Slashdot Mirror


Uber's Self-Driving Trucks Division Is Dead (techcrunch.com)

Uber is shuttering its self-driving unit, reports TechCrunch. The company will reportedly stop development of self-driving trucks and instead focus its efforts on self-driving cars. "We recently took the important step of returning to public roads in Pittsburgh, and as we look to continue that momentum, we believe having our entire team's energy and expertise focused on this effort is the best path forward," Eric Meyhofer, head of Uber Advanced Technologies Group, said in an emailed statement. From the report: Uber Freight, a business unit that helps truck drivers connect with shipping companies, is unaffected by this decision. "Rather than having two groups working side by side, focused on different vehicle platforms, I want us instead collaborating as one team, according to an email reviewed by TechCrunch that was sent by Meyhofer to employees. "I know we're all super proud of what the Trucks team has accomplished, and we continue to see the incredible promise of self-driving technology applied to moving freight across the country. But we believe delivering on self-driving for passenger applications first, and then bringing it to freight applications down the line, is the best path forward. For now, we need the focus of one team, with one clear objective." The company will pivot employees focused on self-driving trucks to other work that revolves around self-driving technology.

35 of 61 comments (clear)

  1. Wrong direction by Archfeld · · Score: 1

    I have no citations to present but I think they are going about it backwards. Concentrate on the long haul trucks and the shipping industry, then leverage that know how and experience into self driving cars. The behaviors and needs of big trucks seems more predictable, the trucking industry seem more receptive and more likely to finance and adopt the technology. Then after having success there they can spread to other areas.

    --
    errr....umm...*whooosh* *whoosh* Is this thing on ?
    1. Re:Wrong direction by HornWumpus · · Score: 1

      Sure automating the long haul part is about the easiest of the tasks. Especially if you have the software build virtual truck trains, with an awake driver in the lead vehicle.

      But their are laws in place and economic interests to protect. It will take 20 years to make it legal for a trucker to rest while his truck is rolling as part of a train. Until they do, it's a waste of effort.

      --
      John McAfee 'It was like that time I hired that Bangkok prostitute; to do my taxes, while I fucked my accountant'
    2. Re:Wrong direction by PPH · · Score: 1

      the trucking industry seem more receptive

      Teamsters.

      The soccer mom's union doesn't have nearly as much clout.

      --
      Have gnu, will travel.
    3. Re:Wrong direction by HornWumpus · · Score: 1

      Get rid of the second driver and the teamsters will lose close to half their members/power.

      Yes I know, many trucks go slower these days and only have one driver, but you get the point. In that case, you get rid of half the trucks too.

      --
      John McAfee 'It was like that time I hired that Bangkok prostitute; to do my taxes, while I fucked my accountant'
    4. Re:Wrong direction by seven+of+five · · Score: 1

      Oh, Teamsters will LOVE autonomous trucks. It's not like they're going t take a pay cut.

      That's right! Trucking companies will be happy to pay a premium for self-driving trucks AND pay the Teamster's driver the same money they received when they were behind the wheel.

      Uh, no. Push will eventually come to shove.

    5. Re:Wrong direction by HornWumpus · · Score: 1

      What % of warehouse workers are teamsters?

      Decades ago, when I worked a summer doing appliance deliveries, it was one of six of the warehouses we got product from. God they sucked big wet donkey balls. I'm sure it's even lower now, they were putting their employer out of business back then.

      --
      John McAfee 'It was like that time I hired that Bangkok prostitute; to do my taxes, while I fucked my accountant'
    6. Re:Wrong direction by aaarrrgggh · · Score: 1

      Yeah... but if they did that, they would have less time to stall. By focusing on something a decade out, think of how many more quarters they can obfuscate their true performance...

    7. Re:Wrong direction by rtb61 · · Score: 1

      Pfft, it was all a pump and dump. Create the illusion of economic sector dominance, do the IPO, sell and run before it all explodes.

      Vehicle manufacturers who dominate the vehicle leasing market, will similarly dominate the automated vehicle market, for much the same reason.

      --
      Chaos - everything, everywhere, everywhen
  2. Teamsters or Driver's unions? by magarity · · Score: 1

    One would think automated delivery truck would be much easier; they go between some loading dock and some other loading dock, not random street corners. And deliveries are made in the early hours of the morning when there's much less traffic.

    1. Re:Teamsters or Driver's unions? by 110010001000 · · Score: 1, Insightful

      Anyone who really knows anything wouldn't think that autonomous driving is possible at all with todays (or any reasonably near future) technology. Autonomous driving falls into the "80% is easy" part. The last 20% is the tough part, and might never be solved. People are way too optimistic about automated driving and technology in general.

    2. Re:Teamsters or Driver's unions? by HornWumpus · · Score: 3

      Until that time, it remains vapor.

      Also good luck with your 'better driver than a teenager' advertising campaign. I don't think you'll sell many.

      --
      John McAfee 'It was like that time I hired that Bangkok prostitute; to do my taxes, while I fucked my accountant'
    3. Re:Teamsters or Driver's unions? by 110010001000 · · Score: 2, Insightful

      Yes you do. And that might not be possible to do. It isn't there yet, and it might never be. The next 20% needs to be solved. You guys are way too optimistic about this stuff.

    4. Re:Teamsters or Driver's unions? by 110010001000 · · Score: 1

      Right. Because one thing is possible, all things are possible. I can't wait for my unicorn made out of gold (on Mars).

    5. Re:Teamsters or Driver's unions? by HornWumpus · · Score: 1

      Put your money where your mouth is chump.

      --
      John McAfee 'It was like that time I hired that Bangkok prostitute; to do my taxes, while I fucked my accountant'
    6. Re:Teamsters or Driver's unions? by slack_justyb · · Score: 1

      I don't think you'll sell many.

      I don't think you understand how companies work. If saving 10% here and 8% there adds up, it adds up. If automating trucks reduces lawsuits for pain and injury in a wreck by xx% and the cost of rolling out that automation is less than that xx%, then only stupid companies wouldn't roll it out. So the massive question is, "Will the automation ever drop in price?" And the answer is undoubtedly yes, it will. If there is one thing about technology is that the cost of implementing it always goes down.

      Also good luck with your 'better driver than a teenager' advertising campaign

      Yeah, I get that's not the best way parent could of put it, but the sentiment is still true. As soon as the technology is "good enough/cost effective enough" companies will start rolling it out. That has always been the case since the 1700s, and there is always nay sayers that go "nuh-uh", and those nay sayers are always wrong. This isn't a "oh well your opinion is incorrect and this is why I think that" kind of thing, this is a "there is literally three centuries of history that strongly points to XYZ being an outcome of this." Now that's not a 100% sure fire thing, and I'll admit that, but I think it's a pretty safe bet that ultimately truck drivers are going to be "mostly" gone within the next four to five decades. And I'm just going to point to the last 300 years of "if there is some device/machine that's cheaper and slightly better than man at doing something, that device/machine always wins."

      There is significant economic investment in this kind of thing and while Uber leaving might cast some doubts into the continued economic investment, in the trucking company industry. Uber wasn't even a bit player. They were a non-existent "oh hey let's try this out" player. Their departure should be seen with about as much regard as one might think about blood loss from a single mosquito bite. There are other people doing this very thing of automating a truck while it is on the road, 3PL providers looking at getting easy on/off access to highways, auto makers developing tracks that mimic some of the most used corridors such as I80/I90/I40, and the list goes on of people who are investing hand over fist more than Uber ever had hoped to dream of injecting into the industry. And the investments aren't fruitless, there are test trucks that are driving down large stretches of Interstate 80 with only a person in the cab to observe. Yes it is easy to point out flaws in something that is so infantile and wonder "How on Earth could that ever take the place of a human driver?" And then one day, that job no longer exists. This has been the norm of how things "do" in this world since the Industrial Revolution and I find it humorous how everyone just thinks that somehow this is some exception, that driving is some sort of endeavour to which only humans can do, which is something that they thought of Chess and Go and sewing and paralegals and programming and so many other things and then machine started doing those things. I get it, it is difficult to accept new things and it is made that much harder for things that are not yet "in production". But there is motion, there is investment, there is research, and there is so many other things that point toward an outcome that will soon spell the end of this profession for a large majority. 100%, maybe not, but 80%, oh yeah you better believe it.

      And there are those who would point to something like an 80/20 rule for driving. "Yeah, we can automate the 80% but the 20% is the hard stuff." If an investment of $x solves at least 90% of that 20% that's difficult, then companies are going to do that investment. If Amazon thinks they'll be able to convince the government to paint the far right lane blue to indicate automation traffic only, then Amazon is going to make an investment to convince the government to get some blue paint ready. If a logistics company can purchase the land right off

    7. Re:Teamsters or Driver's unions? by slack_justyb · · Score: 1

      The next 20% needs to be solved.

      I'm honestly curious, what's this mythical 20% that needs to be solved?

    8. Re:Teamsters or Driver's unions? by 110010001000 · · Score: 1

      It is the part where the car does, you know, autonomous driving. You know, the kind where it doesn't require a human driver. Didn't you ever wonder why these "autonomous" cars always have at least TWO drivers in them? I guess everyone misses that part because, you know, progress is inevitable. Keep spending those billions though, autonomous driving is right around the corner!

    9. Re:Teamsters or Driver's unions? by HornWumpus · · Score: 1

      We only see the successes. For every 'airplane' there 100 'uranium water infusers', 'anti fapping pajamas/cereals', 'porn blockers', 'electric boner belts' and female paroxysm inducing vibrating devices (that require a medical licence to operate). Engineering success is _not_ inevitable, no matter how much someone wants porn filters to work, they don't. Unpost.exe and rStabInEye.exe are not 'just around the corner'.

      Level 5 automated driving requires strong AI. We don't even know where to start in the task of building strong AI.

      The state of the art is currently more like the 'boner belt' than the wright flyer.

      --
      John McAfee 'It was like that time I hired that Bangkok prostitute; to do my taxes, while I fucked my accountant'
    10. Re:Teamsters or Driver's unions? by Namarrgon · · Score: 2

      Autonomous vehicles have rounded the corner and already arrived. Right now they're doing, you know, autonomous driving - the kind that doesn't require a human driver.

      These are merely Class 4 AVs, only capable of driving in a limited area under specific conditions, so still a long way from 100%. But those "specific conditions" include full city driving among pedestrians, buses, trucks, pets, distracted drivers, roadworks.. clearly an already-useful percentage of human capability.

      --
      Why would anyone engrave "Elbereth"?
    11. Re:Teamsters or Driver's unions? by Green+Mountain+Bot · · Score: 1

      No, you have to make a car that buyers will perceive to be a better driver than *themselves*. People aren't buying the car for the idiots they see on the road - they're buying it for themselves. Given that most people think they are far better drivers than the examples that you give, that's a much higher bar to clear than you present.

    12. Re:Teamsters or Driver's unions? by Namarrgon · · Score: 1

      That's a good point; I agree that many will feel they can drive better than the car can. Yet I'd bet most people consider themselves better than their last taxi driver too, and that doesn't stop them taking taxis. You watch, they'll still be willing to let the car drive for them on an increasing number of occasions - commuting in traffic, after a couple of drinks, when they'd rather play that new mobile game etc - and trust will build.

      --
      Why would anyone engrave "Elbereth"?
    13. Re:Teamsters or Driver's unions? by HornWumpus · · Score: 1

      Your realize that American freight rail is _better_ than Europe's? By any metric you choose.

      A rail spur still costs millions more than a loading dock, if it's possible at all. I trust businesses to make those decisions for themselves.

      --
      John McAfee 'It was like that time I hired that Bangkok prostitute; to do my taxes, while I fucked my accountant'
  3. They're saving money by viperidaenz · · Score: 4, Informative

    They know they'll never develop viable self-driving technology.
    They need to keep pretending they can, as it's the end-goal to turn a profit on their main business, ride-sharing.
    They'll lose investors if they give up on that. The company is still running at a loss, without constant investment it's going to collapse.

    1. Re:They're saving money by 110010001000 · · Score: 1, Insightful

      That is the same trick that Tesla has been using. Eventually investors figure it out, the stock falls, then the lawsuits start.

    2. Re:They're saving money by Comrade+Ogilvy · · Score: 1

      Maybe. But Uber has no choice but to be in this game. If someone else arrives with reliable automated passenger car technology and Uber is far behind, Uber stock price can go to zero overnight.

      The vision of the future is that you can order your ride with a convenient app, which could be Uber or Lyft or Google or Apple or Walmart -- nobody cares who is the provider when the service is a cheap commodity.

      Even if that vision is a looooong way off, Uber cannot look like it will allow itself to be written out of the narrative, for want of a lousy few billion dollars. Cheaper to throw a bit of money down the toilet than let fear slash the Uber valuation in half.

  4. It's not dead... by Gravis+Zero · · Score: 1

    ...it has simply hit a wall. *rimshot*

    I'll see myself out. ;)

    --
    Anons need not reply. Questions end with a question mark.
  5. dog whistle : pivot == lay off by Anonymous Coward · · Score: 1

    Translation: "the path to making this stuff work is a lot harder than we thought. So the path to profits is a waaaaays out. So, we're going to "combine" these units, and expect us to dribble out layoffs as we 'realize synergies' "

  6. Why does the link go to a gold pimping site? by vik · · Score: 1

    The link in that article goes to goldseek.com, which has dick all to do with Uber trucks.

    What is going on here? Are that many people really ranting without clicking on the link, or has it been subsequently hacked?

    1. Re:Why does the link go to a gold pimping site? by 110010001000 · · Score: 1

      There is no reason to read any article on this topic. The headline says it all. This is totally expected as the interest in self-driving cars will eventually fade out. Just like it did in the 1970s and 80s. And the editors aren't paying attention at this point to whatever they post.

    2. Re:Why does the link go to a gold pimping site? by OzPeter · · Score: 1

      The link in that article goes to goldseek.com, which has dick all to do with Uber trucks.

      What is going on here? Are that many people really ranting without clicking on the link, or has it been subsequently hacked?

      The actual link *is* in the tech crunch name on the heading.

      But shame on the editors for the pimping. Not that they actually care what we think.

      --
      I am Slashdot. Are you Slashdot as well?
  7. That's enough by Namarrgon · · Score: 1

    I'm sure a lot of people would be just fine if merely 80% of their driving needs were automatable, or their truck drove itself 80% of the way while they napped.

    Very few humans can do 100% of all driving tasks too (our accident rate is proof of that). Self-driving cars may never be perfect, but they're already good enough today for Waymo's taxis to be carrying unaccompanied passengers in Phoenix as we speak.

    --
    Why would anyone engrave "Elbereth"?
  8. Re:Otto? by The+Cynical+Critic · · Score: 2

    It was probably to be expected seeing how all of Otto's tech was built on stuff the founder had pilfered from Google's self-driving car project after leaving it and when Google found out (IIRC trough a shared supplier accidentally cc:ing them some of their own documents) they got the courts to put a stop to using any of it.

    --
    "Why should I want to make anything up? Life's bad enough as it is without wanting to invent any more of it."
  9. Re:How many will have to die? by Gavagai80 · · Score: 1

    How many human lives will have to be lost before we give up on this 'self driving car/truck/taxi/whatever' nonsense?

    More than the 1.3 million people a year human drivers kill.

    --
    This space intentionally left blank
  10. Re:BSD Is D E A D by Type44Q · · Score: 1

    Lennart, you won (somehow, sort of...); now fuck off, eh? ;)

  11. Good by Huge_UID · · Score: 2

    That means fewer people will be.

    And before you rant about how autonomous cars and trucks will save lives because human drivers suck, I agree. I just want competent companies building them, not Uber.