Toyota Research Institute head says full autonomous driving is “not even close” • Hacker News Board

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#131

Post by Piskvorrr » Mon, Aug 28, 2017, 4:01 PM UTC

limeblack wrote: post #5

One thing I would love someone to explain is how can people assume cars are going to be automated soon when both planes and trains haven't been automated for various reasons. They are different issues I have gathered but essentially shouldn't trains be dead simple but to this day at least the trains(not metras though) often have someone to stop the train in case of emergencies. Planes are more complex but have lower…

Oh, new trains are full auto, no worries. Retrofitting an entire fleet and tracks, now there's the issue - follow the money, and you'll see why even automatic trains are only programmed to follow the usual route, and for diversions, the driver would need to step in. Plus having a person in the cockpit to push the Big Red Button helps with all sorts of legal issues.

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#132

Post by rayiner » Mon, Aug 28, 2017, 4:01 PM UTC

redbeard0x0a wrote: post #94

People have a habit of thinking linearly, however technology builds on top of technology (leading to exponential improvements). Half way through the Genome project, they had only mapped out 1% of the genome. A lot of people in that field and working on that project were thinking the same thing - "not even close". But the next 7 years, they completed the next 99%. This is the power of exponential progress. The only th…

Technological curves tend to look exponential at first, then logarithmic. We went from first manned flight to first supersonic manned flight in just 44 years. Seventy years later we still don't have supersonic commercial flights. In the energy sector, the thermal efficiency of coal plants improved dramatically from 1880 to 1950, then pretty much hasn't improved at all since then. The number of years someone can expect to survive after qualifying for Social Security at age 65 has gone up only about five years since the age was first set to 65 back in 1935.

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#133

Post by throwawayjava » Mon, Aug 28, 2017, 4:01 PM UTC

limeblack wrote: post #5

One thing I would love someone to explain is how can people assume cars are going to be automated soon when both planes and trains haven't been automated for various reasons. They are different issues I have gathered but essentially shouldn't trains be dead simple but to this day at least the trains(not metras though) often have someone to stop the train in case of emergencies. Planes are more complex but have lower…

The economics couldn't be more different.

Driver salary is 26% of the cost of trucking.

Driver salary is 52% of the cost of a taxi.

Those numbers are in the single digits for aircraft and trains.

We could replace pilots and train operators, technologically speaking. But the incentives are not (yet) large enough to overcome the regulatory and cultural barriers. We'd rather have that extra order of magnitude of safety, because it costs so little compared to other expenses.

The cost/benefit calculation for drivers is fundamentally different.

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#134

Post by nradov » Mon, Aug 28, 2017, 4:01 PM UTC

SEJeff wrote: post #54

limeblack wrote: post #5

One thing I would love someone to explain is how can people assume cars are going to be automated soon when both planes and trains haven't been automated for various reasons. They are different issues I have gathered but essentially shouldn't trains be dead simple but to this day at least the trains(not metras though) often have someone to stop the train in case of emergencies. Planes are more complex but have lower…

Planes are mostly automated. The pilots are involved for takeoff and landing or emergencies, but little else. All modern Airbus or Boeing planes are mostly flown with autopilot. The human component is still involved primarily for the "what if it breaks" case where the software isn't there today. A computer would likely have never considered putting a plane into the Hudson river, but that pilot saved a lot of people b…

Military UAVs have a high crash rate.

There's no autopilot which could have successfully landed an Airbus A320 in the Hudson River.

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#135

Post by Piskvorrr » Mon, Aug 28, 2017, 4:02 PM UTC

Eridrus wrote: post #45

limeblack wrote: post #5

One thing I would love someone to explain is how can people assume cars are going to be automated soon when both planes and trains haven't been automated for various reasons. They are different issues I have gathered but essentially shouldn't trains be dead simple but to this day at least the trains(not metras though) often have someone to stop the train in case of emergencies. Planes are more complex but have lower…

If we had autonomous planes it wouldn't significantly reduce cost or increase usage. But there is a lot of investment in autonomous drones. Planes also have a very high safety bar to begin with, which cars do not. Same thing about trains, there's very little value in doing so, and resistance in the form of unions in some locations. Some trains are fully automated though, here's an article with some more info: https:/…

MTA has 100+ years of legacy track. Imagine trying to retrofit anything into a 100-year-old codebase. Ouch.

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#136

Post by 0xbear » Mon, Aug 28, 2017, 4:02 PM UTC

First step should be a car with driver assistance so sophisticated, a human driver can't crash it. Seems like a much more tractable problem, and one I personally would pay quite a bit of money for in a few years, when my son is old enough to drive.

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#137

Post by jason_pomerleau » Mon, Aug 28, 2017, 4:06 PM UTC

strebler wrote: post #75

dperfect wrote: post #57

I've been saying this for a while. The problem is that people are mostly familiar with the rate of overcoming traditional technical challenges (e.g., making your current computer/phone more compact, improving graphics quality, making cameras and screens with more pixels, etc.). What people don't realize is that full autonomous driving will require more than just faster/smaller/cheaper technical innovation - it will r…

You've hit the nail on the head - audio recognition is quite poor after (a lot of) concerted effort over decades. Driving a car safely is significantly more difficult (and dangerous). That being said, Toyota's (and Subaru's) approach is the smart way forward - add sensors and capabilities that augment the human, but leave the human always responsible and in control. In order for a crash to happen, the human AND the m…

Once the machines are better than humans at a macro level, insurance companies will notice, and I think this will cause widespread adoption pretty quickly.

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#138

Post by saosebastiao » Mon, Aug 28, 2017, 4:06 PM UTC

redbeard0x0a wrote: post #94

People have a habit of thinking linearly, however technology builds on top of technology (leading to exponential improvements). Half way through the Genome project, they had only mapped out 1% of the genome. A lot of people in that field and working on that project were thinking the same thing - "not even close". But the next 7 years, they completed the next 99%. This is the power of exponential progress. The only th…

Futurists have a habit of thinking exponentially for things that ultimately end up with asymptotic progress.

Driving cars is ultimately about learning and learning is asymptotic. A child typically progresses from not knowing how to control their vocal cords to 50% of their peak vocabulary within 5 years. It takes several times that amount of time to be able to capably work a white collar job (a feat many fail at), and decades more to be able to make a meaningful and attributable contribution to society (something almost all fail at).

The genome project is a terrible example because mapping progress is a second order effect to the actual technological progress of learning how to map the genome. It would be reasonable to project that once we know how to build a level 5 car that AVs will consume 100% of miles travelled within a very short period of time. But learning how to get to level 5 could very well be asymptotic in the same way that it takes decades of asymptotic refinement to turn an exponentially learning toddler into a PhD.

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#139

Post by rainbowmverse » Mon, Aug 28, 2017, 4:06 PM UTC

alphonsegaston wrote: post #88

QAPereo wrote: post #65

Earlier quoted context omitted.

It really makes the insane gamble of the Uber VC's ultimately terminal.

Maybe for this particular investment, but I'm pretty sure the wider intention was to legitimize business practices where the poor live tenuous lives as an app-directed servant class. Uber made this kind of neo-feudalism so socially acceptable that we now celebrate "Uber-for-X" as "innovation."

It already happened before Uber, except now the app is in the hands of the person being exploited, not the managers creating optimized schedules that make it near impossible for individuals to hold enough jobs to pay the bills.

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#140

Post by Piskvorrr » Mon, Aug 28, 2017, 4:07 PM UTC

Bartweiss wrote: post #80

limeblack wrote: post #5

One thing I would love someone to explain is how can people assume cars are going to be automated soon when both planes and trains haven't been automated for various reasons. They are different issues I have gathered but essentially shouldn't trains be dead simple but to this day at least the trains(not metras though) often have someone to stop the train in case of emergencies. Planes are more complex but have lower…

One point would be that both planes and trains are professionally operated already. That means three of the biggest incentives to automate are missing. First, they have few human-error crashes; they're both significantly safer than automobiles, so unless automation can do better-than-human disaster recovery (not likely yet) there's minimal safety advantage. You have to outperform well-trained humans instead of random…

Good points. Automated train control maps are not even 2D: they're directed graphs, precisely mapped to sub-foot precision, with all permissible edges fully checked in advance, at design and at mapping. Far easier to navigate than a road network, which can be simplified and approximated to a directed graph, but 1. at the cost of some precision, and 2. orders of magnitude more likely to change unnoticed.