On yet another one of those long, pointless road trips that seemed to punctuate my life starting when I got my license, I was plying the roads somewhere in eastern Pennsylvania with a friend. He told me that on long trips he’d often relieve the boredom by finding another car from the same state as his destination, and then just follow it. I wasn’t sure then how staring at the same car, hour after hour, mile after mile, would do anything but increase the boredom while making you look sort of creepy, but it seemed to work for him.
What works for college kids in cars also works for long-haul truckers, and the concept of a convoy has long been a fact of life on the road and a part of popular culture. Hardly a trip on the US Interstate goes by without seeing a least two truckers traveling in close formation, partly for companionship and mutual support but also for economic reasons. And now technology is poised to take convoying to the next level, as platooning becomes yet another way to automate the freight.
I’m Not Tailgating, I’m Drafting
The physics of platooning are simple: things moving through the air experience drag. Aerodynamic drag increases along with speed and directly correlates to the amount of energy needed to keep moving forward. The more surface area that a body presents to the air it’s moving through, the more drag it experiences.
Truckers have always taken advantage of drafting as a way to reduce their fuel costs. By driving in the partial vacuum in the slipstream of a lead vehicle, the following vehicle can realize significant fuel savings. The lead vehicle experiences reduced drag, too. This is because the drag-inducing wake turbulence normally present at the trailing edge of the semi trailer is transferred to the rear of the following vehicle. Drafting is an aerodynamic win for all the trucks that participate in a convoy.
But two vehicles operating in close proximity at high speeds can be a recipe for disaster, especially if the lead driver needs to stop quickly. That’s where platooning comes in. Platooning is really just drafting on steroids – a technology assist for what truckers are already doing. Multiple companies are looking into systems that coordinate platooning for long-haul truckers, and one, Peloton Technology, has fielded a working system:
The PlatoonPro system provides both the wireless systems needed to find platoon partners and coordinate them into position, and the sensor suite and vehicle controls needed to safely operate the platoon. It uses a dedicated short-range communications (DSRC) link to keep platoon members together and operate the vehicles safely. Platoons are prevented from forming in congested areas by geofencing, which is determined by a network operations cloud (NOC) that also serves to find platoon partners and to warn drivers of approaching road hazards.
Wherever You Go, I Shall Follow
PlatoonPro is currently commercially available and in use by six customers, and platooning is allowed by law in 18 states in the USA. But as impressive as PlatoonPro is, it’s really just an intervehicle cruise control system. While the fuel savings of platooning can be considerable — up to 7% between the platoon partners — both trucks in the platoon still require drivers.
As we’ve pointed out many times before in the “Automate the Freight” series, drivers are expensive; not only must the company pay their salary and benefits, drivers take sick time and vacation, are subject to quit at a moment’s notice, and potentially bring a raft of personal problems with them to the driver’s seat. From a trucking company’s point of view, the fewer drivers they have to employ, the fewer the headaches they’ll have. So while some companies are in pursuit of fully-autonomous long-haul trucks, Peloton sees value in replacing only half of them.
Enter automated following, recently announced by Peloton. In an automated following platoon, only the lead vehicle has a driver. The following vehicle, equipped with the same suite of sensors and linked to the lead through the same vehicle-to-vehicle (V2V) system as PlatoonPro, is driverless. The following vehicle takes commands for accelerating, braking, and turning from the driver in the lead vehicle. An automated following platoon instantly doubles the freight a single driver can haul, halving personnel costs and still realizing the increased fuel economy of drafting.
While PlatoonPro qualifies as an SAE Level 1 or “driver assist” automated driving system, the fact that there’s no safety driver in the following vehicle makes automated following a Level 4 system. That’s a far cry from fleets of Level 5 driverless trucks plying the highways with cargo, but as Peloton points out, it’s far more doable in the short term.
There are clear benefits to a Level 4 system other than the fuel and personnel savings. Shipping companies will benefit from more flexible logistics, with follow-trucks teamed to multiple leads over the course of a long route. The lead drivers would benefit by operating shorter routes, which would increase recruitment and retention. In addition, the lead driver would need more training and accept more responsibility, and therefore command more in the way of compensation.
Automated following seems like a clear win for both shipping companies and drivers, at least those with the skills needed to adapt to the new system. It’s an interesting idea that should serve to bridge the gap between where automated shipping currently is and where it can be someday.
So basically concrete railroads.
Yeah, but with exits every couple of miles that connect to a network of secondary and tertiary roads that provide an almost infinite number of paths to any address. So, in some ways better than railroads, and certainly more suited to the geography and culture of a country like the USA.
This is not how a lot logistics works though. The local trucks delivery to a distribution hub, another truck takes from hub to hub, and then another truck takes from hub to local. Each stop at a hub allow for sorting and consolidation. It is very unlikely that a long haul style trucker that would favor drafting would also make the delivery to the end destination unless it is a full truckload such as industrial equipment. Those then have their own disadvantages of typically not being a preferred aerodynamic shape.
I don’t understand why we downplay trains so much in America. We are even going so far to talk about ridiculous electric planes when we could build more rail systems.
Long distance rails project are ridiculously complicated to plan and build in many areas. The current estimate for California’s High Speed Rail project is nearly a trillion dollars, up from the initial $33 billion estimate. While it makes more logistical sense to ship by rails, it’s less of a headache in many ways to use existing infrastructure.
No one sane uses the CA High Speed Rail system as example of anything but fraud, waste and abuse.
BTW the typical costs for one mile of class one rail is about $1 million.
Where are you getting those numbers? My google fu says phase one is expected to cost 77B up from the initial 40B.
Trains are not sexy and that’s the problem. Techies don’t care that a intermodal train with a crew of two can haul 1-3 miles worth of cargo that far surpasses any rube goldberg Platooning scheme and that’s limited to very specifical geographical regions.
Trains got screwed over here by GM and the rest of the automotive/oil industry a long time ago. It’s sadly extremely unlikely that the situation will change.
>”would also make the delivery to the end destination”
I worked in a restaurant once, and the beer truck would do exactly that – a big box truck would drive hundreds of kilometers end to end and stop at all the little gas stations, hostels, supermarkets etc. along the way until it was empty. Each place would usually take one or two towers of cases, so there were like 12 – 15 different deliveries in the same truck along a stretch of highway with the end destinations branching out like a christmas tree.
Concrete railroads that haul way less cargo per “train,” use way more fuel, tear up our highway network, and use untested technology that, if it fails, will result in loss of life and limb. But it lets the trucking companies lay off workers, so it must be great!
I used to do the same. College in the 1990’s through Ohio to PA to WV, Speed limits were 65, 55, 65. I’d turn on my CB radio and pull my little 4 cylinder in behind the 18 wheelers and go slow up the hills and fast down the hills in time with them.
(These are obviously aimed at the Peloton people, not the author of this article…)
1. What about when the human needs to make a pit stop? Do all the automated trucks try to follow him into the truck stop single-file?
2. What happens in case of mechanical failure of any vehicles in the “platoon”?
3. What if an impatient jerk weaves in between the lead and the follower?
4. How long before a security hole allows “virtual hijacking” (causing the platoon to start following the wrong vehicle)?
If I were designing Peloton’s system, maybe my answers would be:
1. Breaking down the platoon when entering a truck stop could be tricky, but it’s probably made less tricky by the low speeds and relatively constrained environment of truck stop. With acres of blacktop to play with and pull-through parking spots, and perhaps a little infrastructure to guide the following trucks into parking spots – or fueling bays; that’s going to be a need too – it should be possible. Just need a little AI to avoid the lot lizards ;-)
2. Sticky problem. As a first response, I’d think the lead driver gets a warning and pulls the whole platoon over. But then what? Does he abandon the follower vehicle to wait for service and continue with the rest of the platoon? Not sure how best to approach that.
3. They covered that pretty well in the video – the platooned vehicles automatically increase their distance to allow the idiot with a death wish to be as safe as possible given his or her inexplicable desire to be wedged between 800 tons of truck and cargo. I imagine the same logic would apply for driverless followers.
4. Probably not much longer than existing security holes that are exploited every day by hijackers. The mob needs to make a living too ;-)
For #2, once the platoon is safely off the road, that’s a business decision with some input from law enforcement, not a technical issue. I assume the system will automatically notify dispatch if the platoon stops unexpectedly and that the driver will stay around for cleanup if something catastrophic happens to one of the trucks.
I’m curious about where you got the 800 tons number. Are you expecting platoons of 20 trucks?
>”the platooned vehicles automatically increase their distance”
You know what happens after that?
Leaving a safety gap for the idiot who just squeezed in front of you makes you slow down, which means the next idiot sees you driving slowly and overtakes you as well, and squeezes into the gap you just made. If this process is automated, you can bet that the trailing truck will keep falling behind more and more until they fall out of the convoy.
To simplify the problem I think that a mechanical coupling between the trucks and a system to relay the brake and acceleration signal could be added. There is the problem of the steering, this is a bit harder to solve but I think that some guide could help to follow a straight line, you can actually get rid of the steering wheel of the first truck. You need some extra infrastructure and could be a bit difficult to take detours, this is a big problem. But if the guiding structure is made with steel, one could get rid of tyres and use steel rims reducing the friction, and with a clever shaping of the supports and the rims you could eliminate the differential.
Having to follow metallic supports makes also easy to have an electric engine, you could add a wire or a third support to get electrical contact.
At this point, you can also having a system to force the trucks to brake automatically before reaching a red light, and also you can make them brake automatically if the convoy ahead is stopped for whatever reason. You can use the system to adapt the speed of the convoy.
Yes I think it could be feasible, providing that some infrastructure will be built.
Hmm I think someone had this idea already, about 200 years ago…
“There is the problem of the steering, this is a bit harder to solve but I think that some guide could help to follow a straight line, you can actually get rid of the steering wheel of the first truck. You need some extra infrastructure and could be a bit difficult to take detours, this is a big problem. But if the guiding structure is made with steel, one could get rid of tyres and use steel rims reducing the friction, and with a clever shaping of the supports and the rims you could eliminate the differential.
Having to follow metallic supports makes also easy to have an electric engine, you could add a wire or a third support to get electrical contact.
At this point, you can also having a system to force the trucks to brake automatically before reaching a red light, and also you can make them brake automatically if the convoy ahead is stopped for whatever reason. ”
You realise you just described a railway, right?
I believe that was his intent.
That’s the joke, yes.
Yes, that was the joke.
More seriously, if the problem is more freight to transport, having automated convoys loos as a solution. But there are older and more boring solutions that are available. Leaving existing railways to rot, convert them in bike lanes or roads, instead of renew them, make easy to move freight on railways and then proposing the automated convoy doesn’t seem to me a good idea. More effective is develop an automated contained load and unload between lorries and wagons.
Some railway: https://youtu.be/S-snlsDDMqM?t=1400 as you could see for a long stretch this new high speed railway is running near a motorway for a long stretch, this motorway is full of lorries going end to end, some of them are arriving from abroad and the question is why isn’t used for freight?
And only 1-2 Million (USD) per mile for that infrastucture. Plus all the extra miles to make all the branches and lines to individual businesses, plus the cost of having to buy all the land for that new infrastucture to be built on…and if you only have one lane of that new infrastucture, any breakdown/maintenance of that infrastructure or vehicle using it completely shuts down that connection.
Road Trains https://en.wikipedia.org/wiki/Road_train
Makes me think of the Australian truck trains… https://www.youtube.com/watch?v=0iFkKRh5kcM
yes, I can’t understand the point of this if you are only doing two or three trucks, as you may as well just connect them together like we do here in Aus.. 3 trailer ones are very common here…
Mind you, I do see quite a few of them traveling as a group as well, so maybe they could then be platooned..
If it is part of a organised group I can see this all being a good concept – however when I was recently doing a long drive I had a car catch up to me then sit maybe 10m from the back of my car no matter what speed I did. I suspect he had active cruise control, and it must have been adjustable to that quite short distance..
Very annoying.. When you are doing 120km/h and may have kangaroos jump out in front of you, you don’t want a car that close behind.. I slowed down to 50km/h and he still wouldn’t stop it, so I eventually had to put the foot down and go faster than he could to get rid of him…
It doesn’t have to be that way, I’ve done long distance (say 1000km/day) with a group of other cars (of people I didn’t know) and it has worked well – especially at night. You all take turns being the front car – which is much more work at night (the front car also with high beam lights on) – then the cars behind follow the lights of the car in front and can also see a long way down the road..
One of the inevitable solutions will be to have a driver in each truck, but they get paid a different rate when they are ‘in command’ of the vehicle, which is only needed at times like rest stops/refueling, etc. The rest of the time they can sleep and not count against their log hours, which gives them more run time. Think aircraft that rotate crews through the cockpit on very long flights. Only in this case, you swap out the lead truck with one that has a driver with hours to burn.
I don’t think the unions would go for that… ironically eliminating a number of drivers all together is easier and more cost effective then working with the union to get a more equitable deal for a larger number of drivers…
A truck stop outside of major destinations is what’s really needed…
Not your typical truck stop, but something redesigned for depositing a convoy of trucks so that waiting drivers can move them the last mile, or drop off trucks waiting to get on the highway as part of a convoy…
Get close to the city, pull off into the stop, your following trucks that need to detach will do so, and any trucks heading the same direction will attach.
Also a point where long haul truckers and swap out and maybe get their required rest…
Of course doing such would require massive coordination among all the organizations operating trucks on the highway… and that seems unlikely.
So it would only work somewhere desolate like Wyoming or idaho because somewhere like chicago the cut ins would be constant thus rendering it useless, also what about grades example lead truck is much heavier loses lots of speed on the climb but truck in rear is light and is now coming up on leader atleast 30mph faster than lead, car in left lane so rear cant do anything but brake hard to try and keep distance making all fuel saving pointless because now momentum has to be regained to climb hill. There is reason we are told to keep at minimum four truck lengths in front of us and it isn’t to annoy the driving public. By the way that entire scenario would be constant in desolate places like wyoming, so where exactly is this supposed to work?
Driving the truck is only ONE of the reasons for having a truck driver. They’re also there to find problems with the trailers (Trailer breakdowns are a more common reason for trucks to stop/break down than problems with the traction), load, lash, unlash and unload cargo, make sure the administration is in order, detect potential problems with the truck before it becomes an acute problem, refuel the vehicle, perform basic service, guard the cargo, etc.
It’ll be a LONG time before we don’t have anyone riding in the cab, I even doubt it’ll ever happen.
There might be an existing (Retrotacular?) HaD article on a related concept (https://en.wikipedia.org/wiki/Overland_train) (https://www.techeblog.com/overland-train-is-the-worlds-longest-off-road-vehicle-here-are-5-interesting-facts/). Basically, each trailer had its own steering and drive system and would follow a path made by the cab.
Just a question about drafting, when ducks fly in formation the lead duck provides some aerodynamic advantage to the following ducks, which lets them fly farther, but he uses more energy so every so often the lead duck switches off with another duck who becomes lead. OK enough ducking shit, my question is could fuel usage between lead and following trucks drafting be kept more constant by switching off leader and drafter? Just asking.
I read about a year ago that Japan was going to try platooning this year.
Multiple EU companies have been trialling this in the Europe for a few years now:
https://www.theverge.com/2016/4/7/11383392/self-driving-truck-platooning-europe
The idea is to have a Europe-wide standard so a Volvo truck can follow a Scania (for example), and they don’t all have to be from the same manufacturer.