6 ms·

~3-5 years is way too soon. The technology might be ready but the governments (think regulations, traffic laws, car insurance companies, car manufacturers, etc

by esMazer 12y ago

~3-5 years is way too soon. The technology might be ready but the governments (think regulations, traffic laws, car insurance companies, car manufacturers, etc etc etc ) in addition to the public (which is not ready either). Is going to take at least 10+ years for it to resemble something like you imagine it.

Dolimiter 12y ago
The technology is not nearly ready. Not in a city. It's like flying drone delivery. Easy to demonstrate in a single controlled setting, but nearly impossible to actually implement in the real world.
- ghaff 12y ago
  It's hard for me to even imagine how a self-driving car could manage somewhere like large swaths of Manhattan where a certain amount of, umm, aggression is needed to make any forward progress. Though it will likely happen some day (probably multiple decades from now). As it is, I'd even settle for voice control on my phone that could reliably understand me.
  - IanCal 12y ago
    > a certain amount of, umm, aggression is needed to make any forward progress They added a slight "aggression" back in 2011, if other cars aren't letting it out it starts pushing forwards a bit: > Sometimes, however, the car has to be more "aggressive." When going through a four-way intersection, for example, it yields to other vehicles based on road rules; but if other cars don't reciprocate, it advances a bit to show to the other drivers its intention. Without programming that kind of behavior, Urmson said, it would be impossible for the robot car to drive in the real world. http://spectrum.ieee.org/automaton/robotics/artificial-intelligence/how-google-self-driving-car-works http://spectrum.ieee.org/automaton/robotics/artificial-intel... I really don't understand why people think small behavioural differences like that require decades of work. Things that humans find difficult and stressful are likely to be the things that are significantly easier when you have a reaction time measured in milliseconds and full 360 degree vision.
    - ghaff 12y ago
      I hope you're right but the very uneven progress of AI generally--especially when it involves real world interfaces--makes me skeptical. As I wrote elsewhere, even voice recognition is very much still a mixed bag.
      - IanCal 12y ago
        But voice recognition is something we take many years to learn, even when are brains are highly plastic. I find it hard to understand people when they speak often. We disambiguate heavily based on context and expectation, very complex issues.
        ghaff 12y ago
        Oh, I fully agree. The fact that it's been the subject of so much research and remains only very incompletely solved demonstrates just how difficult a problem it is. I'm just saying that navigating a potentially dangerous machine through a physical world populated with erratic human beings is also very complex and difficult--albeit probably far less so in some contexts (limited access highways in good weather) than others (Manhattan, Boston).
- baddox 12y ago
  It's not really like flying drone delivery. Flying drone delivery is far simpler and far closer than general use self-driving automobiles.
  - zobzu 12y ago
    Actually landing UAVs is far harder than automatically parking a car. The driving vs flying after take off is what is easier. Landing and take off.. nope. Note that drone could mean a car in this case. UAV being an unmanned aerial vehicule.
    - baddox 12y ago
      With a modest infrastructure of marked landing pads on roofs, I think UAV landings are easier than automated car parking. If you're expecting delivery UAVs to fly under the tree line along sidewalks and yards to land directly on a residential porch, then yes, that's extremely difficult. But I highly doubt that this is a remotely feasible plan within a decade or so.
alonmower 12y ago
I agree that 3-5 years is a bit optimistic but I also bet that when it does happen it will transition far faster than we can imagine. You have a really nice set of aligned incentives here: - As a 'driver' I get a car that let's me free up large chunks of my time - As a company like Uber the driver 'input' to my model is now far more consistent, predictable, and inexpensive - As an insurance company I can start undercutting my competition by charging less to insure this substantially safer means of transportation while still pocketing a nice margin - As a car manufacturer I now have an extremely compelling reason to sell cars to people that would otherwise wait between 4-6 years in between car purchases - As a government I can (if I'm smart) try and pass laws that will result in drastic reductions in accidents, deaths, etc... I'll also likely be bringing in a lot of business and money by being on the leading edge of this (less traffic and congestion, new opportunities for businesses that leverage these cars, etc...). Whatever revenue I lose in traffic tickets I can make up for in other ways (like a reduced need for traffic enforcement). As soon as any of the above are able to make/save money this is going to snowball. The lobbying power and influence of the teamsters and taxi unions is probably the single largest hurdle but if self driving cars are as safe and cheap as it appears they will be this will be a speed bump, not a roadblock.
- georgemcbay 12y ago
  I'm not so sure the future is as bright as you make it out for all those groups. At the very least this change is going to bring with it a mountain of uncertainty. As a driver, you probably lose your job. Not with the 3-5 year out model, but the one that comes 10 years after that. Even with the 3-5 year model, suddenly the competition for your job becomes extremely tight and starts to focus on things like people skills and entertaining, since the automated car will be doing most of the "heavy lifting" most of the time in a way that is automated into a commodity. As Uber, you save a lot of money but you also lose a lot of moat, it becomes much easier/cheaper for other companies to suddenly field a fleet of cars. As a car insurance company things look really iffy for you, again not so much in the 3-5 year timeframe but a bit further out when the human is completely unnecessary since at that point the insurance will be more about insuring the companies making the cars and car software and less about personal insurance, which means there is still significant money to be made, but probably far less than now (if the safety of automated cars turns out to be as great as it should) and by far fewer players. As a car manufacturer, you'll do well at the beginning but the eventual model of a networked driverless transportation system should require far fewer people to own fewer cars and still live as if they do own cars, which is a long term problem for you when you've built a business optimized to sell cars to broad market consumers on a 5-ish year cycle. As a government you should, in an ideal world, be in pretty good shape for the reasons you outlined, but good luck dealing with the lobbyists from industries this may "disrupt". I do think driverless cars are quite a bit closer than most people think, for a lot of reasons, I just think they are going to upset a lot of applecarts on the way in, which isn't necessarily a bad thing unless you're an applecart vendor.
alonmower 12y ago
I agree that 3-5 years is a bit optimistic but I also bet that when it does happen it will transition far faster than we can imagine. You have a really nice set of aligned incentives here: - As a 'driver' I get a car that let's me free up large chunks of my time - As a company like Uber the driver 'input' to my model is now far more consistent, predictable, and inexpensive - As an insurance company I can start undercutting my competition by charging less to insure this substantially safer means of transportation while still pocketing a nice margin - As a car manufacturer I now have an extremely compelling reason to sell cars to people that would otherwise wait between 4-6 years in between car purchases - As a government I can (if I'm smart) try and pass laws that will result in drastic reductions in accidents, deaths, etc... I'll also likely be bringing in a lot of business and money by being on the leading edge of this (less traffic and congestion, new opportunities for businesses that leverage these cars, etc...). Whatever revenue I lose in traffic tickets I can make up for in other ways (like a reduced need for traffic enforcement). As soon as any of the above are able to make/save money this is going to snowball. The lobbying power and influence of the teamsters and taxi unions is probably the single largest hurdle but if self driving cars are as safe and cheap as it appears they will be this will be a speed bump, not a roadblock.

by esMazer 12y ago

Dolimiter 12y ago

The technology is not nearly ready. Not in a city. It's like flying drone delivery. Easy to demonstrate in a single controlled setting, but nearly impossible to actually implement in the real world.

ghaff 12y ago
It's hard for me to even imagine how a self-driving car could manage somewhere like large swaths of Manhattan where a certain amount of, umm, aggression is needed to make any forward progress. Though it will likely happen some day (probably multiple decades from now). As it is, I'd even settle for voice control on my phone that could reliably understand me.
- IanCal 12y ago
  > a certain amount of, umm, aggression is needed to make any forward progress They added a slight "aggression" back in 2011, if other cars aren't letting it out it starts pushing forwards a bit: > Sometimes, however, the car has to be more "aggressive." When going through a four-way intersection, for example, it yields to other vehicles based on road rules; but if other cars don't reciprocate, it advances a bit to show to the other drivers its intention. Without programming that kind of behavior, Urmson said, it would be impossible for the robot car to drive in the real world. http://spectrum.ieee.org/automaton/robotics/artificial-intelligence/how-google-self-driving-car-works http://spectrum.ieee.org/automaton/robotics/artificial-intel... I really don't understand why people think small behavioural differences like that require decades of work. Things that humans find difficult and stressful are likely to be the things that are significantly easier when you have a reaction time measured in milliseconds and full 360 degree vision.
  - ghaff 12y ago
    I hope you're right but the very uneven progress of AI generally--especially when it involves real world interfaces--makes me skeptical. As I wrote elsewhere, even voice recognition is very much still a mixed bag.
    - IanCal 12y ago
      But voice recognition is something we take many years to learn, even when are brains are highly plastic. I find it hard to understand people when they speak often. We disambiguate heavily based on context and expectation, very complex issues.
      - ghaff 12y ago
        Oh, I fully agree. The fact that it's been the subject of so much research and remains only very incompletely solved demonstrates just how difficult a problem it is. I'm just saying that navigating a potentially dangerous machine through a physical world populated with erratic human beings is also very complex and difficult--albeit probably far less so in some contexts (limited access highways in good weather) than others (Manhattan, Boston).
baddox 12y ago
It's not really like flying drone delivery. Flying drone delivery is far simpler and far closer than general use self-driving automobiles.
- zobzu 12y ago
  Actually landing UAVs is far harder than automatically parking a car. The driving vs flying after take off is what is easier. Landing and take off.. nope. Note that drone could mean a car in this case. UAV being an unmanned aerial vehicule.
  - baddox 12y ago
    With a modest infrastructure of marked landing pads on roofs, I think UAV landings are easier than automated car parking. If you're expecting delivery UAVs to fly under the tree line along sidewalks and yards to land directly on a residential porch, then yes, that's extremely difficult. But I highly doubt that this is a remotely feasible plan within a decade or so.

alonmower 12y ago

I agree that 3-5 years is a bit optimistic but I also bet that when it does happen it will transition far faster than we can imagine. You have a really nice set of aligned incentives here: - As a 'driver' I get a car that let's me free up large chunks of my time - As a company like Uber the driver 'input' to my model is now far more consistent, predictable, and inexpensive - As an insurance company I can start undercutting my competition by charging less to insure this substantially safer means of transportation while still pocketing a nice margin - As a car manufacturer I now have an extremely compelling reason to sell cars to people that would otherwise wait between 4-6 years in between car purchases - As a government I can (if I'm smart) try and pass laws that will result in drastic reductions in accidents, deaths, etc... I'll also likely be bringing in a lot of business and money by being on the leading edge of this (less traffic and congestion, new opportunities for businesses that leverage these cars, etc...). Whatever revenue I lose in traffic tickets I can make up for in other ways (like a reduced need for traffic enforcement). As soon as any of the above are able to make/save money this is going to snowball. The lobbying power and influence of the teamsters and taxi unions is probably the single largest hurdle but if self driving cars are as safe and cheap as it appears they will be this will be a speed bump, not a roadblock.

georgemcbay 12y ago
I'm not so sure the future is as bright as you make it out for all those groups. At the very least this change is going to bring with it a mountain of uncertainty. As a driver, you probably lose your job. Not with the 3-5 year out model, but the one that comes 10 years after that. Even with the 3-5 year model, suddenly the competition for your job becomes extremely tight and starts to focus on things like people skills and entertaining, since the automated car will be doing most of the "heavy lifting" most of the time in a way that is automated into a commodity. As Uber, you save a lot of money but you also lose a lot of moat, it becomes much easier/cheaper for other companies to suddenly field a fleet of cars. As a car insurance company things look really iffy for you, again not so much in the 3-5 year timeframe but a bit further out when the human is completely unnecessary since at that point the insurance will be more about insuring the companies making the cars and car software and less about personal insurance, which means there is still significant money to be made, but probably far less than now (if the safety of automated cars turns out to be as great as it should) and by far fewer players. As a car manufacturer, you'll do well at the beginning but the eventual model of a networked driverless transportation system should require far fewer people to own fewer cars and still live as if they do own cars, which is a long term problem for you when you've built a business optimized to sell cars to broad market consumers on a 5-ish year cycle. As a government you should, in an ideal world, be in pretty good shape for the reasons you outlined, but good luck dealing with the lobbyists from industries this may "disrupt". I do think driverless cars are quite a bit closer than most people think, for a lot of reasons, I just think they are going to upset a lot of applecarts on the way in, which isn't necessarily a bad thing unless you're an applecart vendor.

alonmower 12y ago