3 ms·

The US trucking industry is ripe for disruption. The role of a driver is minimal as it is: s/he has to just steer the truck along the highway, with periodic fue

The US trucking industry is ripe for disruption. The role of a driver is minimal as it is: s/he has to just steer the truck along the highway, with periodic fuel stops. There's nothing preventing an automated system from doing the same. Self-driving cars may be years away, but self-driving trucks which automatically attach themselves to trailers, drive 2500 miles and drop them off, only to do the same in return, 24x7 ? That'll be huge.

mdorazio 11y ago
Keep in mind that the role of a modern truck driver covers far more than just hauling a trailer on the freeway. Drivers also have to navigate city/suburb streets to an actual end destination (usually a warehouse), work with the loading dock workers, sign paperwork, etc. Personally, I think we will see a system where trucks like this do all the long-distance driving and stop at a drop-off location just outside cities. Then the role of truck drivers changes to that of delivering cargo the last mile to the final destination (the part that's much harder to automate). That, too, will eventually be automated, but not for a while.
- a_imho 11y ago
  You don't need to transport the driver for that, you can have workers residing in the cities to finish the work. But you still have to pay them, so you gain nothing at all. We already have quite a few more or less useful driving aids, I just don't get the obsession of making it fully autonomous. Legislations are lacking behind tech, ethics problems won't go away either, autonomous trucks could be prime target for hacking etc.
  - mdorazio 11y ago
    I don't understand how it doesn't save money. If the human drivers are only driving the last few miles and working with loading operations, they can handle many more trucks in a day than they can currently, so the number of drivers needed drops quite a bit. Salaries for those drivers might go up, but not enough to compensate, so the trucking companies stand to benefit. Let's say a popular shipping route is 300 miles from dock to warehouse, of which 280 are "easy" open road perfect for automating. For the sake of example, let's also say it takes half an hour at the loading/unloading point on either side, the 20 "city miles" take an hour to drive, and the middle 280 miles take 4.5 hours to drive. The total time is then 6.5 hours for a single truck, all of which is currently human labor. Now let's say the truck drives itself the middle 280 miles, and the human driver is handling the rest. Whereas before the human driver would have been able to handle a single truck in 6.5 hours, the end point operations now only require 2 hours of driver time per truck, so we've cut the need for 2/3 of our human driver hours because one driver could handle 3 trucks in the same time as before. With good scheduling and tracking, that means we've also cut the need for 2/3 of our human drivers. Obviously it's more in favor of human drivers for shorter hauls, and more in favor of self-driving trucks for longer hauls.
virmundi 11y ago
I kinda think this too. One thing that's particularly hard is driving through snow on a mountain. Even going through Tennessee is a challenge. I don't automation is there yet, and probably won't be for a while. Clean roads, where we know what's going on seem to be the way of automation right now. If you don't train your humans to drive in bad conditions and good, they won't help here either.

mdorazio 11y ago

Keep in mind that the role of a modern truck driver covers far more than just hauling a trailer on the freeway. Drivers also have to navigate city/suburb streets to an actual end destination (usually a warehouse), work with the loading dock workers, sign paperwork, etc. Personally, I think we will see a system where trucks like this do all the long-distance driving and stop at a drop-off location just outside cities. Then the role of truck drivers changes to that of delivering cargo the last mile to the final destination (the part that's much harder to automate). That, too, will eventually be automated, but not for a while.

a_imho 11y ago
You don't need to transport the driver for that, you can have workers residing in the cities to finish the work. But you still have to pay them, so you gain nothing at all. We already have quite a few more or less useful driving aids, I just don't get the obsession of making it fully autonomous. Legislations are lacking behind tech, ethics problems won't go away either, autonomous trucks could be prime target for hacking etc.
- mdorazio 11y ago
  I don't understand how it doesn't save money. If the human drivers are only driving the last few miles and working with loading operations, they can handle many more trucks in a day than they can currently, so the number of drivers needed drops quite a bit. Salaries for those drivers might go up, but not enough to compensate, so the trucking companies stand to benefit. Let's say a popular shipping route is 300 miles from dock to warehouse, of which 280 are "easy" open road perfect for automating. For the sake of example, let's also say it takes half an hour at the loading/unloading point on either side, the 20 "city miles" take an hour to drive, and the middle 280 miles take 4.5 hours to drive. The total time is then 6.5 hours for a single truck, all of which is currently human labor. Now let's say the truck drives itself the middle 280 miles, and the human driver is handling the rest. Whereas before the human driver would have been able to handle a single truck in 6.5 hours, the end point operations now only require 2 hours of driver time per truck, so we've cut the need for 2/3 of our human driver hours because one driver could handle 3 trucks in the same time as before. With good scheduling and tracking, that means we've also cut the need for 2/3 of our human drivers. Obviously it's more in favor of human drivers for shorter hauls, and more in favor of self-driving trucks for longer hauls.

virmundi 11y ago

I kinda think this too. One thing that's particularly hard is driving through snow on a mountain. Even going through Tennessee is a challenge. I don't automation is there yet, and probably won't be for a while. Clean roads, where we know what's going on seem to be the way of automation right now. If you don't train your humans to drive in bad conditions and good, they won't help here either.