Posted on 04/27/2016 12:26:09 PM PDT by JOAT
A convoy of self-driving trucks recently drove across Europe and arrived at the Port of Rotterdam. No technology will automate away more jobs — or drive more economic efficiency — than the driverless truck.
Shipping a full truckload from L.A. to New York costs around $4,500 today, with labor representing 75 percent of that cost. But those labor savings aren’t the only gains to be had from the adoption of driverless trucks.
Where drivers are restricted by law from driving more than 11 hours per day without taking an 8-hour break, a driverless truck can drive nearly 24 hours per day. That means the technology would effectively double the output of the U.S. transportation network at 25 percent of the cost.
And the savings become even more significant when you account for fuel efficiency gains. The optimal cruising speed from a fuel efficiency standpoint is around 45 miles per hour, whereas truckers who are paid by the mile drive much faster. Further fuel efficiencies will be had as the self-driving fleets adopt platooning technologies, like those from Peloton Technology, allowing trucks to draft behind one another in highway trains.
Trucking represents a considerable portion of the cost of all the goods we buy, so consumers everywhere will experience this change as lower prices and higher standards of living.
While the efficiency gains are too real to pass up, the technology will have tremendous adverse effects as well.
In addition, once the technology is mature enough to be rolled out commercially, we will also enjoy considerable safety benefits. This year alone more people will be killed in traffic accidents involving trucks than in all domestic airline crashes in the last 45 years combined. At the same time, more truck drivers were killed on the job, 835, than workers in any other occupation in the U.S.
Even putting aside the direct safety risks, truck driving is a grueling job that young people don’t really want to do. The average age of a commercial driver is 55 (and rising every year), with projected driver shortages that will create yet more incentive to adopt driverless technology in the years to come.
While the efficiency gains are real — too real to pass up — the technology will have tremendous adverse effects as well. There are currently more than 1.6 million Americans working as truck drivers, making it the most common job in 29 states.
The loss of jobs representing 1 percent of the U.S. workforce will be a devastating blow to the economy. And the adverse consequences won’t end there. Gas stations, highway diners, rest stops, motels and other businesses catering to drivers will struggle to survive without them.
The demonstration in Europe shows that driverless trucking is right around the corner. The primary remaining barriers are regulatory. We still need to create on- and off-ramps so human drivers can bring trucks to the freeways where highway autopilot can take over. We may also need dedicated lanes as slow-moving driverless trucks could be a hazard for drivers. These are big projects that can only be done with the active support of government. However, regulators will be understandably reluctant to allow technology with the potential to eliminate so many jobs.
Yet the benefits from adopting it will be so huge that we can’t simply outlaw it. A 400 percent price-performance improvement in ground transportation networks will represent an incredible boost to human well-being. Where would we be if we had banned mechanized agriculture on the grounds that most Americans worked in farming when tractors and harvesters were introduced in the early 20th century?
We often discuss the displacement of jobs by artificial intelligence and robots in the abstract, as something that we’ll have to eventually tackle in the far distant future. But the recent successful demonstration of the self-driving truck shows that we can’t afford to put off the conversation on how we’re going to adapt to this new reality.
Only 300...
Only?
300 incidents in a million miles is actually 160 times the accident rate for a standard automobile.
And, since Google cars don’t make an attempt in bad weather like snow or heavy rain, the numbers are skewed even worse.
This put the Google experiment squarely in the ‘test’ stage, and nowhere near ready for consumer use.
It’s a new technology. It’s isn’t supposed to be perfect yet. But not being perfect yet is a far cry from “never going to happen”. It’s like saying we’ll never get to the moon because Apollo 13 had problems.
Google is still squarely in the test stage. Meanwhile, Tesla, Mercedes, BMW have shipped.
I don’t know what to tell you...Tesla et al have most definitely NOT shipped a completely self driving car. Period.
Google is tricking you to some extent, with their ‘million miles’. Here’s how:
For decades, experiments have been attempted at what would be considered a ‘guided’ car. Usually the experiments would involve a series of sensors along a road, or a wire beneath. For a variety of reasons, this won’t work large scale...and what Google purports to be working on is different: self driving, unguided, autonomous cars.
And that’s what their press release says. And that’s what you believe...and repeat...’a million miles driven’.
Here’s the Paul Harvey part: It is STILL guided, and not really autonomous at all. What have they replaced the wire under the road with? Well, prior to the Google car making a trip, a fleet of Google Street cars goes through the route, making multiple 3D scans of the route, and correlating them together. Every driveway, lamp post and stop light is scanned, filtered through elaborate algorithms to create ‘shapes’ out of point clouds, and developed into a map. Said map is then programmed into the Google car.
You see, the car really isn’t as autonomous as the press release. The map is just about as restricting as a railroad track is to a train. Without it, the car cannot proceed.
Now raise your hand if you have made 3D LIDAR scans, correlated them together, and used software to reduce the point clouds into recognized features....
...I had to quit typing for a minute there, because my hands were raised. I’ve used this type of equipment, and I know exactly what it takes (angles, overlap, max distance between control points) to make an accurate map. And, I know that the storage and processing of the point clouds take racks of computers...and a lot of time...and ALOT of human review of the map. Probably thousands of hours for each new mile ‘mapped’.
Oh, and since the world constantly changes, areas have to be re-mapped. The only really effective way to manage this is a national registration system for the maps, and requirements that municipalities register changes (new stop sign, construction zone, etc) ahead of time, so a Google Street car can go re-map the new feature. This exact requirement has been placed on the nation’s major railroads already (under a mandate called Positive Train Control)...and they have successfully lobbied against enforcement for 4 years now, citing cost (and they already have fixed control in place). Now, do you really think that within 5 years we will have you fully autonomous luxury automobile on the market...which would mean:
1) A majority of states will have passed laws allowing fully autonomous vehicles
2) Google will have expanded its mapping from a current 0.5% of the nation’s roads to at least 50%
3) A national map registration system will be in place - affecting every township, county, and city in the nation
Of course not.
Beyond 5 years? I still say never. The self driving safety features on the Tesla (which you repeatedly confuse for an autonomous vehicle) will become widespread, and will be useful/improve safety....and generally make commutes easier. Given these improvements, and the large technical obstacles to an autonomous vehicle, I just don’t see consumers needing or wanting the latter. Absent government intervention, it just won’t happen.
Oh...going to the moon was orders of magnitude simpler. And you may recall that final adjustments at re-entry were actually manually controlled :)
Google stuff might have started that way. But it isn’t anymore. And Tesla folks aren’t doing that at all, yes they’re primarily leveraging the collision avoidance system, but they’re on the fly. It’s interesting because they’re basically taking opposite approaches, Google is going from scratch and trying to go to the end, Tesla et al are saying “look we already have most of the pieces in places as feature, let’s incremental our way to it”. So yeah Tesla and company aren’t fully self driving, but they are to the point that you can just let go of the wheel for long periods of time and let the car do it, and it can handle most stuff. It’s not just the safety features it’s LEVERAGING the safety features into: push the button, don’t touch the wheel, the car drives itself and can handle most issues. That IS a self driving car. At worst right now what we have are self driving cars that aren’t very good drivers, they’re on a learners permit, but they’re driving.
Self driving cars is orders of magnitude harder than going to the moon. But we have orders of magnitude more computing power to throw at it. The modern smartphone has more computing power than NASA had. We are in a world where computing power is no longer a barrier to pretty much anything, we can throw a couple hundred GFLOPS into a car without impacting the cost of the car. It’s now just a matter of gathering the data from sensors and doing something with it. And we’re already doing it, it’s just a matter of how much. Declaring never is ignoring what is already happening.
“but they are to the point that you can just let go of the wheel for long periods of time and let the car do it, and it can handle most stuff...That IS a self driving car.”
From Tesla’s web site:
“Tesla requires drivers to remain engaged and aware when Autosteer is enabled. Drivers must keep their hands on the steering wheel.”
Swing and a miss.
Fine, touch but don’t use. You’re not driving the car. The car is driving itself. All the quibbling on specific words doesn’t change the fact that what you insist can never happen is happening right now on America’s roads. Cars are RIGHT NOW AT THIS MINUTE driving themselves.
Does a Tesla know to stop at a Stop Sign?
Nope.
It does a reasonable good job of staying between the lines...with caveats of course. Its nowhere near autonomous.
NHTSA has proposed a rating system from 0 (no special features) to 4 (completely autonomous). The Tesla is a 2.
It will. It’s iterative. Technology progresses. The fact that a 2 is available to consumers right now proves your statement of never is doomed. It’s happening now.
Attaining level 2 has nothing to do with attaining level 4. A camera that reads the road lines has nothing to do with the AI necessary to traverse an intersection. They are completely different paths, not any sort of continuum. They are distinctly different problems. An analagy would be changing a tire - I could reach perfection in jacking up the car, but if I can’t get the lug nuts off, I still can’t change my tire. And my jack does nothing to get the nuts off. I have to go back to square 1 and find a tire tool...and if I cant find one I’m just stuck.
Tesla’s concentration on level 2 is more indicative that they have less interest in attempting level 4.
No it is a continuum. The camera is part of the data to gather. Reacting to the camera is part of the processing of the data. In the end that’s ALL this is: gather data, use it, do it all fast enough to safely control a vehicle at highway speeds.
Tesla’s focus on level 2 is indicative of the fact that they’re willing to ship in iterations. They’re climbing the hill. And so is BMW, Google, Mercedes, Volvo and Ford. And whoever did these trucks. And that’s just the ones we know about.
In the end all purely technical problems are solvable. It’s not like they have to break relativity or thermodynamics. It is, at it’s core, a big data problem, and big data is a happening place right now.
“It is, at it’s core, a big data problem”
Its an entropy problem. You can’t collect ‘big data’, unless you know what constitutes data...and the real world offers an infinite set of variables...you can never know all of it.
Does it have to be perfect? No. But it has to be close to it...and it can’t be.
You seem to have moved the goal post around like its on a 3 dimensional chess board. So lets recap and define some terms.
“there will be a mass produced luxury driverless vehicle on the market within 5 years”
“THEY’RE ALREADY OUT THERE. Millions of miles driven on America’s roads by self driving cars already”
“TESLA IS ON THE MARKET”
“The only reason there’s a driver in them is the law. They mostly don’t do anything, they’re ballast.”
“Tesla, Mercedes, BMW have shipped”
“That [Tesla] IS a self driving car.”
First, the NHTSA classifications:
Level 0: The driver completely controls the vehicle at all times.
Level 1: Individual vehicle controls are automated, such as electronic stability control or automatic braking.
Level 2: At least two controls can be automated in unison, such as adaptive cruise control in combination with lane keeping.
Level 3: The driver can fully cede control of all safety-critical functions in certain conditions. The car senses when conditions require the driver to retake control and provides a “sufficiently comfortable transition time” for the driver to do so.
Level 4: The vehicle performs all safety-critical functions for the entire trip, with the driver not expected to control the vehicle at any time. As this vehicle would control all functions from start to stop, including all parking functions, it could include unoccupied cars.
Tesla, BMW, Mercedes...none of them have ‘shipped’ a car that does better than level 2. They are not ‘driverless’ or ‘self driving’ in any way, shape, or form.
You have predicted level 4 will be attained in 5 years. To me, its obvious that’s impossible. We shall see.
80,000 lbs at 75 mph with no human involved, even as a safety valve? Yea, that’s a plan for success.... (Rolling Eyes)
We know what constitutes the data: the surroundings of the vehicle and the characteristics of the vehicle. The real problem comes in because any way there is to get the surrounding (visually, LIDAR, whatever else you got) is going to bring in a LOT of data, and 90% of it is worthless. The human brain has a lot of built in stuff to make it so we instinctively ignore that 90%. That’s the challenge of the self driving car, in some way shape or form it’s going to get probably 1GB of data a second, and maybe 100MB of that data is useful, it’s gotta find, understand it, and use it, all fast enough to not crash the car AND get the next second’s GB. It’s a pure data processing problem.
For driving the real world doesn’t offer an infinite set of variables. Because most of it doesn’t matter. Those cars on the other side of the street obeying the law and presenting no threat don’t matter, the buildings off the street not falling down don’t matter, the stoplight that hasn’t changed color since first observed doesn’t matter. The trick really is in ignoring all the right data. Which is ALWAYS the trick in big data. The fact is we only commit a couple of dozen actual acts a minute while driving (adjusting speed, steering, etc), it really isn’t that tough, once you’re correctly filtering the info.
The collision evasion stuff that’s in Mercedes today comes pretty close to level 3.
But really what it boils down to is this:
Once upon a time landing on the moon was impossible
Once upon a time leaving the earth’s atmosphere was impossible
Once upon a time breaking the sound barrier was impossible
Once upon a time flight was impossible
Once upon a time self driving cars were impossible
The course of human history is people like you declaring things impossible and being wrong.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.