August 07, 2015
Autonomous Operation Coming For Trucks First?

Full autonomous operation of commercial vehicles is seen as still over a decade away. But commercial fleet operators are rolling out lots of driver-assist technologies that take over a rising portion of driving tasks.

the move to autonomy in commercial and industrial vehicles is far ahead of the autonomous systems offered on most passenger vehicles.

The automated assist technologies are cutting accident rates and improving fuel efficiency. Full autonomous driving will also slash labor costs. This might well happen first with long haul trucks because highways are an easier autonomous driving challenge than local streets. Also, there is a way to cut labor costs without complete autonomy: let one human truck driver play the roll of leader for a bunch of unmanned trucks. The lead truck could also have full autonomous tech. But the human could be ready to intervene when something up ahead is beyond what the computers can handle.

Long haul highway trucking seems like the most likely first candidate for autonomous trucking. Humans can bring the truck to the on-ramp and humans can service the truck at successive truckstops along its way. Then near the end of the trip the truck can pull over at a new style of truck stop to allow a human to get in and drive the trickier pieces of road to a warehouse.

Share |      Randall Parker, 2015 August 07 07:32 PM 


Comments
Engineer-Poet said at August 8, 2015 1:15 PM:

That's a half-solution even for the highway segments.  It would be far better to lay down rail in the medians, use dual-mode trucks, supply power either through an overhead wire or a segmented and switched flush third rail, and eliminate both the drivers and most of the liquid fuel consumption.

This also provides a way to shift electricity consumption.  The medium-haul trips are done by autonomous trucks at night; they leave from a depot and travel to another depot mostly under electric power.  During the workday drivers take the trucks from depot to destination and return to the depot, so they're home every night.  Long haul still goes depot to depot but runs 24/7 until it arrives.

bob sykes said at August 9, 2015 5:02 AM:

Go over to antiplanner (http://ti.org/antiplanner/) for a critique of trains vs. trucks and busses. Trains are good for long-haul, depot-to-depot operations, but the complete transport from maker to user requires trucks. The capital investment needed for trains is so large that long-haul trucking, with its flexible routes, usually is economically preferred, which is why there are so many long-haul trucks in service.

On a lighter note, John Barnes "Directive 51" contains a description of what happens to a 100 +/- truck "train" when an EMP hits.

Engineer-Poet said at August 9, 2015 12:00 PM:

Trucks can run on rails just as easily as trains can.  Rails take care of 95% of the guidance problem and also lend themselves to electrification.

Fifty Ville said at August 14, 2015 11:56 AM:

My prediction? Teamsters beating truck programmer geeks into pulp.

_Jim said at August 15, 2015 11:25 AM:

bob sykes, fortunately, EMP is over-rated by most.

For instance, see this paper by Mario Rabinowitz titled:

"Effect of the FAST NUCLEAR ELECTROMAGNETIC PULSE on the
Electric Power Grid Nationwide: A Different View"

Mario is with Electric Power Research Institute.

A copy can be seen here: http://arxiv.org/ftp/physics/papers/0307/0307127.pdf

bobby b said at August 15, 2015 2:01 PM:

Here's the main problem:

http://www.npr.org/sections/money/2015/02/05/382664837/map-the-most-common-job-in-every-state

Check out in how many states in the USA "truck driving" is the number one most common occupation.

It shocked me, frankly.

I can see all kinds of problems with a push to automate trucks.

Larry said at August 16, 2015 11:42 AM:

The truck thing is going to happen. Cost and safety considerations demand it. Drivers have no power. Big firms will control even more of the industry. Their scale will permit them to learn faster than smaller firms and will drive the technology forward very rapidly. I expect it to have the typical tech uptake curve. It will not happen much for awhile and then boom, mass adoption. I get a new car every 5-8 years. I'm about to get a new one. That one won't drive itself (much). The next one probably won't have a steering wheel. Can't wait.

Randall Parker said at August 16, 2015 2:19 PM:

bobby b,

Yes, great link. In most states the most common job is truck driver. Just as that chart shows secretary ceasing to be a common job (since computers automate so much office work) so the same will happen with truck driver. In 20 years time that chart will show few or no states with truck driver as the most common job. Ditto for farmer. Robots will plow the fields and harvest the crops.

Larry,

Yes, the big fleets will jump on each successive step of autonomous truck tech as soon as it become available.

E-P,

Autonomous electric trucks could go into robotic battery-swapping truck stops every few hours and get a battery swap done and hit the road again. Its a trade-off between the cost of extra batteries sitting in truck stops vs time trucks spent stopped and recharging.

But if oil stays below $50 a barrel that isn't going to happen. Perhaps in a few years oil prices will shoot back up again to $100 per barrel and in the mean time battery costs will drop.

Engineer-Poet said at August 16, 2015 3:08 PM:

Diesel fuel is about $2.55/gallon at the corner near me.  At 140,000 BTU/gal and 40% efficiency, that's about 15.5¢/kWh delivered at the crankshaft.  Batteries cannot compete (amortization alone costs that much), but wholesale grid power can cost half that or less especially if it's off-peak.  Power through overhead wires or third rails is competitive even today.  OTOH, small batteries to provide a few miles of range from a rail-lane of freeway to a depot and back (recharging in motion while on the electrified rail system) would not cost all that much and could reasonably be required by localities interested in controlling noise and air emissions.

What I'd like to see is a system that is safer, quieter, cleaner and allows things like semi-automated electric RV transport over the network.  Automated rail-changing was demonstrated with Bladerunner prototypes quite a few years ago, which would already have given us self-driving trucks had we begun the conversion back then.  There's no time like the present.

Post a comment
Comments:
Name (not anon or anonymous):
Email Address:
URL:
Remember info?

                       
Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright