwartzen 8y ago
      I also suspect that the average human driver is getting worse not better. We have way more distractions available in the cabin than in the past mostly due to cell phones. I also notice a lot of younger people are either forgoing driving altogether or getting their licenses much later due to the availability of uber/lift as an alternative so would imagine they have less experience.
- thx11389793 8y ago
  If we have a hundred years to burn, how about reorganizing our cities into forms that obviate the need for cars (self-driving or not) in the first place? Anecdotally, I've heard people make the point that public transit is infeasible in the states because the population density is too low, and people are too spread out. I'm not sure that it is -- the US Census Bureau stats show the share of Americans living in urban areas growing steadily throughout the history of the country, with 80% of the population currently hailing from urban areas. We are and have been urbanizing since the birth of the country, I reckon if were smart about building infra now, we can remove the problem SDVs are trying to solve in approximately as long as it would take for them to become ubiquitous were we to take no action.
  - chiefofgxbxl 8y ago
    Agreed. If car crashes kill so many in the U.S., wouldn't it make sense to make the country a less car-dependent nation if we want to reduce fatalities? Walking and biking generally won't get people killed, so why not promote that? I think part of the problem is that self-driving vehicles is marketed as a life-saving technology, when it is the novelty and futurism people are attracted to. (It is indeed an impressive and challenging problem.) The furthest mainstream press release I can recall making the case for autonomous vehicles was Google's efforts. This makes me wonder if they had a solution in search of a problem, so they took road deaths in the U.S to sell the idea. Maybe they were chasing the next iteration of intense computing application, e.g. Deep Blue -> Watson -> self-driving cars. The 40k+ fatalities per year is a solvable problem without autonomous vehicles: (1) promote equity in other methods of transportation, (2) enhance driver-assistance technology, (3) design roads to be built for safety rather than convenience/speed [0] Regarding point 2 above, note that Chevrolet has "teen-driver technology" which can set a maximum speed, and provide visual and audible warnings if the (teen) driver exceeds the speed limit [1]. Presumably the vehicle has cameras to read speed limit signs. Why isn't this being applied to all drivers? It seems to be the classic case of "let's blame teens for everyone's bad driving" (juvenoia). [0] https://www.strongtowns.org/journal/2018/1/25/speed-kills-so-why-do-we-keep-designing-for-it https://www.strongtowns.org/journal/2018/1/25/speed-kills-so... [1] https://www.chevrolet.com/teen-driver-technology https://www.chevrolet.com/teen-driver-technology
- Semirhage 8y ago
  Right now I’d be thrilled with doing just as well as humans without fudging the numbers. We’re orders of magnitude away from that, so frankly this is all a non-issue. We’re far from demonstrating reliability because they’re far from reliable. Maybe someday they will be, maybe not, but for now we’re playing a dangerous game in the name of a dubious utopia. The most likely outcome is a winter. Edit: I don’t mind the downvotes, but I wish they came with a rebuttal rather than radio silence.
  - dpcx 8y ago
    Pretty sure you're being downvoted by making the statement that we're orders of magnitude away, but not providing any data to back up your claim. Even with the number of accidents / injuries / deaths caused by self-driving cars, without actual data, they at least seem to being doing as well as humans - though that's completely anecdotal from my side.
    - Semirhage 8y ago
      Fair enough. https://www.insurancejournal.com/news/west/2018/02/01/479262.htm https://www.insurancejournal.com/news/west/2018/02/01/479262... The latest disengagement report showed that Waymo vehicles, in tests conducted from December 2016 through November 2017, on average logged 5,596 miles without Waymo’s safety drivers disengaging the system and retaking the wheel. Its next closest contender was Cruise, owned by General Motors Co, at 1,214 miles on average between disengagements. And later Waymo, for example, drove 352,545 miles in the state during the period with only 63 disengagements. Cruise vehicles drove about a third less, at 127,516 miles, and had 105 disengagements. The third best performance came from Nissan Motor Co , which drove 5,007 miles and had 24 disengagements, meaning that its vehicles had disengagements on average every 208 miles. The numbers fall off sharply after Nissan, with Baidu Inc at an average rate of every 41 miles, chipmaker Nvidia Corp at 4.6 miles on average, and Mercedes, with disengagements every 1.3 miles on average. Some further discussion and context: https://www.scientificamerican.com/article/are-autonomous-cars-really-safer-than-human-drivers/ https://www.scientificamerican.com/article/are-autonomous-ca... Between mistakes a human wouldn’t make unless impaired, such as accelerating into a crash barrier or into a fire truck, and the high rate of human intervention, compared to the rate of fatal accidents per x miles humans have (1.27 per 100 million miles even including drunks, the elderly and new drivers) the argument for “safer than humans” seems utterly insupportable at this time.
stefan_ 8y ago
This is just bad math to justify the "imperative to develop adaptive regulations" aka "lets just let Uber drive without emergency braking". No one is fooled. It didn't take a billion miles to expose Ubers system as broken, though it did unfortunately cost someone their life. It is also not needed to take a graduate class of statistics to look at Waymos tests and realize none of them involve driving rain in pitch black night to come to the conclusion that maybe they aren't giving the entire picture and that "driving miles" are not created equal.
ck2 8y ago
it's a perfect chain of self-destruction public thinks technology and computers are "magic" and perfect self-driving cars are somehow easily possible, not having a clue about the extensive limitations of radar and other sensors vs software managers think they'll just fix the code by pushing a new update later, so just get it out the door because someone higher up is breathing down their neck government forces their agencies to work with minimal budgets so testing is too basic, never updated and never re-reviewed, maybe even corrupted by lobbyists to do self-regulation without oversight victims often won't be the drivers who early-adopted or abuse the technology out of ignorance, laziness or greed but instead innocent pedestrians, cyclists, or other drives in regular vehicles doing nothing wrong when the victims sue, juries will be made to easily believe technology can't be wrong and it was human error by either the victim or the driver remember toyota spaghetti code for electronic fly-by-wire steering and brakes? sadly I think it's going to take some lawyers with billion dollar class-action suits to get all this to slow down beta testing 2-10 ton vehicles moving fast in the realworld with plenty of possible victims demands space-grade-level testing (and even then plenty of rockets blow up on the launch pad or never make it to orbit) meanwhile I hope independents like consumer-reports come up with an ever-improving test site of obstacles and events for these cars to deal with and that they are very thorough and harsh about the tests every year
- brokenmachine 8y ago
  > sadly I think it's going to take some lawyers with billion dollar class-action suits to get all this to slow down That will come in short order. Uber was lucky that the person they killed was homeless and not a politician's son or daughter.
vladislav 8y ago
How do we test human readiness for driving? We give them a driving test, which takes all of 40 minutes. This is possible only because we know a lot about what it means to be a human. To test self-driving technology we need to understand a lot about the AI that is behind it, which is an active area of research with much progress. Moreover, there is adversarial testing in simulation and on raw sensor data, which can tell you a lot about your system before any large scale real world testing begins, including when it might be safe to start real-world testing. Real-world testing at such large scales is neither a practical nor a necessary component to validating self-driving systems.
- namibj 8y ago
  Take a look at Germany, and how to get a drivers license there. Hint: if you are bad at parallel parking, you will fail the exam.
- ux-app 8y ago
  >We give them a driving test, which takes all of 40 minutes 40 minutes + 16 years to prime the neural net for - object recognition and classification - developing a theory of mind so that the NN can anticipate how other NNs will behave - etc The NN can't sit the 40 minute test unless they prove they can do successful object recognition and avoidance at low speed (while in bipedal mode). That they have the sufficient fine/gross motor skills to navigate a complex and changing 3D environment etc. It's going to take a while before we've got digital systems that can compare.
rdm_blackhole 8y ago
We should not make "better" the enemy of perfect or something of the essence. If we can have self-driving vehicles that are twice as safe as the average human, then they should be allowed on the streets. https://www.rand.org/blog/articles/2017/11/why-waiting-for-perfect-autonomous-vehicles-may-cost-lives.html https://www.rand.org/blog/articles/2017/11/why-waiting-for-p... I realize that accepting that a machine may kill you when you cross the street is a bit unsettling but if you think about it, it is no different than accepting that the person in the car behind you may have a moment of absence or not pay attention to the road end up killing you or a pedestrian. Yet having another human kill another human seems to be accepted by society? If car manufacturing companies take on the responsibility and financial burden on them to compensate families that are killed/injured by a self-driving car, just like insurance companies do nowadays on the behalf of the drivers at fault, then the problem becomes non-existent unless I am missing something.
- rohit2412 8y ago
  I am appalled at statements such as "twice as safe as human". We should be comparing driverless technologies with humans assisted by the same technology running in passive mode. Otherwise, you're implyjng that the developer of that technology would either release the technology as completely driverless or not release it at all.
  - gsam 8y ago
    > Otherwise, you're implyjng that the developer of that technology would either release the technology as completely driverless or not release it at all. In some ways that statement is true. Ford and a number of other companies refuse to implement level 3 automation -- because of the switch-over costs and human-vehicle communication impedances which would almost certainly lead to more accidents.
    - Sophistifunk 8y ago
      Yup. If the computer works fine 99.9% of the time, there's just no way to keep the meat paying attention so it's there when that .1% situation pops up.
nmca 8y ago
This paper consistently assumes a fleet size of 100, and 24/7/365 driving. These are silly in different directions, but I think the former invalidates a lot of their claims. As a slightly not-random example, GM produces around 30,000 Chevy Bolts a year. Things start to look a little expensive, but the paper claims "This analysis shows that for fatalities it is not possible to test-drive autonomous vehicles to demonstrate their safety to any plausible standard". I think they are neglecting "without increasing fleet sizes beyond 100 vehicles".
- nmca 8y ago
  Ah, and indeed it seems like substantial (>10x) fleet size increases are planned. https://www.theverge.com/2018/1/30/16948356/waymo-google-fiat-chrysler-pacfica-minivan-self-driving https://www.theverge.com/2018/1/30/16948356/waymo-google-fia...
- notahacker 8y ago
  Their assumptions are just one way of reaching the 275 million mile mark needed to hit that 95% confidence interval. If we're looking at actual numbers, Waymo's recently boasted of hitting the 7 million mile mark over several years of testing so if anything, Rand's assumption of 275M miles in 12.5 years for a test programme looks to be erring on the generous side. Sure, if Chevrolet decided to use their entire production run of Bolts to test autonomous vehicle performance you'd get 275 million miles pretty quickly, but they might struggle to get consumer buy in and regulator approval for selling fully autonomous driving capability to 30,000 consumers unless and until they can produce evidence that their autonomous mode accident rate is lower than human drivers... And there's a much bigger problem, in that the software being tested is continually evolving, and modifications which result in improvement in general driving conditions don't necessarily tends towards stable (or monotonically reducing) numbers of rare accident-inducing bugs.
- amputect 8y ago
  FWIW the paper mentions that Google's autonomous driving fleet currently has 55 cars, so I suspect the 100 fleet size was picked to reflect a significant improvement over a notable competitor in the field. The article mentions Google's self-driving fleet size but I didn't notice other companies. I wonder what the fleet size is at other companies that are working on this, anyone know? The reason I ask is that I'm curious if "making more cars" is really the bottleneck. It's totally possible that the Waymo team doesn't have the headcount to keep up with the data output from the existing 55 units, or the fleet logistics bandwidth to keep up more than 50 or 60 heavily custom self-driving vehicles. It may very reasonably be a staffing issue rather than a money one, and it's possible that even if Google bought the rest of the self driving industry they might not be able to maintain more than a couple hundred cars. Disclaimer: I work for Google, but not on self driving cars. I actually know almost nothing about our self driving efforts, sorry.
8bitsrule 8y ago
First show me a machine that won't run into a barrier, or a human, or a parked fire truck, or a parked police car at 40-60mph without braking. Then ask this (for now ridiculously abstract) question again.
- IanCal 8y ago
  These things seem to happen with humans too. But what you're asking leads directly to the question being asked here! How long would something have to drive for before you believe me it's unlikely to do one of those things? Is it even possible for me to do this?
  - diggernet 8y ago
    It would have to drive fully autonomous in all conditions that humans drive in, for enough miles to produce a statistically valid lower rate of accidents than humans. Since it can't drive fully autonomous until it has proved itself, it would be allowed to have a safety observer onboard. But here is the catch: Any intervention by the observer counts as an accident, since we are simulating what it would do without a human. Assume worst case. Oh, and software updates reset the clock.
mgamache 8y ago
Totally unsupported by examples, but it seems like the huge economic benefits will cause society to underestimate to risk of autonomous vehicles. We'll debate and distort the relative risk and provide plausible deniability for the right people. In the end, we'll tweak the AI based on real world failures until it's good enough to win the PR battle. This is assuming it can be tweaked to work. The key will be protecting (or limiting) the liability of car manufactures from lawsuits. If AVs are involved in 37,000 fatalities a year (the 2016 human stats) the lawsuits will be untenable. Maybe there's a plan to handle this, and I just haven't seen it. Just my opinion...

21 points by quickfox 8y ago

>Self-driving vehicles must be driven for hundreds of millions of miles to demonstrate reliability, which would take 10-100 years. So we can A) accept higher fatalities for convenience, B) demand much better performance from self-driving vehicles than we expect of humans, or C) wait.

dwighttk 8y ago
What’s the difference between B and C?
- manicdee 8y ago
  The difference is in how you measure "better than humans." Which also means looking closely at how we measure how "good" a human driver is. It's the same argument as "what is AI" which inevitably devolves to "must be demonstrably smarter than humans in every way before we consider it to actually be intelligent."
  - gehwartzen 8y ago
    I also suspect that the average human driver is getting worse not better. We have way more distractions available in the cabin than in the past mostly due to cell phones. I also notice a lot of younger people are either forgoing driving altogether or getting their licenses much later due to the availability of uber/lift as an alternative so would imagine they have less experience.
thx11389793 8y ago
If we have a hundred years to burn, how about reorganizing our cities into forms that obviate the need for cars (self-driving or not) in the first place? Anecdotally, I've heard people make the point that public transit is infeasible in the states because the population density is too low, and people are too spread out. I'm not sure that it is -- the US Census Bureau stats show the share of Americans living in urban areas growing steadily throughout the history of the country, with 80% of the population currently hailing from urban areas. We are and have been urbanizing since the birth of the country, I reckon if were smart about building infra now, we can remove the problem SDVs are trying to solve in approximately as long as it would take for them to become ubiquitous were we to take no action.
- chiefofgxbxl 8y ago
  Agreed. If car crashes kill so many in the U.S., wouldn't it make sense to make the country a less car-dependent nation if we want to reduce fatalities? Walking and biking generally won't get people killed, so why not promote that? I think part of the problem is that self-driving vehicles is marketed as a life-saving technology, when it is the novelty and futurism people are attracted to. (It is indeed an impressive and challenging problem.) The furthest mainstream press release I can recall making the case for autonomous vehicles was Google's efforts. This makes me wonder if they had a solution in search of a problem, so they took road deaths in the U.S to sell the idea. Maybe they were chasing the next iteration of intense computing application, e.g. Deep Blue -> Watson -> self-driving cars. The 40k+ fatalities per year is a solvable problem without autonomous vehicles: (1) promote equity in other methods of transportation, (2) enhance driver-assistance technology, (3) design roads to be built for safety rather than convenience/speed [0] Regarding point 2 above, note that Chevrolet has "teen-driver technology" which can set a maximum speed, and provide visual and audible warnings if the (teen) driver exceeds the speed limit [1]. Presumably the vehicle has cameras to read speed limit signs. Why isn't this being applied to all drivers? It seems to be the classic case of "let's blame teens for everyone's bad driving" (juvenoia). [0] https://www.strongtowns.org/journal/2018/1/25/speed-kills-so-why-do-we-keep-designing-for-it https://www.strongtowns.org/journal/2018/1/25/speed-kills-so... [1] https://www.chevrolet.com/teen-driver-technology https://www.chevrolet.com/teen-driver-technology
Semirhage 8y ago
Right now I’d be thrilled with doing just as well as humans without fudging the numbers. We’re orders of magnitude away from that, so frankly this is all a non-issue. We’re far from demonstrating reliability because they’re far from reliable. Maybe someday they will be, maybe not, but for now we’re playing a dangerous game in the name of a dubious utopia. The most likely outcome is a winter. Edit: I don’t mind the downvotes, but I wish they came with a rebuttal rather than radio silence.
- dpcx 8y ago
  Pretty sure you're being downvoted by making the statement that we're orders of magnitude away, but not providing any data to back up your claim. Even with the number of accidents / injuries / deaths caused by self-driving cars, without actual data, they at least seem to being doing as well as humans - though that's completely anecdotal from my side.
  - Semirhage 8y ago
    Fair enough. https://www.insurancejournal.com/news/west/2018/02/01/479262.htm https://www.insurancejournal.com/news/west/2018/02/01/479262... The latest disengagement report showed that Waymo vehicles, in tests conducted from December 2016 through November 2017, on average logged 5,596 miles without Waymo’s safety drivers disengaging the system and retaking the wheel. Its next closest contender was Cruise, owned by General Motors Co, at 1,214 miles on average between disengagements. And later Waymo, for example, drove 352,545 miles in the state during the period with only 63 disengagements. Cruise vehicles drove about a third less, at 127,516 miles, and had 105 disengagements. The third best performance came from Nissan Motor Co , which drove 5,007 miles and had 24 disengagements, meaning that its vehicles had disengagements on average every 208 miles. The numbers fall off sharply after Nissan, with Baidu Inc at an average rate of every 41 miles, chipmaker Nvidia Corp at 4.6 miles on average, and Mercedes, with disengagements every 1.3 miles on average. Some further discussion and context: https://www.scientificamerican.com/article/are-autonomous-cars-really-safer-than-human-drivers/ https://www.scientificamerican.com/article/are-autonomous-ca... Between mistakes a human wouldn’t make unless impaired, such as accelerating into a crash barrier or into a fire truck, and the high rate of human intervention, compared to the rate of fatal accidents per x miles humans have (1.27 per 100 million miles even including drunks, the elderly and new drivers) the argument for “safer than humans” seems utterly insupportable at this time.

stefan_ 8y ago

This is just bad math to justify the "imperative to develop adaptive regulations" aka "lets just let Uber drive without emergency braking". No one is fooled. It didn't take a billion miles to expose Ubers system as broken, though it did unfortunately cost someone their life. It is also not needed to take a graduate class of statistics to look at Waymos tests and realize none of them involve driving rain in pitch black night to come to the conclusion that maybe they aren't giving the entire picture and that "driving miles" are not created equal.

ck2 8y ago

it's a perfect chain of self-destruction public thinks technology and computers are "magic" and perfect self-driving cars are somehow easily possible, not having a clue about the extensive limitations of radar and other sensors vs software managers think they'll just fix the code by pushing a new update later, so just get it out the door because someone higher up is breathing down their neck government forces their agencies to work with minimal budgets so testing is too basic, never updated and never re-reviewed, maybe even corrupted by lobbyists to do self-regulation without oversight victims often won't be the drivers who early-adopted or abuse the technology out of ignorance, laziness or greed but instead innocent pedestrians, cyclists, or other drives in regular vehicles doing nothing wrong when the victims sue, juries will be made to easily believe technology can't be wrong and it was human error by either the victim or the driver remember toyota spaghetti code for electronic fly-by-wire steering and brakes? sadly I think it's going to take some lawyers with billion dollar class-action suits to get all this to slow down beta testing 2-10 ton vehicles moving fast in the realworld with plenty of possible victims demands space-grade-level testing (and even then plenty of rockets blow up on the launch pad or never make it to orbit) meanwhile I hope independents like consumer-reports come up with an ever-improving test site of obstacles and events for these cars to deal with and that they are very thorough and harsh about the tests every year

brokenmachine 8y ago
> sadly I think it's going to take some lawyers with billion dollar class-action suits to get all this to slow down That will come in short order. Uber was lucky that the person they killed was homeless and not a politician's son or daughter.

vladislav 8y ago

How do we test human readiness for driving? We give them a driving test, which takes all of 40 minutes. This is possible only because we know a lot about what it means to be a human. To test self-driving technology we need to understand a lot about the AI that is behind it, which is an active area of research with much progress. Moreover, there is adversarial testing in simulation and on raw sensor data, which can tell you a lot about your system before any large scale real world testing begins, including when it might be safe to start real-world testing. Real-world testing at such large scales is neither a practical nor a necessary component to validating self-driving systems.

namibj 8y ago
Take a look at Germany, and how to get a drivers license there. Hint: if you are bad at parallel parking, you will fail the exam.
ux-app 8y ago
>We give them a driving test, which takes all of 40 minutes 40 minutes + 16 years to prime the neural net for - object recognition and classification - developing a theory of mind so that the NN can anticipate how other NNs will behave - etc The NN can't sit the 40 minute test unless they prove they can do successful object recognition and avoidance at low speed (while in bipedal mode). That they have the sufficient fine/gross motor skills to navigate a complex and changing 3D environment etc. It's going to take a while before we've got digital systems that can compare.

rdm_blackhole 8y ago

We should not make "better" the enemy of perfect or something of the essence. If we can have self-driving vehicles that are twice as safe as the average human, then they should be allowed on the streets. https://www.rand.org/blog/articles/2017/11/why-waiting-for-perfect-autonomous-vehicles-may-cost-lives.html https://www.rand.org/blog/articles/2017/11/why-waiting-for-p... I realize that accepting that a machine may kill you when you cross the street is a bit unsettling but if you think about it, it is no different than accepting that the person in the car behind you may have a moment of absence or not pay attention to the road end up killing you or a pedestrian. Yet having another human kill another human seems to be accepted by society? If car manufacturing companies take on the responsibility and financial burden on them to compensate families that are killed/injured by a self-driving car, just like insurance companies do nowadays on the behalf of the drivers at fault, then the problem becomes non-existent unless I am missing something.

rohit2412 8y ago
I am appalled at statements such as "twice as safe as human". We should be comparing driverless technologies with humans assisted by the same technology running in passive mode. Otherwise, you're implyjng that the developer of that technology would either release the technology as completely driverless or not release it at all.
- gsam 8y ago
  > Otherwise, you're implyjng that the developer of that technology would either release the technology as completely driverless or not release it at all. In some ways that statement is true. Ford and a number of other companies refuse to implement level 3 automation -- because of the switch-over costs and human-vehicle communication impedances which would almost certainly lead to more accidents.
  - Sophistifunk 8y ago
    Yup. If the computer works fine 99.9% of the time, there's just no way to keep the meat paying attention so it's there when that .1% situation pops up.

nmca 8y ago

This paper consistently assumes a fleet size of 100, and 24/7/365 driving. These are silly in different directions, but I think the former invalidates a lot of their claims. As a slightly not-random example, GM produces around 30,000 Chevy Bolts a year. Things start to look a little expensive, but the paper claims "This analysis shows that for fatalities it is not possible to test-drive autonomous vehicles to demonstrate their safety to any plausible standard". I think they are neglecting "without increasing fleet sizes beyond 100 vehicles".

nmca 8y ago
Ah, and indeed it seems like substantial (>10x) fleet size increases are planned. https://www.theverge.com/2018/1/30/16948356/waymo-google-fiat-chrysler-pacfica-minivan-self-driving https://www.theverge.com/2018/1/30/16948356/waymo-google-fia...
notahacker 8y ago
Their assumptions are just one way of reaching the 275 million mile mark needed to hit that 95% confidence interval. If we're looking at actual numbers, Waymo's recently boasted of hitting the 7 million mile mark over several years of testing so if anything, Rand's assumption of 275M miles in 12.5 years for a test programme looks to be erring on the generous side. Sure, if Chevrolet decided to use their entire production run of Bolts to test autonomous vehicle performance you'd get 275 million miles pretty quickly, but they might struggle to get consumer buy in and regulator approval for selling fully autonomous driving capability to 30,000 consumers unless and until they can produce evidence that their autonomous mode accident rate is lower than human drivers... And there's a much bigger problem, in that the software being tested is continually evolving, and modifications which result in improvement in general driving conditions don't necessarily tends towards stable (or monotonically reducing) numbers of rare accident-inducing bugs.
amputect 8y ago
FWIW the paper mentions that Google's autonomous driving fleet currently has 55 cars, so I suspect the 100 fleet size was picked to reflect a significant improvement over a notable competitor in the field. The article mentions Google's self-driving fleet size but I didn't notice other companies. I wonder what the fleet size is at other companies that are working on this, anyone know? The reason I ask is that I'm curious if "making more cars" is really the bottleneck. It's totally possible that the Waymo team doesn't have the headcount to keep up with the data output from the existing 55 units, or the fleet logistics bandwidth to keep up more than 50 or 60 heavily custom self-driving vehicles. It may very reasonably be a staffing issue rather than a money one, and it's possible that even if Google bought the rest of the self driving industry they might not be able to maintain more than a couple hundred cars. Disclaimer: I work for Google, but not on self driving cars. I actually know almost nothing about our self driving efforts, sorry.

8bitsrule 8y ago

First show me a machine that won't run into a barrier, or a human, or a parked fire truck, or a parked police car at 40-60mph without braking. Then ask this (for now ridiculously abstract) question again.

IanCal 8y ago
These things seem to happen with humans too. But what you're asking leads directly to the question being asked here! How long would something have to drive for before you believe me it's unlikely to do one of those things? Is it even possible for me to do this?
- diggernet 8y ago
  It would have to drive fully autonomous in all conditions that humans drive in, for enough miles to produce a statistically valid lower rate of accidents than humans. Since it can't drive fully autonomous until it has proved itself, it would be allowed to have a safety observer onboard. But here is the catch: Any intervention by the observer counts as an accident, since we are simulating what it would do without a human. Assume worst case. Oh, and software updates reset the clock.

mgamache 8y ago

Totally unsupported by examples, but it seems like the huge economic benefits will cause society to underestimate to risk of autonomous vehicles. We'll debate and distort the relative risk and provide plausible deniability for the right people. In the end, we'll tweak the AI based on real world failures until it's good enough to win the PR battle. This is assuming it can be tweaked to work. The key will be protecting (or limiting) the liability of car manufactures from lawsuits. If AVs are involved in 37,000 fatalities a year (the 2016 human stats) the lawsuits will be untenable. Maybe there's a plan to handle this, and I just haven't seen it. Just my opinion...