Is my 2020 date for fully autonomous self-driving trucks on the highways optimistic enough?
Every month, another test comes out, each more successful than the last. The latest Daimler tests showcased its new capabilities with a platoon of three “Pilot Connect-Equipped” Mercedes trucks on the German Autobahn.
Fuel efficiency improved 2% for the lead truck, 11% for the middle truck and 9% for the third in a platoon test.
Highway Wi-Fi Connect Enables Truck Platooning
Please consider Daimler Outlines Vision for the Future of Vehicle Connectivity.
“There is a revolution going on, and the truck is in its center,” said Wolfgang Bernhard, member of the Board of Management for Daimler AG, Daimler Trucks & Buses. “We have a vision of the future where the entire transportation process is completely seamless. Where the flow of goods on the road is mirrored by a flow of information from the internet in real time.”
Bernhard envisions a supply chain with completely digital information exchange between driver and shipper, a Formula 1-like service environment with parts and technicians on-hand to make repairs as the truck arrives, automatic rest stop reservations and routing when the vehicle senses the driver is fatigued, and exchange of information between vehicles and infrastructure.
Highway Pilot Connect enables truck platooning
As an example of increased connectivity, Daimler debuted its Highway Pilot Connect, the next step in it quest for autonomous truck operation. In an on-highway demonstration on the Autobahn on the outskirts of town, Daimler showcased its new capabilities with a platoon of three Highway Pilot Connect-equipped Mercedes Actros trucks.
Highway Pilot Connect builds on Daimler’s Highway Pilot autonomous truck platform it introduced last year but adds Wi-Fi communication for vehicle-to-vehicle (V2V) communication, allowing all three vehicles to automatically adapt to changing traffic and road conditions.
In the demonstration, the following distance is reduced from 50 meters to 15 meters when the three trucks link into a single platoon, which Daimler says can decrease emissions and improve the combined fuel efficiency of the three vehicles by as much as 7 percent (2 percent improvement for the lead truck, 11 percent for the second truck and 9 percent for the third truck).
When a Mercedes passenger car entered the gap between the second and third truck, the Highway Pilot Connect automatically increased the following distance of the third truck to 50 meters until the car exited the lane, after which the system automatically closed the gap back to 15 meters.
Platooning Solves Problems
That last paragraph above is amusing because many people told me platooning would not work because cars could not get by.
My experience is the opposite. I am continually frustrated by trucks passing each other, one going 2 miles-per-hour faster than the one ahead of it, blocking up traffic for miles.
“Highway Pilot Connect is a perfect example to illustrate the possibilities in V2V communication,” said Sven Ennerst, head of product engineering & global procurement, Daimler Trucks. “Compared to Highway Pilot, Highway Pilot Connect unveils further fuel savings potential and reduces CO2 emissions and takes the burden off the shoulders of drivers on monotonous routes and will increase transport efficiencies in areas with low traffic density.”
Once again, there are backup drivers in these trucks. But those drivers will soon be unneeded.
Last Mile Concerns
Last-mile concerns and some loose ends on regulations are only thing holding up long-haul automation. Local drivers might be needed for the last mile.
I still envision large truck stop style facilities at major highway intersection points, near but well outside major city limits.
The intersection of I-80 and I-55 is the perfect spot for a truck service center hub.
Google Maps View of I55 I80 intersection
Google Earth Shot of I55 I-80 Intersection
There is plenty of vacant space and farmland near the intersection of I55 and I80 to support a major truck hub should a last-mile or service-center hub solution be needed. Clearly more are needed, but there are ample spots in the Western and Southern states.
Farmland prices in key areas just might skyrocket.
Mike “Mish” Shedlock
“WASHINGTON – You could soon be sharing the sidewalk with robots in D.C. A test project will be presented to District leaders to allow delivery robots to operate on city sidewalks.
These robots could be delivering your takeout meal or maybe some fresh groceries. The company that created them say they are safe, smart and conserve energy.”
http://www.fox5dc.com/news/local-news/111109311-story
overlay the rail map over the interstate map before you peg the primary focal point mike and make sure the primary ideal spot t connects to the waterway that leeds to the MISS and the lake
There is no “final” spot
there will be hundreds of them
Pingback: Robotics In the Labour Market | thePOOG
http://www.bloomberg.com/news/articles/2016-02-25/why-mercedes-is-halting-robots-reign-on-the-production-line
But also, notice what just happened in Brussels today. Wifi? lets see how many we can jack-knife with a little jamming, or worse take it over.
Illinois is the last place anyone would want to open any business, so I-80 – I-55 would be a bad idea for that reason alone. If there is one, it will be in California where the Tech companies are and where weather isn’t an issue.
If Hubs needed, there will be at least one in every state
Mish
Oh, did I miss the post on the driverless google car running into a bus a few weeks ago??
Apparently not
You just forgot to think how meaningless it was.
Hummmm,
https://nakedsecurity.sophos.com/2016/03/21/warning-issued-by-fbi-over-dangers-of-car-hacking/
http://www.pymnts.com/news/security-and-risk/2016/fbi-warns-of-increased-vehicle-hacking-vulnerability/
Daimler says …
I bet they do … would love to see independent 3rd party conduct the tests.
The rigged anti missile – missile tests dancing in my mind.
Google is now saying it may take 30 years to develop commercial versions of their autonomous cars that can drive in all weather conditions and locations:
“…Not only might it take much longer to arrive than the company has ever indicated—as long as 30 years, said Urmson—but the early commercial versions might well be limited to certain geographies and weather conditions. Self-driving cars are much easier to engineer for sunny weather and wide-open roads, and Urmson suggested the cars might be sold for those markets first…”
http://spectrum.ieee.org/cars-that-think/transportation/self-driving/google-selfdriving-car-will-be-ready-soon-for-some-in-decades-for-others
please be serious
Drivers worst in snow, ice, bad conditions
Automation will be much better
non-tech people think machines dont make mistakes, machine dont make mistake when all the rules are clearly defined and there are no unknown variables. Machine breaks down when when unexpected situation occurs.
@meofios
Another issue is tho mode of failure. Biological systems tend to fail more gradually and gracefully. Rather than either working 100%, or absolutely not at all. And, the kind of heuristics and algorithms that enable biology style graceful failure, also brings with it biology style imprecision during normal operation.
Which wouldn’t necessarily be so bad, as long as the end result still trumped a representative sample of humans. But, both the social and legal tolerance for machine failure is much lower than for living things, and the honeypot available to the lawyers are almost always much bigger. Think; the difference between your dog biting a kid, versus a Google made robot pet doing the same thing……
You’re in thrall to technology.
I trust you would disagree with J.H. Kunstler…but it’s generally true nonetheless that the more complex, the larger the scale and the more inflexible things are the greater the systemic risk. Earthquakes? Rabbits? A human in a traffic accident? A breakdown mid-highway? Who knows? They’re akin to Black Swans.
Absent massive industry lobbying for control of the government gun this likely will never fly, and even if it does how is it going to raise the purchasing power and improve the living standards of the now less likely to be employed masses?
These are not power looms or cotton jennies: times have changed in that real-terms economic growth, not measured in dollars, has now stopped.
Another spoon feeding coming up.
Productivity is bound to rise. Precisely the result of taking 10 manhours of work instead of 50 to get from point a to point b.
Costs will drop (or at least should barring central bank intervention)
More goods, cheaper prices, in less time – the essence of rising standards of living
Once the driver is gone speed is no longer the key cost driver. Put extra hard treadless wheels on to reduce rolling friction. Drop speed to thirty mph to reduce wind drag and still achieve adequate braking and handling with treadless hard wheels. I see such robo trucks lumbering across country consuming one quarter the fuel while arriving in the same time window because the computer never eats or sleeps. 75% energy savings is just for starters.
they will need to go the minimum
Typically 45 MPH
45 mph times 24 hours = 1,000 miles per day
likely 1/2 the fuel consumption
1/3 wear and tear on machine
1/5 wear and tear on road
you certainly correct the first application will be freight running on interstates w/o humans, then once exit interstate, will be human driven for last ‘mile’
2020 will likely see beta testing
2025 significant sales
2030 US – EU market share of new sales ~80%
Wind drag loss is power law in speed typically quadratic. You cross over to rolling friction being dominant as speed falls but you get the idea. The true autonomous vehicle revolution will occur not when the first robo trans arrive (which will be very soon) but when the last human driver leaves. This will happen one controlled region at a time. Region by region the rules will change to suit the special needs/opportunities of the robo’s. It will be a race to early adopt because the economic advantage to early adopters will be huge.
This sounds a like a railroad?