Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror
Get HideMyAss! VPN, PC Mag's Top 10 VPNs of 2016 for 55% off for a Limited Time ×
Transportation AI

How Much Will Autonomous Cars Really Help? (theconversation.com) 211

An anonymous reader writes: An opinion piece at The Conversation questions the common belief that autonomous vehicles will easily solve a host of problems with road-based travel, including safety and traffic. "Assuming autonomous vehicles were one meter apart and traveling at 100 kilometers per hour (an aim that has been stated as the ultimate hope) this would mean around 25,000 people per hour could be taken down a freeway lane. While impressive, this movement capacity is only half that of a train. But getting to this capacity means 100% of vehicles are under control of a guidance system, with none under independent control. As soon as one car does this, the whole system would slow down considerably, as is seen on freeways now." The writer argues that a better role for autonomous cars might be to take passengers to and from hubs for public transportation.
This discussion has been archived. No new comments can be posted.

How Much Will Autonomous Cars Really Help?

Comments Filter:
  • We keep getting told driverless cars are crash-proof and theoretically perfect drivers or as close as you can possibly get to it but I'd say that's mostly hype. There will still be accidents, including fatal ones and I would think a worse, more catastrophic breed of accident will appear once they start having cars drive in very close formation.

    Safety standards will slip, there will be more of a drive to improve fuel efficiency and more risks taken. Redundant systems will eventually be scrapped to save cos
    • by NotInHere ( 3654617 ) on Saturday December 05, 2015 @04:30PM (#51063687)

      Well, it eliminates whole classes of cases for accidents. All those people who drive while they are drunk, or drive too fast because they are too late, or drive too fast because they like driving fast, or drive too fast because they don't know better, or etc. There are tons of accidents caused by older people who are too senile to drive a car. This can be helped by taking away their license, but staying at home surely isn't a good therapy for old people to stay healthy.

      Also, if all cars are driverless, they always know when faster cars can get before slower cars on a one lane per direction road.

      This won't solve all accidents, but it will certainly improve the situation.

      • by ShanghaiBill ( 739463 ) on Saturday December 05, 2015 @05:17PM (#51063849)

        This won't solve all accidents, but it will certainly improve the situation.

        Indeed. The naysayers seem to forget that self-driving cars already have millions of miles of testing. If they were accident-prone, the data would show that, and it does not. Driving in close formation, or "platooning" is well tested. I remember seeing test car platoons on I-5, north of San Diego, in the 1990s. TFA is mostly nonsense and conjecture. It says that a single non-autonomous car will "slow the system down considerably". I see no reason that would be true. A human driven car would just mean one car would have a normal gap, but that wouldn't slow down other cars. I have heard the opposite: That even a few autonomous cars can make a big difference in preventing congestion, since they have more information about traffic conditions ahead, and can react quicker, so they smooth out the "accordion effect" for themselves as wells as all the cars behind them.

        Comparing self-driving cars to trains is idiotic. I can't take the train to the grocery store. A stream of self-driving cars may have half the bandwidth of a passing train, but not if you consider the gaps between trains, which are usually far more than the length of the train. A mile of passenger rail costs about $100M. A lane of asphalt costs about $1M per mile.

        • Re: (Score:3, Insightful)

          by WarJolt ( 990309 )

          The article suggests a new mode of driving only possible by driverless cars, but 1 meter apart is kinda ridiculous. A tire failure in any one of those cars could cause a pileup of unimaginable purportions. I'd settle for autonomous cars driving at human following distances because we know humans can do it. Even at human following distances autonomous cars can improve things because even the simplest actions on the road have huge unseen consequences from the drivers perspective. You can avoid the problem whe

        • Comparing self-driving cars to trains is idiotic. I can't take the train to the grocery store. A stream of self-driving cars may have half the bandwidth of a passing train, but not if you consider the gaps between trains, which are usually far more than the length of the train. A mile of passenger rail costs about $100M. A lane of asphalt costs about $1M per mile.

          Actually, construction cost for a mile of high speed railroad track is $1M-$2M. For interstate highway, it is $1M -$5M. Neither of those figures includes land acquisition costs.

          In reading the article, they aren't comparing cars to trains for going to the grocery store, but instead for commuting to the workplace, which is the majority of congestion/accident problems they are trying to solve.

        • Comparing self-driving cars to trains is idiotic. I can't take the train to the grocery store.

          I can, as can millions of others. If something seems idiotic, usually it's because you haven't grapsed the full scope of the issue.

          A stream of self-driving cars may have half the bandwidth of a passing train, but not if you consider the gaps between trains, which are usually far more than the length of the train. A mile of passenger rail costs about $100M. A lane of asphalt costs about $1M per mile.

          And what does a mile of cars cost?
          The most efficient rail system can transport 75000 people per hour per line, and it returns a net profit to its owners. Please show me an example of a road that comes even close to that?

      • It's not just people who drive too fast. After 30 years of driving and observing traffic patterns, the most frequent cause of sudden speed changes I see is people driving too slowly as they try to merge onto the freeway. That causes people already on the freeway to have to slow down or change lanes to avoid them, which increases the risk of an accident.

        And TFA comparing car and train capacity is silly because it excludes time spent stopping to load/unload passengers. The whole reason people drive cars
        • the most frequent cause of sudden speed changes I see is people driving too slowly as they try to merge onto the freeway. That causes people already on the freeway to have to slow down or change lanes to avoid them, which increases the risk of an accident.

          It is even worse if you are joining the freeway immediately behind that person because you are the one that the truck behind will crash into.

          And TFA comparing car and train capacity is silly because it excludes time spent stopping to load/unload passengers. The whole reason people drive cars instead of take public transportation is because (1) they're sick of waiting for the bus/train to show up, and (2) they're sick of the trip taking 2x-3x longer than if they drive because the bus/train has to stop at a bunch of places they're not interested in going.

          That is obviously an American view; I understand trains there are rather slow, and if people are that slow getting on and off them it is probably because they are not used to them enough. I try to make workaday journeys by train in the UK because I am sick of sitting in my car in traffic jams taking 2x-3x longer than the train. The last time I drove from Bristol to

          • A London Underground railway line, in its 12ft diameter tunnel, even with its stops, can shift more people per hour than a three-lane highway, both at max capacity.

            This seems to be the crux of the issue. Robot cars are an American thing, becasue for some reason American can't see past the car as a mode of transport.
            Real world numbers (not from TFA) are a peak of 2000 vehicles per hour per lane of freeway compared to 75000 per hour for rail. It's a no brainer.
            Any sufficiently large and dense city would be far better of with a metro rail system like London, Paris, Hong Kong, Singapore or Tokyo.

    • by Macman408 ( 1308925 ) on Saturday December 05, 2015 @04:54PM (#51063765)

      Something north of 90% of accidents are preventable; take a look at table 8 here: http://www-nrd.nhtsa.dot.gov/P... [dot.gov]
      That table shows the 'critical event' in an accident, which is what made it unavoidable. Just 1.4% of accidents are from an object or animal in the road. Likewise, only 1.2% are due to a vehicle problem, although a large percentage of those are improper maintenance, which would be solved by some autonomous vehicle business models where they are owned and maintained by a fleet company (such as Uber).

      So we can prevent 90% of accidents, but you think it's not worthwhile because the other 10% still happen?

      Furthermore, if the fleet model is adopted, it actually becomes more likely that safety improvements will make more financial sense; far fewer cars are needed in the fleet, so the costs are amortized over more people. But in either case, safety standards are set by the government, and we can choose to raise or lower them as we see fit, completely orthogonally from whether cars are autonomous or not.

      • People make mistakes. A huge percentage of accidents occur to drivers aged 25-65 (60%) on a clear day (74%). Driverless cars will make an impact by going after the dominant sources of accidents.

        Personally, I suspect that partially interactive and/or assistive technologies will be deployed first. That way the care doesn't have to handle every case properly, particularly winter and icy conditions.

        • I suspect that partially interactive and/or assistive technologies will be deployed first.

          This has ALREADY HAPPENED. Tesla Autopilot is available to consumers, and does 80% of what you would expect a self-driving car to do.
          Here is a video [youtube.com] of some idiot who climbed in the backseat while his Tesla was driving on a busy freeway.

      • "Something north of 90% of accidents are preventable;"

        Which is why they shouldn't be called accidents. They're crashes. Call them that.

    • It's pretty certain that autonomous cars will cause a lot less accidents than human drivers do.To the extent that liability insurance premiums are expected to drop significantly.

      You can speculate all you like about "a worse, more catastrophic breed of accidents", on average the amount of damage and loss of life and limb can only be expected to decrease sharply.

      But yes, there will probably be accidents. It would be a bloody miracle if there weren't any. Actuator malfunctions, programming errors, hardware

    • There will still be accidents, including fatal ones and I would think a worse, more catastrophic breed of accident will appear once they start having cars drive in very close formation.

      I'd assume catastrophic accidents would be less likely as car AIs could detect the rare conditions that could lead to one (as opposed to human drivers who might lack the knowledge) and react to stop it from propogating.

      Safety standards will slip, there will be more of a drive to improve fuel efficiency and more risks taken. Redundant systems will eventually be scrapped to save costs and we'll be back to (or worse) than we are now.

      There's certainly going to be some back and forth between safety, speed, and cost. But long term it should be safer since driverless cars make safety cheaper.

      Above all the fact remains that we live in an imperfect world where sh1t doesn't always go according to plan. Moose will still jump infront of robo cars and get killed, as will children - you just can't stop a lump of metal traveling at 100kph in 0 time using software alone (and even if you could, doing so would kill the occupants)

      No but you can vastly improve upon the reaction time and quick decisions of a human. You could get something like for every two childre

    • by thej1nx ( 763573 ) on Sunday December 06, 2015 @03:03AM (#51065971)

      Looks like you are arguing from a bias. First, your demand that there should be zero accidents is an idiotic one. Statistics and tests have already proven that there are less number of accidents with automated cars.

      Secondly, your pulled-out-of-your-ass argument about dropping safety standards seems to never happened to say flights, or industrial machinery. You put people's lives at risk, your product doesn't sells and you get sued too. Hell of a dis-incentive.

      All the argument about being unable to stop a lump of metal travelling at 100kph in 0 time is the most moronic thing I have heard. Do you have some special telekinetic powers to be able to do this, if you had manual control?

      The key thing you are missing is that the software is not getting distracted while texting, is not going to be drunken driving and is not going to get into a drag race with others on the road. Its 100% focus is on avoiding collisions while getting you where you want to go.

      • "Statistics and tests have already proven that there are less number of accidents with automated cars."

        So far _every single crash_ involving an automated car had a human in control of the other vehicle and it was determined to be the human's fault.

      • Statistics and tests have already proven that there are less number of accidents with automated cars.

        Statisitcs and tests from controlled tests which are completely unrelated to an uncontrolled environemnt that is a public road.

        The key thing you are missing is that the software is not getting distracted while texting, is not going to be drunken driving and is not going to get into a drag race with others on the road. Its 100% focus is on avoiding collisions while getting you where you want to go.

        The key thing you seem to be missing is that I have an interest in getting places quickly, and can do that now without the need to purchase the product you are selling.
        Really, the robot car crowd is starting to sound an awful lot like the TSA. Perception of safety, at any cost!

    • more catastrophic breed of accident will appear

      Yes, in the current condition autonomous cars will make it safer to drive and be more efficient... short term after the hype is over and the real autonomous cars come out. But the players in the industry will sell new hype and the system will get push to its limits. That's when accidents will occur... much like moving to drive-by-wire tech (e.g. unintended acceleration).

      Having worked with the technology--I see the new breed of accident will not be car-car, but

  • by bullgod ( 93002 ) on Saturday December 05, 2015 @04:08PM (#51063575)

    ... is getting back from the pub after I've had a skinful.

    • by 93 Escort Wagon ( 326346 ) on Saturday December 05, 2015 @04:18PM (#51063615)

      Actually, this (loosely) has been what I see as the most beneficial use of the technology - giving mobility back to people who can't / shouldn't drive.

      My mom still drives, and has her faculties. But, at 78, it's not a given that will continue for too many more years. I dread when we'll have to say "mom, you can't drive anymore... we're worried you're going to hurt someone or yourself". An autonomous car would go a long way in helping her maintain some independence, when she reaches that point.

      • A 30 minute commute 5 days a week has convinced me human beings are too prone to mistakes to drive a car safely. I see close calls almost daily.

  • by zippthorne ( 748122 ) on Saturday December 05, 2015 @04:18PM (#51063611) Journal

    With autonomous cars, the train actually has a chance of achieving it's 2x 1-lane capacity (noting that heavily traveled freeways are currently 2-4-lanes per direction of travel, though...) because people can take an autonomous car from their home to the train station, and then from the train station to their destination - the high cost of taxis rides isn't to support to the cost of the vehicle and its support, it's mostly supporting the cost to support the control system (a.k.a. the driver)

    • Yeah really autonomous cars imo should be viewed as circulators. Take mass transit to a nearby hub and then hop on a private circulator to do the last mile jaunt to your house.

  • While true enough autonomous cars to public transport makes a degree of sense, the author suffers from thinking most people are urbanites like them.

    Unless you want the cost and upkeep of laying rail to the sticks, public transportation only works with a high enough degree of population density or in-between routes from major cities. Everyone else is left behemoth vehicles to carry supplies twice a month where public transport falls short. Not to mention you haven't solved congestion issues like the author s

    • by Calydor ( 739835 )

      Pretty much this.

      If I were to drive to the nearest train station from where I live, I pass at least three supermarkets on the way. That train station furthermore doesn't have direct links to the places I usually go.

      What is the benefit of taking the train for me?

      • I have light rail almost in my back yard (3 houses away as the bird flies, but a 6 minute walk in reality). My work also has a light rail station only a 10 minute walk away. Great right? Even though it is 25 minute drive, and a 40 minute bike ride, the train is about 50 minutes door-door. I can easily swing by a grocery store on the way home using my bike or car, but not with the light rail. So the light rail is my worst option even though I have a nearly ideal situation.

        To be honest I would rather see

  • by godrik ( 1287354 ) on Saturday December 05, 2015 @04:28PM (#51063675)

    I do not really believe that self driving cars will significantly reduce my transportation time. But I expect them to reduce the number of traffic accident. In a traffic jam, drivers can frustrated and bump in each other. I highly doubt self driving car would do that. Also, I do not care as much being in a traffic jam if I am not the one driving the car. Finally, if the car drive itself, then I can take more long distance trips easily: push the buttons, go to sleep, wake up in a different state.

    This is the real reason I loved riding public transportation so much when I was living in France. It might not be the fastest way of moving around. But it was definitely the way that was consuming the less of my attention time. Made me arrived at work after 30 minutes of playing the nintendo DS. Much better than after 20 minutes of dealing with traffic congestion.

    • by khasim ( 1285 ) <brandioch.conner@gmail.com> on Saturday December 05, 2015 @04:51PM (#51063751)

      I do not really believe that self driving cars will significantly reduce my transportation time.

      I do.

      But I expect them to reduce the number of traffic accident. In a traffic jam, drivers can frustrated and bump in each other. I highly doubt self driving car would do that.

      And that is why I believe that they will improve the commute time. Fewer accidents to avoid. Fewer accidents that the idiots have to slow down and look at. And if the idiots really have to look at the remaining accidents, the car can still do the driving.

      • Oh this times 1000. The majority of freeways are not suffering because they can't get cars 1m apart, the majority are suffering because people don't know how to navigate, they try to cut to the front, cross 3 lanes of traffic within 100m of an exit, or (my personal favourite here) two trucks speed limited to a lower max speed than the motorway try to race each other in a contest to see who's speed limiter is calibrated slightly higher while traffic queues up behind them.

        And that's before an accident which c

    • by lorinc ( 2470890 )

      This is the real reason I loved riding public transportation so much when I was living in France.

      Jeez! Your post was pretty believable until that one sentence, since French public transportation is probably one of the things that concentrate most of the world hatred.

      At least it's good to know that the French government is spending money on lobbyists at Slashdot. Either that, or you are a pervert mind on revenge that want to make other people suffer the way you suffered, by having them believe they can take French trains without getting stuck until the end of time due to labor strikes, overaged failing

      • by godrik ( 1287354 )

        Yeah, for 3 weeks a year transportation is terrible in Paris.
        In the rest of the country, there are far fewer issues. But you know it in advance so you plan for it. In regular operation, the network transports you reasonably fast and for relatively cheap.

        There a bunch of things that sucks in France (administration opening hours and slugishness, the country being at a standstill on sundays and from July 15 to August 15); Public transportation is not one of them.

        • by vovin ( 12759 )

          Of the dozen or so cities I have used the metro/subway system Paris is by far the worst [croweded and rude people] and least reliable (multiple breakdowns). Mexico City has a better run metro system and that's just sad.
          Considering that Barcelona, Madrid, Rome, and Milan are all relatively close comparables and infinitely better in pretty much all aspects in my experience.
          If you think the Paris subway is acceptable much less 'good' then I just have to assume you are a gallophile that has no comparable experi

    • In a traffic jam, drivers can frustrated and bump in each other. I highly doubt self driving car would do that.

      True. In a bad traffic jam, how would frustrated people who have nothing to do at all but look out the windows react? Let's hoe we don't find out, because 88% of those people have guns.

  • 25,000 people per hour could be taken down a freeway lane. While impressive, this movement capacity is only half that of a train

    The train comparison is completely fatuous since no train can carry 25,000 and the smaller ones don't run frequently enough to sustain that level of movement. Plus, last time I checked, I can't get a train from right outside my doorstep.

    Trains have some uses, but they lack the versatility of cars, and far more expensive to build and operate and they are only comparably efficient when full or nearly so.

    • The article is from Australia and I'm not terribly familiar with the attitude toward public transportation there, but at least in the US, apart from a few pockets in big cities, you will not pry cars from their owners without at least a generational change. Also, the author seems to have no clue just how advanced these prototype vehicles have become; they are very able to navigate among unpredictable obstacles on city streets without being slowed to a crawl. The premise is decent - that autonomous vehicles

    • Yes, the problem with poor implementations of public transport is that they're slow, expensive, and inconvenient. Having lived in places, including smaller towns connected by rail services, that have coherent public transport systems and a strong focus on rail, I've found getting around faster, cheaper, and more convenient than cars by a long way, e.g. getting to and from work is at least twice as fast by rail as by car. The biggest complaint in those places is that there aren't enough taxis available at pe

    • " Plus, last time I checked, I can't get a train from right outside my doorstep."

      The funny thing is: London expanded to where the train stations were built.

      Not the other way around - and apart from the USA or other countries with appalling public transport structures, this is the normal way things happen.

      So, you may not be able to get a train right from your doorstep, but if you live in a metropolis like London or Paris, you almost certainly CAN find one within 5-10 minutes walk (10-15 in my case but I'm in

  • This suggests the most efficient use of the roadway is to get the vehicles as closely packed together as possible. Its an assumption many motorists seem to make while flow theory states otherwise.
  • An opinion piece at The Conversation questions the common belief that autonomous vehicles will easily solve a host of problems with road-based travel, including safety and traffic. "Assuming autonomous vehicles were one meter apart and traveling at 100 kilometers per hour (an aim that has been stated as the ultimate hope) this would mean around 25,000 people per hour could be taken down a freeway lane.

    Just to put that in perspective, that's almost 35 times what is currently possible (~700 cars per freeway l

    • There are drawbacks of increased speed even if accidents can be cut to zero, such as increased gas usage. This can be solved by moving to electric vehicles, possibly running off power rails on the highways etc.

      However, in urban areas speed is major factor for noise generation.
      I personally live next to a major road where vehicles go past at 70+ km/h and the noise is very bothersome. It hovers around 60-65 dBA, peaks at at 75 when someone with a case of lead foot powers by. The majority of the noise is simply

      • by dgatwood ( 11270 )

        The increase in energy usage won't entirely go away by moving to electric, because the wind resistance will still increase as your speed does. With that said, the wind noise will be greatly improved by reducing the space between vehicles, as will the increased energy usage caused by drag.

    • 25,000 people. Not 25,000 cars. They seem to be assuming 5 people per car, so that's 7 times current capacity.

      I don't know where they get their 50,000 people per hour on heavy rail though.

  • In densly populated areas mass transit is the right technology to transport people and goods. In future we will augment that with other short and mid range transport technologies, like bikes, e-bikes, low speed autonomous cabs etc.

  • "movement capacity is only half that of a train"

    Call me when the train can take the next left. Until then, trains can kind of go screw themselves for everything by dense urban people movement, in areas where you don't have to buy up insanely expensive real estate to lay down new tracks. This is not a socialist utopia, as in Sim City, where you can tear down a stadium because Skywatch One is reporting heavy traffic, in order to improve traffic flow.

  • by kaur ( 1948056 ) on Saturday December 05, 2015 @06:07PM (#51064069)
    The authors misunderstand the point of autonomous cars.
    They won't be here for efficiency, safety or speed.
    They will free up the time for the driver.
    Instead of keeping my hands on the wheel, I can work, shave or have sex.
    THIS will be the benefit.

    It also means that the decision to drive or not to drive will be much cheaper. Today, a 30 minute drive will take 30 minutes off from my life. Tomorrow, it won't. I can still do what I want while being driven - which means that I will "drive" much more. The ones who can allow to own / rent robot cars will suddenly start moving around a lot more. This will create more traffic, maybe exponentially so. The green, eco-friendly vision of reduced traffic via autonomous vehicles is all wrong.

    It will also affect urban planning in ways that nobody can yet comprehend nor predict.
    • Instead of keeping my hands on the wheel, I can work, shave or have sex.

      This is Slashdot.

  • The other thing that self driving vehicles will do is allow us to timeshift some of the traffic to when freeways aren't busy. E.g. freight can be moved to 10pm - 5am in urban areas, since we won't need to worry about the driver's exhaustion level. Fright would also be more efficient as driverless trucks don't need to take rest stops, and can be lighter because they don't require human amenities.

    Also, if I had a driverless car, and a comfy seat, I'd not mind sleeping while my car drives me somewhere overni

  • They'll help me reclaim the hours I spend every day in traffic. Safety and speed will be nice, but those won't be significant until most of the cars on the road are autonomous. Until then, I'll be more than happy with the extra time to sleep, read, write, watch TV, play games, and so on.
  • of the cars blows a tire, hits a deer or piece of trash that fell off a pickup, or has some other problem. Then you get a 50 car pileup, even with all of the cars operating under control of a single system. Traveling that close assumes that the cars all have similar braking characteristics, tires, engines, and suspension in good condition, etc. Look at cars going down the road today. Every 3rd or 4th car spews smoke, is rusty, and probably has other problems due to little to no maintenance.

    The cars will

    • by MrL0G1C ( 867445 )

      Exactly this, the cars will only be able to slow down as quickly as the car in the car-train with the slowest brakes, if the slowest car can't slow down fast enough to avoid the accident then it hits all of the cars in front of it*.

      If the accident involves a car forcefully being stopped at quicker than the braking speed of the cars in the car train then a lot of autonomous cars (30+ or even 90+ for more lanes) will end up in a pile-up*.... Instead of 1 to 2 cars if they are leaving a 2-second gap.

      *if the ve

  • Most people look at the fairly obvious cost savings that autonomous cars will provide but miss the less obvious ones.

    For instance in many cities there is little competition among grocery stores because a few early movers grabbed up crap land where they could then afford to put up a huge store with a massive parking lot. Then the city either grew to surround their store, or the value in that section of the city went way up. Thus the barrier to entry is impossibly high. The only grocery store competition t
  • Driverless cars aren't about traveling at high speeds, packed together like sardines. Commercial aircraft fly themselves these days, but traffic controllers still keep them 3 miles apart. Such close formations of driverless cars would still result in massive pileups when one of the cars malfunctions and crashes.

    The point of driverless cars is to let me do something else while I'm traveling. I won't care so much about my one-hour commute if I can read the news or get started on my work day while I'm on th

  • Volvo has really advanced the game. The play is a kind of super cruise control. The car will drive itself safely when driving is boring. If the situation gets too complicated it turns control back to the driver, who can take control when he or she wants anyway. If the car can't wake the driver up the car will slow and stop when it is safe to do so. Volvo's CEO stated publicly that the company accepts liability in cases where the autopilot is in control when an accident takes place. He says any car company

"355/113 -- Not the famous irrational number PI, but an incredible simulation!"

Working...