Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie

Self driving buses, trains, trucks etc

Options
1232426282933

Comments

  • Closed Accounts Posts: 1,857 ✭✭✭TheQuietFella


    https://www.rte.ie/news/analysis-and-comment/2018/0918/994606-broadband-plan-analysis/

    This link is about the broadband plans and then some people expect automated vehicles to be mobile on Irish road within 30 years! :D


  • Closed Accounts Posts: 2,891 ✭✭✭prinzeugen


    plodder wrote: »
    I'd say if the driver is not driving, it is a self-driving vehicle (at least some of the time).

    The point of my post was that the technology has already begun creeping in and will continue to do so.

    The article below says we would have self-driving convoys by the end of this year. So, three years seems a cautious enough prediction.

    https://www.theguardian.com/politics/2017/aug/25/semi-automated-truck-convoy-trials-get-uk-go-ahead-platooning

    Trials. Trials that will end in failure.

    If you search YouTube for "British Transport Films", there are a few videos from the late 60's & 70s.

    British Rail were working on self driving trains for years. Too risky. They also patented a flying saucer.

    A company saying something will happen does not mean it will.


  • Registered Users Posts: 15,704 ✭✭✭✭RayCun


    plodder wrote: »
    I predict they will within three years though :pac:

    a trial, maybe, but "commercially operated (not a controlled test, but trucks carrying normal freight for an existing distributor)"? Doubt it
    plodder wrote: »
    On the definition of "self driving" there seems to be a consensus on how it is classified with five different levels.

    Levels 4 and 5 are the only ones that count as self-driving.


  • Registered Users Posts: 7,133 ✭✭✭plodder


    prinzeugen wrote: »
    Trials. Trials that will end in failure.

    If you search YouTube for "British Transport Films", there are a few videos from the late 60's & 70s.

    British Rail were working on self driving trains for years. Too risky. They also patented a flying saucer.

    A company saying something will happen does not mean it will.
    Self driving trains are already in service (eg Copenhagen Metro). I understand the scepticism about level 4 and 5 automation, but the truck convoys are barely 2 point something. The technology is quite straightforward really. Whether they come to pass commercially soon, will be just down to economics. I'd be fairly sure the technology itself will be proven.


  • Posts: 0 [Deleted User]


    New legislation puts motor insurers in the driver’s seat of autonomous cars

    https://www.bankingtech.com/2018/11/new-legislation-puts-motor-insurers-in-the-drivers-seat-of-autonomous-cars/
    driverless cars expected to become common on Britain’s roads, the next few years will be transformative for the motor insurance industry, which faces both challenges and opportunities in making sure its products support new technology and meet consumers’ needs.

    In the UK, motor insurance is taken out by and for the driver, not the car – a logical approach given that over 90% of road traffic accidents are caused by driver error. Take that driver out of the equation with an autonomous vehicle, and who then is at fault in the event of a crash? And who, or what, should be insured?

    This question is further complicated by the varying levels of automation that these vehicles will provide. Thus far, cars have been created equipped with so-called level three automation, which means that they hand back control to the human driver when facing situations they can’t handle. In a crash, either the driver or the car itself might be at fault. This would then create an onerous process to establish who was legally in charge of the vehicle at any specific moment.

    The Automated and Electric Vehicles Act 2018, passed this summer, places the UK at the forefront of untangling the insurance implications of this new automotive technology. After consultation with the insurance industry, the government has put into place legislation that effectively places legal responsibility for road traffic accidents at the feet of the insurer. Consequently, no matter whether an accident is caused by an inattentive driver or a bug in the car’s software, the victim will not need to prove which was at fault. Their claim will be against the insurer directly, providing a clear legal path to liability.

    Although much remains to be worked out, the law also provides insurers with a new legal right to recovery, allowing them to claim costs back from motor manufacturers, software companies or other parties in cases where the vehicle or technology was found to have been at fault.

    Insurers have long welcomed advances in driving technology and automation. After all, less scope for human error means collisions – and associated claim pay-outs – become rarer. Many companies already incentivise the use of autonomous emergency braking (AEB), for example, to the extent that any vehicle where the technology is fitted as standard receives a lower insurance group rating, reflecting the lower risk and reducing the premium charged to customers.

    But calculating premiums in the event of a fully autonomous car is a more difficult task. Currently, insurers use factors such as occupation, age and driving record to weigh up how risky an insured driver may be. When the risk presented by that particular driver becomes reduced due to full automation, it’s back to the drawing board.

    The advent of self-driving cars means insurers will effectively have to create a whole new insurance sector without any established underwriting actuarial experience or data. To do so in a meaningful way, they will need access to the data which is available – be that information on vehicle specifications and what technology is enabled on a vehicle, or post-crash reports. Careful consideration will therefore need to be given as to how that data is stored and accessed, as the potential is clear for conflicts between the right to privacy and the need for information.

    However, if done correctly, the potential benefits to insurers are immense, from being able to accurately assess risks to reducing fraudulent claims.

    But are insurance companies ready to become big data aggregators and analysts? Within the financial services sphere, insurers have been relatively slow to digitise: it has only been in the last few years, for example, that online claims handling has become standard. While some firms are acquiring technologies – or indeed technology providers themselves – who are able to crunch vast quantities of data, others are still focussed on the task of bringing legacy systems up to standard in order to handle new inflows of information. All, though, are seeking to avoid being disrupted out of their own industry by newcomers, from start-ups to digital giants moving into the sector. Traditional insurers still have the edge, thanks to their underwriting expertise, insurance models and long years of experience, but how long this will continue to be the case remains to be seen.

    While some onlookers may question the need for motor insurance in a future where traffic accidents are all but eliminated, there are as yet no moves to change the existing legislative framework which makes such policies obligatory. What is clear is that the business model will shift, as insurers will pay out less by way of claims, there will be less opportunity for fraud, and there will be lower fractional costs in managing the book, potentially driving increased profitability.

    Whatever happens, it is clear that the future of car transportation in five or ten years’ time will look very different to today. The challenge that insurance companies are facing today is to design products that are fit for purpose and that meet the needs of the changing legislative and commercial environment, ensuring that insurers remain in the driver’s seat for many years to come.


  • Advertisement
  • Posts: 0 [Deleted User]


    In an article related to the opening of a new office for GM, there were some pretty interesting numbers. It's a good indication of the amount of money, brains and companies pursuing the goal of full autonomous vehicles

    GM’s self-driving unit Cruise is expanding to Seattle

    https://techcrunch.com/2018/11/19/gms-self-driving-unit-cruise-is-expanding-to-seattle/

    ............ The hunt for qualified people with backgrounds in software engineering, robotics and AI has heated up as companies race to develop and deploy autonomous vehicles. Founders from several self-driving car startups have regaled TechCrunch with stories of generous, even outrageous, compensation packages and poaching as they fight over a small pool of talent.

    For instance, there are more than 60 companies that have permits from the California Department of Motor Vehicles to test autonomous vehicles in the state.

    GM Cruise has one of the most aggressive timelines among companies hoping to deploy a commercial self-driving vehicle service. The company has said it plans to launch a commercial autonomous ride-hailing service in 2019.

    GM Cruise has the funds needed to pump up its workforce — the company received a $2.25 billion investment by SoftBank’s vision fund in May. More recently, Honda committed $2.75 billion as part of an exclusive agreement with GM and Cruise to develop and produce a new kind of autonomous vehicle.

    As part of that agreement, Honda will invest $2 billion into the effort over the next 12 years. Honda also is making an immediate and direct equity investment of $750 million into Cruise, which pushes the company’s valuation up to $14.6 billion. Honda’s investment gives the automaker a 5.7 percent stake in Cruise


  • Registered Users Posts: 15,704 ✭✭✭✭RayCun


    But are insurance companies ready to become big data aggregators and analysts?

    Insurers are, like, the original big data aggregators and analysts


  • Posts: 0 [Deleted User]


    Buses and taxis to lead UK self-driving public transport push

    Full article at https://www.theguardian.com/technology/2018/nov/22/buses-and-taxis-to-lead-uk-self-driving-public-transport-push
    The first driverless public transport services on the UK’s roads will be in action by 2021, the government has announced, led by an autonomous bus service crossing the Forth estuary in Scotland, and self-driving taxis in up to four London boroughs.

    The services should be on the roads within three years, under three pilot schemes announced by the Department for Business, Energy and Industrial Strategy. Full-sized driverless buses will run on a 14-mile route between Fife and Edinburgh, while Addison Lee and Jaguar Land Rover will lead two separate trials of driverless taxis in London before launching public services, again by 2021.


  • Posts: 0 [Deleted User]


    Further indication that these are nearly a reality

    Here come the self-driving taxis

    https://www.economist.com/gulliver/2018/11/23/here-come-the-self-driving-taxis
    FOR THE first time in decades, a futuristic transportation technology is on the verge of going mass market. It is not quite flying cars, but autonomous taxis—or “robo-taxis”, as they are sometimes called. They will transform cities and business travel, and will hit some other industries hard on the way too.

    Waymo, a subsidiary of Alphabet, a tech firm, is planning to launch the world’s first commercial driverless taxi service early next month in Phoenix. Waymo, which has yet to announce a name or launch date for the service, is also aiming to put autonomous taxis on the streets of California next year. Daimler and Bosch, a carmaker and engineering firm from Germany, announced this month that it would launch a self-driving ride-hailing service in San Jose, California, next year. General Motors, an American carmaker, is also hoping to launch a service next year using its Cruise vehicles. Zoox, a tech startup, aims to put its own all-electric self-driving taxis on the road by 2020.


  • Closed Accounts Posts: 7,907 ✭✭✭Stephen15


    Buses and taxis to lead UK self-driving public transport push

    Full article at https://www.theguardian.com/technology/2018/nov/22/buses-and-taxis-to-lead-uk-self-driving-public-transport-push

    The self driving buses may well still have drivers in them though just in case.


  • Advertisement
  • Closed Accounts Posts: 1,452 ✭✭✭Twenty Grand


    Stephen15 wrote: »
    The self driving buses may well still have drivers in them though just in case.

    Wouldn't be surprised if there are in the short term.
    Not drivers now, but someone just to push the red button if something goes wrong.

    I like the idea of train captains, which is where you remove the driver and have a customer service agent moving around the carriages checking tickets and the like.


  • Closed Accounts Posts: 7,907 ✭✭✭Stephen15


    plodder wrote: »
    Self driving trains are already in service (eg Copenhagen Metro). I understand the scepticism about level 4 and 5 automation, but the truck convoys are barely 2 point something. The technology is quite straightforward really. Whether they come to pass commercially soon, will be just down to economics. I'd be fairly sure the technology itself will be proven.

    The likes of the Copenhagen Metro, DLR etc. are not full railways with complex signalling systems like the DART or any other heavy rail line shared between local, intercity and freight traffic which would be the vast majority of railway systems.


  • Closed Accounts Posts: 1,288 ✭✭✭HonalD


    Stephen15 wrote: »
    The likes of the Copenhagen Metro, DLR etc. are not full railways with complex signalling systems like the DART or any other heavy rail line shared between local, intercity and freight traffic which would be the vast majority of railway systems.

    Someone can correct me if I’m wrong but heavy rail is not a candidate for self driving. Metros around the world are autonomous and are primarily “closed systems”


  • Posts: 0 [Deleted User]


    HonalD wrote: »
    Someone can correct me if I’m wrong but heavy rail is not a candidate for self driving. Metros around the world are autonomous and are primarily “closed systems”
    In theory, Rail should be one of the lower hanging fruit as far as automation is concerned.



    After all the vehicle is running on a segregated route, the route is externally pre-programmed by a signalman, the telemetry can be relayed to the vehicle by a whole array of means (RF, optical & electromagnetic).
    Unlike an autonomous road vehicle that has to rely on a considerable number of sensors, cameras to try to anticipate the movements of other vehicles & people/animals that share the roads.



    And to top it all, the vehicle is on tracks! What is preventing it from being automated?


  • Registered Users Posts: 12,363 ✭✭✭✭Del.Monte


    Is there a point to rail automation or is it just another fetish with perceived modernity - like the Bertie 'laughing stock of Europe' e-voting machines.


  • Posts: 0 [Deleted User]


    Del.Monte wrote: »
    Is there a point to rail automation or is it just another fetish with perceived modernity - like the Bertie 'laughing stock of Europe' e-voting machines.
    I suppose you could ask that of all vehicles, is there really a point or is it just to provide services without the inconvenience of humans being in the way!


    No pay and conditions to worry about for starters...


    And don't care what happens to those displaced by such automation either.


    The people who pull the strings are in a different world.


  • Closed Accounts Posts: 1,452 ✭✭✭Twenty Grand


    HonalD wrote: »
    Someone can correct me if I’m wrong but heavy rail is not a candidate for self driving. Metros around the world are autonomous and are primarily “closed systems”

    Heavy rail, especially in remote areas is probably going to be the poster child for train automation

    The Australians are way ahead.
    https://www.ft.com/content/b71db1fa-ed3d-11e8-89c8-d36339d835c0


  • Posts: 0 [Deleted User]


    Heavy rail, especially in remote areas is probably going to be the poster child for train automation

    The Australians are way ahead.
    https://www.ft.com/content/b71db1fa-ed3d-11e8-89c8-d36339d835c0
    Or without paywall
    https://www.bloomberg.com/news/articles/2018-07-13/robot-trains-are-slashing-mining-costs-in-australian-outback


  • Closed Accounts Posts: 1,452 ✭✭✭Twenty Grand




  • Posts: 0 [Deleted User]


    Thanks! Didn't have a paywall for me :confused:
    Maybe it doesn't like my ad blocker!


  • Advertisement
  • Closed Accounts Posts: 2,891 ✭✭✭prinzeugen


    In theory, Rail should be one of the lower hanging fruit as far as automation is concerned.



    After all the vehicle is running on a segregated route, the route is externally pre-programmed by a signalman, the telemetry can be relayed to the vehicle by a whole array of means (RF, optical & electromagnetic).
    Unlike an autonomous road vehicle that has to rely on a considerable number of sensors, cameras to try to anticipate the movements of other vehicles & people/animals that share the roads.



    And to top it all, the vehicle is on tracks! What is preventing it from being automated?

    Rail came up before. There was an accident recently in the UK (Watford tunnel landslide) that proved there will always need to be a human present on heavy rail.

    The outcome of that would have been different if it was not for the quick and non standard thinking of one of the drivers.

    Another case was from a driver doing 100mph on the East Coast route. Spotted a cow on the line ahead. AI would have hit the brakes. Him? He accelerated so when the train hit it, the cow would explode or been thrown clear and prevent a derailment. It worked.


  • Closed Accounts Posts: 2,891 ✭✭✭prinzeugen


    Heavy rail, especially in remote areas is probably going to be the poster child for train automation

    The Australians are way ahead.
    https://www.ft.com/content/b71db1fa-ed3d-11e8-89c8-d36339d835c0

    Wont work in the highlands of Scotland with a passenger train though. Especially in the winter.


  • Registered Users Posts: 3,522 ✭✭✭paleoperson


    It won't work anywhere. Even Steve Wozniak has said this:

    https://www.dailymail.co.uk/sciencetech/article-6220185/Apple-founder-Steve-Wozniak-says-doesnt-believe-AI-smart-drive-car.html

    For many years I've been saying it will never work with the almost child-like simplicity that some people seem to think it will, especially in bad weather or roads that aren't in immaculate condition. But even I didn't imagine that it would have collapsed and failed as hard as it has done.

    So much for "closer than most people think", more like "further away than anyone could have imagined". Think about this next time someone tells us far-fetched tales of what's down the line.

    So-called "driver AI" turns out to be nothing more than a glorified radar system where a car is programmed to avoid other cars, such a cod.


  • Closed Accounts Posts: 2,891 ✭✭✭prinzeugen


    It won't work anywhere. Even Steve Wozniak has said this:

    https://www.dailymail.co.uk/sciencetech/article-6220185/Apple-founder-Steve-Wozniak-says-doesnt-believe-AI-smart-drive-car.html

    For many years I've been saying it will never work with the almost child-like simplicity that some people seem to think it will, especially in bad weather or roads that aren't in immaculate condition. But even I didn't imagine that it would have collapsed and failed as hard as it has done.

    So much for "closer than most people think", more like "further away than anyone could have imagined". Think about this next time someone tells us far-fetched tales of what's down the line.

    So-called "driver AI" turns out to be nothing more than a glorified radar system where a car is programmed to avoid other cars, such a cod.

    I think there is a programme on BBC 2 or 4 in the next day or so that proves this.

    Tomorrow's World.. What really happened.

    So many companies claiming bull. They promised me a flying car. Still waiting!


  • Closed Accounts Posts: 1,452 ✭✭✭Twenty Grand


    prinzeugen wrote: »
    I think there is a programme on BBC 2 or 4 in the next day or so that proves this.

    Tomorrow's World.. What really happened.

    So many companies claiming bull. They promised me a flying car. Still waiting!
    Nobody wants a flying car, least of all one driven by people. Can hardly drive in a 2d environmwnt, be terrible in a 3d.

    Then you have the energy problem of keeping cars in the air. Very inefficient.

    Then you have the issue of flying above ppulated areas.

    Then you have the issues of maintenance. At least you can pull over if your car breaks down. What happens if you're flying?


    Autonomous cars will happen.

    It won't be quick. It'll be in ideal cities with ideal conditions, then it'll spread, slowly but surely.
    Snowy countries, or ones with crap infrastructure are decades away, but they will get there.


  • Registered Users Posts: 12,363 ✭✭✭✭Del.Monte


    ^^^^ Nothing like blind faith.


  • Registered Users Posts: 722 ✭✭✭Carol25


    I would love to see a self driving car in somewhere like bangalore in India.
    But i do think it will be possible eventually.
    Huge banks of data will need to be collected & fed into AI over time before it would be possible.


  • Moderators, Category Moderators, Arts Moderators, Sports Moderators Posts: 48,393 CMod ✭✭✭✭magicbastarder


    prinzeugen wrote: »
    Another case was from a driver doing 100mph on the East Coast route. Spotted a cow on the line ahead. AI would have hit the brakes. Him? He accelerated so when the train hit it, the cow would explode or been thrown clear and prevent a derailment. It worked.
    what a weird point.
    if we know speeding up can prevent a derailment, we program it in. you seem to think AI (or driving autonomy) exists frozen in time and that current challenges can't even be addressed, let alone solved.


  • Registered Users Posts: 1,747 ✭✭✭mdebets


    what a weird point.
    if we know speeding up can prevent a derailment, we program it in. you seem to think AI (or driving autonomy) exists frozen in time and that current challenges can't even be addressed, let alone solved.
    An Ai would be even better at it than the train driver, because it could calculate what would be the best solution, break or accelerate and could it do more precisely, also looking ahead on the line. What good would it have been, if the driver accelerated (more than he would have needed) and moved the cow out of his way, but was then derailed on the next bend, because it couldn't handle the higher speed and he couldn't break fast enough for it?


  • Advertisement
  • Registered Users Posts: 19,728 ✭✭✭✭cnocbui


    what a weird point.
    if we know speeding up can prevent a derailment, we program it in. you seem to think AI (or driving autonomy) exists frozen in time and that current challenges can't even be addressed, let alone solved.

    You can't program in changes to AI that are derived from neural networks as you don't know anything about the 'solution' neural networks come up with.

    It's called the black-box problem and it is of serious concern. If they deploy AI guidance in general and it turns out that they start inexplicably driving straight over people wearing yellow green and pink clothing on crosswalks, there is no way of finding out why and just tweaking the lines of code because there aren't any in human legible form. You would have to try training the neural network to stop doing that and eventually issue a software update, but in the meantime, you would either have to just put up with the deaths, or incapacitate all vehicles with that AI until you had a fix.

    I am guessing that a train driving AI would be near trivial compared to one for cars so it probably would be coder written, so you are probably right.


Advertisement