I have written before about automated cars. Today the British government announced that it will allow a trial of automated lorries on motorways to go ahead next year. The idea here is that a human drives one lorry, and automated ones follow close behind, saving the cost of extra drivers and reducing air resistance.
The Automobile Association complains about it.
A platoon of just three HGVs can obscure road signs from drivers in the outside lanes and potentially make access to entries or exits difficult for other drivers. On the new motorways, without hard shoulders, lay-bys are every 1.5 miles. A driver in trouble may encounter difficulties trying to get into a lay-by if it is blocked by a platoon of trucks going past.
I think they are overstating the problem because there are already convoys of human driven lorries on motorways. It is already a good idea not to drive alongside them for any distance. Something I do see as a problem is reported matter-of-factly by the Telegraph:
The Government has provided £8.1 million funding towards the trials, which will initially take place on a test track before being carried out on motorways.
I left this comment on the Telegraph’s news article:
If some private company was spending their own money I would have no complaint. If it is a good idea, people will do it and they will invest their own money in it. I have no idea why the government thinks it is a good idea to hand out free money to anyone who goes begging with the right story.
As for the idea itself, I can imagine it working. The lorries can drive just inches apart so unlike others I think slipstreaming will work and there is little risk of cars getting in between the lorries. Someone asked about trailers and a powered trailer may also work but I can also easily imagine that some electronics would be cheaper than a heavy mechanical coupling.
The real test of the idea is whether someone can make a profit at it with their own money (third party liability included). It is the government subsidy that is causing the controversy here.
A better idea, and much more realistic, is the one that has already been tried in the US. Automated trucks do have a human driver. While the truck is on the highway, or motorway, it is driven by the automatic driver, the human rests. The human driver takes over for the last couple of miles of access from the motorway to his actual destination (urban driving).
This way truckers can “drive” far more hours without stress or fatigue. This seems to me a very big and significant improvement, and one that is clearly and easily feasible already now.
To be fair, my car drives itself. I switch on cruise control and provided I’m on a motorway, there’s no traffic, and the road is dead straight the car simply drives itself allowing me to rest. Hardly surprising they’ve managed to get this system into a truck as well.
How about improving it further by removing the cab and attaching the second lorry to the first. Aka making the lorry longer???
Who goes to prison as and when someone gets killed by such a truck?
Other countries already have trucks set up somewhat like this. Look up “road train” in Wiki. In Australia they have these “road trains” of up to 4 trailers behind one tractor. In the US, there are trains of up to 2 or 3 trailers (depends on the state) behind one tractor. Right now, there is one driver in the tractor, pulling a bunch of trailers. This proposal wants to make the trailers autonomous.
Oh, and my new Subaru has a “lane minder” option that watches to see when I drift over to one side or the other. I can set it to just alarm and alert me when I drift, or I can set it to nudge the steering wheel to get me back in the lane. It is effectively self steering, but Subaru will not let me take my hands off the wheel for more than a few seconds – it sounds an alarm and tells me to put my hands back on the wheel.
There is also an automatic braking feature that will slow down the car to follow the one ahead automatically. It will bring the car to a complete stop if the traffic stops, and then it will start back up again, automatically (I think – the car is brand new), when they start moving again. Great for heavy traffic.
If they are going to be inches apart, why not do away with the cab/engine of the following trucks and have series of close-coupled trailers instead…? We could call them road trains…
In fact, better still put a mahoosive engine up front, dozens of ‘trailers’ attached behind and put them all on rails… We could call them trains…
Another interesting question is whether your vehicle could be programmed to kill you under certain circumstances. Eg. If it has to choose between driving into a crowd of school children or steering you into a concrete wall.
Surely the advantage of a chain of lorries rather than a lorry and multiple trailers is that they can split up and service different destinations when the bulk of the journey is done. Similarly they can pick up at different points before assembling for the main journey.
Presumably electronic linking/ relinking is simpler than mechanical.
Not sure about the funding, but I would expect the DOT to insist on rigorous testing before licensing the system. And since the licensing is for the public good I don’t object to public funding for this.I would definitely object to public money for rolling the system out once approved.
Much as I am all in favour of the driverless-car revolution I find myself in sympathy with the AA. Here and now, many is the time I have found a sign obscured by a lorry. When you have platoons of the things that is going to make things much harder. Also, how long are these platoons going to be? Much longer than 3 would make it difficult to move into the inside lane at a junction. Will they create a gap if you need to move?
Also, have you seen how close lorries drive to one another with Mark I human drivers at the wheel?
Patrick, maybe they will move apart to let cars through. Maybe that’s another advantage of having them electronically linked instead of mechanically linked in trailers.
Maybe it is a daft idea. Maybe it solves some problems we haven’t thought of. The way you find out is: does it make a profit?
Government subsidy interferes with that, of course, possibly making daft ideas financially viable.
We engineers have been working on AVs for 25 years: this particular idea shows how rail-guided trains and locomotives will be replaced. HS2 is obsolete before it has been built. Companies like Hitachi get to sell us £50+ billion of redundent technology and in 20 years time when the rail lines are all ripped up – they get to do it all over again. Nice work if you can get it.
I predict a new hobby developing amongst bored (or terroristic) hackers: taking control over the unmanned rigs and crashing them.
If they’re not hard-wired together, they’re vulnerable.
@bobby b
There are many ways to get around this: just one way is to deploy a matrix of dedicated and shielded proximity sensors and others both on-board and off-board plus GPS. Basically belt, braces, string and sellotape. all sorts.
There used to be something called “railways” – they went everywhere and were more efficient at carrying heavy cargo than lorries are. The government undermined railways by giving special rights to trade unions (see the “Strike Threat System” by W.H. Hutt – the “Collective Bargaining” system depends on government interventions) and by price controls on the railways and by building a network of “free” roads – drove the railways down and then nationalised them, then the government shut much of the railway network.
The government creates a problem (congested “free” roads), by destroying much of the railways – and now it claims that “driver less trucks” will solve the problem it created. And it charges the taxpayers – yet again.
What a concept! 😆
I live in Australia, home of the “serious” road-train.
Triple-trailers are quite common on the long-distance highway runs, but only single trailers are allowed in towns / built-up areas, hence “marshalling yards” on the outskirts of medium to large population centres.
Triple-trailer fuel tankers are quite common, as are multi-deck, triple-trailer livestock transports.
so if you’re a truck driver in england, you’re paying taxes to help develop the technology that will replace you. what fun! you’re subsidizing your own replacement. kind of a metaphor for the whole collapse of europe.
There are lots of ways to stop hacking- that we know of. Problem is the hacker just needs one mistake…
Do computers need help to fail?
Lots of the ones I encounter still manage to collapse without needing any hackers. Let alone hackers with terroristic resources.
This driverless crap is trendy and the Fish-Faced Hag and her gang are nothing if not wannabe trendys. “Very stylish” as Dirty Harry used to say.
What they should and –of course–won’t do is have a clear plan of response for when self-driven vehicles cause their first monster pile-up with loadsa fatalities.
You might think they could anticipate such an eventuality but we know they won’t. They will just throw our cash at those who survive.
Meanwhile, the government is considering yet another non-automated transport boondoggle manned by very expensive un-sackable personnel with gold-plated final salary pensions who are prone to strike, overtime bans and extended sickness.
Paul is quite right that in Britain, the government made it difficult for the railways to compete, then bankrupted them and then nationalised them.
I suspect that even in a perfect free market roads, with their greater flexiblity and lower maintenance requirements would eventually have re-asserted their dominance. But we’ll never know.
The only two examples of a relatively free market in freight – Japan and the United States – offer extreme contrasts. The American freight railway is extraordinarily successful. The Japanese despite – or perhaps because of – the successful private passenger railways, is tiny.
Are freight trains more cost effective in the US because of the long distances involved? Railways aren’t able to take things all the way to or from their destinations, so the journey usually must involve truck + train + truck. If the journey isn’t that long then it is probably quicker and more cost effective to just use the truck. If the journey is very long, as it is likely to be in the US, then the train part of the journey must be far more cost effective.
Railways. No, the govt did not bankrupt them, but it creamed off massive profits from them during the Second World War, not even allowing them to earn the maximum profit to which they were entitled (‘standard revenue’ – sic) under the 1921 Act that grouped them. Then, of course, having helped itself on nationalisation to the arrears of maintenance funds they rlys built up during the War, it spent very little on them for a decade…
Sorry this it totally OT but I randomly clicked on a link on Tim N’s place to Anna Racoon’s site and there is a note there to say that she passed away a week or so ago. Just thought I’d mention as many of you will be familiar with her work.
How do the companies promoting this autonomous vehicle technology expect to avoid the enormous liability they would, by necessity, assume?
Insurance. Just like the private rocketeers buy before sending their rockets up over us.
There’s no risk that someone won’t underwrite.
The coming technological change will revolutionise travel: electric cars and driverless/connected data technology renders rail lines as a means to guide a vehicle to its destination no longer necessary.
It renders HS2 obsolete before it has been built.See:
As the age of autonomous vehicles nears, why are policy wonks focused on the past?
“The delegates to the 1898 urban planning conference failed to recognize the developments that would transform their world. Today’s transportation infrastructure discussions — about building a $10-billion bus terminal in New York, or a $70-billion high-speed rail system in California — may prove similarly short-sighted. These transportation mega-projects don’t seem to take AVs into account. Yet by the time these initiatives are completed, AVs will be a major part of the transportation landscape.”
http://www.latimes.com/opinion/op-ed/la-oe-1108-strauss-autonomous-vehicles-20151108-story.html
And:
Driverless public transport will change our approach to city planning – and living
“fundamental change could turn transport and urban planning on its head. Autonomous vehicles are likely to be used very differently from the vehicles of today – replacing existing transport businesses and creating new ones.”
https://theconversation.com/driverless-public-transport-will-change-our-approach-to-city-planning-and-living-35520
The fear in the auto industry is not that the major automakers will pull out of the UK because of Brexit but will many of them still be in existence at all in a few years?
TPH–Come back in 50 + years.
Electric cars are crap. Short of a battery breakthrough that can match the energy density of petrol etc they have no future other than some sort of idiot eco-statement.
Driverless cars are a long way off in software terms. The hype is big–and political scum are drooling over the power and control driverless will give them to turn private into public transport BUT the AI required is a long way off–if you look thro’ the hype and plain tripe.
You are entirely right about HS2 however. Which is why Trasher May will build it and waste our cash come Hell or High water.
Whether you consider rail or road freight, the most effective method is to have small individual vehicles, there is no efficiency of scale, and any that exists is completely wiped out by the final part of the journey when the goods have to be delivered, something that is specifically reduced by individual vehicles.
The biggest problem has always been the human in the cab, the most expensive, unreliable and inefficient component of the entire affair, the thinking of big trains and lorries has almost entirely to be concerned with the driver (or other personnel) and minimizing the impact, and lest we fall into the “Five Monkey Analogy”, the automation is a great opportunity to resolve this longstanding issue.
Instead of “training” lorries, just tarmac the railway lines and put automated vehicles on them.
Here in the union stronghold of Southern Rail, yet another period of misery as drivers and guard inflict on us how important their job is and how they are worth every bit of salary and pension we pay them.
What Mr. Ecks said.
I am with Mr.Ecks, just another stupid and expensive idea from the idiot factory, aka parliament. Incidently, HS2 was a directive from the EU under the name TEN-T, a pan EU rail network. But I do worry about the abilities of our politicians, they seem to have the IQ of a plant, or less! Electric cars with 19 hours required to charge the batteries, utter nonsense, as is “driverless cars”. Computers are not God and are thus not infallible as anyone with a computer will know. Crashes, as in the computer world do happen, too often to be put in charge of two ton of motor car. Even in motorways with the ever changing traffic patterns, split second choices have to be made, the lorry that suddenly pulls out in front of you, and then attempts to overtake another doing the same speed…up hill. God preserve us from these kinds of politician, PLEASE!
With regard to battery technology and breakthroughs in same, I offer this. I am no engineer or battery expert, so I can’t offer an opinion on the merits.
@ Mr Ecks
Engineering progress is incremental and the problems will be solved.
Re: Design of driverless cars
When you remove the following from a car:
• Engine
• Engine management
• Engine control
• Gearbox
• Fuel injection, carburettor, fuel pump and fuel management
• Turbocharger or supercharger
• Exhaust system
• Exhaust gas re-circulation
• Fuel lines
• Fuel tank
• Ignition
• Starter motor
• Transmission and drive train
• Water cooling, radiator and fan
• Cooling for air, cooling for oil
• Air filters and oil filters
• Lead acid battery, charger and alternator
• Associated instrumentation
And you insert:
• Battery under floor or in bodywork
• Motors in wheels
• Automated motor control/steering/connected car data software and hardware
• Wireless charging/drive
What you achieve is:
A saving of a great deal of space and weight: which is why Autonomous Vehicles (AVs) look so odd: you don’t need a bonnet because you don’t have an engine or anything else for that matter, so you can do all sorts design-wise. And you reduce the vehicle to a device not much more complicated than a Kenwood food mixer.
The implications are:
Production and design innovation can take off using advanced production techniques, 3D and 4D printing and customised programmed production. A whole raft of new entrants may put most of the established automobile makers out of business. The existing vast production facilities geared to export are replaced by smaller highly automated and flexibly programmable factories sited where the markets are.
The implications for society are immense and not yet understood: transport systems and technologies are merging like TV/PC/mobile/laptop technologies are merging. Will there be any call for car ownership at all? Will self-driven car ownership run parallel with driverless and be a symbol of status? It will impact on insurance, taxi/lorry driving, city design, mass transport systems and very much else.
“Electric cars with 19 hours required to charge the batteries” — no real reason why it has to take 19 hours to recharge and irrelevant anyway when you can swap the battery in an instant.
“the AI required is a long way off” — this is more likely to turn out to be true. I think it’s a solvable problem but software takes twice as long to finish than we expect, even when we take that rule into account. But the right kind of infrastructure could simplify the software, perhaps.
Isn’t this a rather rosy view of coming battery design?
My memory is that, looking at the different models of the Teslas, from the little 2-seater (3000lb) roadster to the sub-SUV X (GVWR of almost 7000lbs), the batteries take up a lot of room and make up about one-third of the vehicles’ very hefty weights.
They’re so massive that they have to be built into the car such that battery replacement is like pulling an engine.
I think we’re far off from popping open the battery compartment and swapping batteries.
Short of a battery breakthrough that can match the energy density of petrol etc they have no future other than some sort of idiot eco-statement.
It doesn’t have to be all battery, does it? A simple generator converting another energy dense fuel will do, start off with a petrol one and swap it out for gas turbine, hydrogen, or even nuclear when they get up to speed, get the rest of the design right and worry about the power pack later. The all-battery concept is an eco-idiot design fixated on the belief we can charge them via windmills and we’ve deluded ourselves the design cannot move forward without one.
Transmission and drive train
This is the most infuriating thing that I find in recent designs, where they simply replace the ICE with an electric motor, in fact, the in-wheel concept (where the motor is the wheel) has value in added stability.
the AI required is a long way off
AI? What AI? This is another of those utterly stupid concepts brought in to muddy the waters, as if the car has to resolve the trolley problem, what utter balderdash, a human can’t even do this right now, so why would an AI be expected do it. Simply having enhanced sensors and a linked transport network, you avoid most of the issues in the first place. The only AI needed here is no more than a Sat Nav provides today.
Just remember, driverless cars will likely lead to no-one owning the car and they are just glorified Uber Cabs. The control of the vehicle, and it’s safety, will most like be centrally managed in tight traffic areas, and given broad directions elsewhere.
TPH–That is just a lot of techno-goobldegook. You might want to swap your car for some sort of glorified milk float but I won’t. Not DON’T–I WON’T. What you describe still needs a battery unless you are suggesting drawing power like a tram.
The rest of your remarks are just a collection of buzz words–you left out stem cells BTW.
Runcie–The eco-freaks rule so your “power brutale” plans are non-starters. And trying to design a vehicle without regard to its power source sounds more than a bit iffy.
Plus–and a far bigger problem–how do “enhanced sensors” and a “linked transport system”–which sounds liked centralised control–get you down the street? Some form of computing needed. Or will lots of girls be brought in to do the job computer free –like old fashioned telephone exchanges?
Mr E, my comment said that the computing power required would match that of today’s Sat Nav, no fancy AI needed, the computing technology is already here.
But how would it even work…?
Imagine my journey today, I jumped in the car with a bag of rubbish, stopped at the bin and dumped it, drove to McDonald’s for breakfast and parked, did a crafty (illegal) turn to save five minutes and headed to the Post Office, parked in an available space on the kerb, posted my mail then headed off to the DIY centre and parked as close to the door as I could, then left there, went to have the car MoTd and headed home on the most efficient route from there (which was different to my route to town)…
How would SatNag do that…? I’d have to programme 6 routes prior to departure. But it couldn’t park at McDonald’s or anywhere else… there would need to be some clever technology at McDonald’s and the post office and everywhere else that told the car how many spaces there were and how many were free…
Yet I worked the whole thing out in a minute while drinking my first cup of tea… Why would I ‘upgrade’ to a system that would be vastly more complex and expensive…?
Quite. I would much prefer an Uber-like system in which I just use my phone to call a self-driving car from anywhere, and just walk away from it when I don’t need it any more.
Tesla were swapping batteries in 90 seconds but gave up because no one used the service. But it’s possible. And robots could do it with the right design of car. Motors in wheels and no need for cobtrols or seat belts surely opens up design possibilities.
https://www.google.co.uk/amp/amp.timeinc.net/fortune/2015/06/10/teslas-battery-swap-is-dead/%3fsource=dam
I imagine Uber type services will be far more cost effective than private car ownership. This is already nearly true with human drivers in some places.
You also would not need to park.
Imagine my journey today, I jumped in the car with a bag of rubbish, stopped at the bin and dumped it, drove to McDonald’s for breakfast and parked, did a crafty (illegal) turn to save five minutes and headed to the Post Office, parked in an available space on the kerb, posted my mail then headed off to the DIY centre and parked as close to the door as I could, then left there, went to have the car MoTd and headed home on the most efficient route from there (which was different to my route to town)…
Imagine your journey, order a car via Uber, jump in the car with a bag of rubbish, say “OK Google, go to McDonald’s”, press pause at the bin and dump rubbish, get to McD’s, abandon the car (no parking, it will toddle off for another job), have a leisurely breakfast, order another car, “OK Google, go to the nearest Post Office”, abandon the car as you’ll be waiting 15-30 minutes to speak to a person with a “what the f*ck do you want” face (some things will never change), post mail, order another car, “OK Google, look up cheapest (insert DIY item here) and go there”, abandon car, during the journey have a quiet chuckle at at the old days when you had to get an MoT and insurance and queue for fuel, get DIY gadget you’ll never use, order car, “OK Google, take me home”
Now can I have all that, but with DuckDuckGo?
Here is an interesting talk (about 54 minutes total) on disruptive technologies, but what’s relevant to this discussion is that starting around minute 37 he talks for about 7 minutes about the end of car ownership. (The rest of it is about batteries, self-driving cars, solar energy, etc.)
So how do you go courting in driverless cars?
@Runcie
I doubt if the customer that followed me would be happy to find the vehicle smelling of freshly cut timber and their clothes covered in sawdust… No more than I wan’t to get in one that smelt of the previous customer’s BO or the contents of their child’s nappy.
You’re under the misapprehension that I’d like to get rid of my three vehicles and take a ‘taxi’ everywhere. In fact, I love driving and even if there were an infinite number of ‘taxis’ and I didn’t have to wait outside McDonald’s the post office or the DIY centre for a free vehicle to arrive I’d still prefer to drive.
The only time I’d want a taxi is to take me to/from the pub… 😉
Now can I have all that, but with DuckDuckGo?
I probably should have said “Ok Uber”, but I’d imagine there would be a number of competing services, unless the government regulates it stillborn.
doubt if the customer that followed me would be happy to find the vehicle smelling of freshly cut timber and their clothes covered in sawdust
I doubt if Uber would want to send you another vehicle, either, your choice, your responsibility.
You’re under the misapprehension that I’d like to get rid of my three vehicles and take a ‘taxi’ everywhere.
You are absolutely wrong on that one, I was explaining how it is possible to achieve a level of independence using driver-less hire cars that matches the requirements of a privately owned vehicle for most people, without the burdens that many of us find frustrating (and unnecessarily expensive). I have no intention of forcing everyone off the roads for the benefit of society and Uber, what do you take me for – a socialist ?
This technology will have far reaching consequences, as much as the internet and mobile phones.
The only time I’d want a taxi is to take me to/from the pub
In the future, everywhere will have booze and weed, unless the religious nutcases get their way.
John Malpas: “Hey, Car, take us to Lookout Point.”
Everything authoritative that I’ve read indicates that the driverless car future needs to move away from allowing any human drivers.
For a crowded traffic scenario, the cars are going to need to be networked together, and a car in the middle that does not obey the schooling behaviors because some dumb human still has control makes it a much less efficient system.
So I think this “you can still have your cars” thing is a very temporary and not very honest way to quell resistance.
I can see inner-city zones becoming driverless-only zones, with parking facilities surrounding them so that you can drive to the city, park, and find a driverless cab. But the two concepts don’t play well together.
Starlings avoid collisions in their murmurations without, it seems, being networked, but by monitoring 6 or 7 proximate starlings. Perhaps this aptitude could be used to model collission avoidance for vehicles?
Very interesting, Mr Ed. Makes me think of peer-to-peer networking, in which ideally there are no single servers (control sources), but rather each machine in the network is “both a client and a server” to others.
However, not to go too far O.T. (moi?!), but this:
strikes me as contradictory. Which is it: the behavior is 100% instinctive, thus uses no birdbrainish intellect and thinking ability at all; or the birds do it almost without thinking, which implies that they do indeed think, just not very much.
. . .
bobby, thus:
First reaction: Oh ye of little faith! :>((
2nd thought: Oh ye who have plenty of faith that with this idea we’ll surely be taken for a ride!
Now let loose fifty crows into that murmuration – crows using their own system (or non-system) and see what chaos it causes.
You end up with large wormholes through the flock as everybody goes banging out of their way to avoid the strangers.
Picture that result in downtown rush hour traffic as the school of fish – er, I mean the driverless cars – swing about to avoid non-plugged-in me as I pilot my own car. Huge efficiency loss.
So I will quickly be regulated out of my driving.
(ETA: Julie, can one be faithfully cynical? If so, that’s me!)
Ever read Neal Stephenson’s “Seveneves”? (My favorite author.) Goes into this fairly deeply for large swarms of small space capsules avoiding collisions and meteorites.
bobby: Oy vey! :>((
So I think this “you can still have your cars” thing is a very temporary and not very honest way to quell resistance.
We still have our horses.
Picture that result in downtown rush hour traffic as the school of fish – er, I mean the driverless cars – swing about to avoid non-plugged-in me as I pilot my own car. Huge efficiency loss.
A driverless car is not going to “get out of your way” just because you are human, they’ll all be driving nose to tail and you wont get the chance. The efficiency loss will be yours, I think, and who cares if the automotive luddite wants to expend his own time and effort accumulating pent-up road rage, even if it makes me a bit late, I’ll just have another beer and download another article to read, or binge-watch another episode, I probably wont even be aware of it, in my little Uber-cocoon.
From what little I’ve read, driverless cars are currently programmed to yield to encroachments, and drivers can have their way with the driverless vehicles simply by somewhat aggressively encroaching on their path. In other words, if I drift around in several lanes, driverless cars will slow and turn to avoid me. Unless they change this, an actual driver will always be a fly in the ointment.
Let’s face it, the politicians and bureaucrats want us using driverless cars that will check the passengers’ tax and social media records and, on detecting a suspected transgression, it will lock the doors and drive the suspect away to an ‘interview’ with the authorities.
They will change this.
In other words, if I drift around in several lanes, driverless cars will slow and turn to avoid me.
Sounds like endless opportunities for monkeywrenching.