Driverless Buses Take To The Road In Scotland

Scotland! It’s the land of tartans, haggis, and surprisingly-warm kilts. It’s also ground zero for the first trial of full-sized driverless buses in the United Kingdom.

It’s not just automakers developing driverless technologies. Transit companies are desperate to get in on the action because it would completely upend their entire existing business structure. Now that self-driving buses are finally approaching a basic level of competence, they’re starting to head out to haul passengers from A to B. Let’s look at how the UK’s first driverless bus project is getting on out in the real world. 

On Schedule

Members of Stagecoach’s Co-Design Panel were invited to an early trial run of the autonomous bus route. Credit: Stagecoach

Scotland’s autonomous bus trial, known as the CAVForth project, has been a long time in the making, and we first looked at it last year. Trials were intended to begin long ago, but faced multiple delays due to the COVID-19 pandemic. Now, the rubber is finally ready to meet the road, with the buses imminently ready to hit the public roads.

The trial will see autonomous buses operated by Stagecoach Group taking to the roads of Scotland. The initial route is a 14-mile run between Fife and Edinburgh Park, which crosses the famous Forth Bridges along the way. The bus service itself has been duly designated “AB1”, standing for “Autonomous Bus 1,” and will be the first full-sized autonomous bus operating on public roads in the UK. The project is being undertaken as a partnership between transit operator Stagecoach, autonomous driving company Fusion Processing, and bus manufacturer Alexander Dennis. Also involved are Transport Scotland, as well as Edinburg Napier University, Bristol Robotics Lab, and the University of the West of England.

Unlike some smaller-scale trials, AB1 is set to be a regular timetabled service open to members of the public, just like any other regular bus route. When it scales to full-time service in the spring, the five autonomous buses involved in the project will have the capacity to take on approximately 10,000 journeys a week.  Notably, though, passengers won’t have to worry about hailing the bus. To avoid having to develop a “hailing detection system,” the bus simply stops at every bus stop along its route.

The buses are capable of driving at up to 50 mph, and have been developed to operate autonomously in mixed traffic. The buses will be capable of SAE Level 4 self-driving. This means there is no need for the human driver to pay attention or keep their hands on the wheel at all times, but the bus can only drive like this in select limited areas. A human driver will remain on board as a monitor for safety reasons, and in part to comply with UK regulations. Beyond that, the buses will also be staffed with a “bus captain” who can assist passengers with ticketing, boarding, and alighting the vehicle.

While this person is hailing the bus, there is no need. Buses on the AB1 route will stop at every bus stop, which eliminates the requirement for the bus to detect hailing passengers. Credit: Stagecoach

An initial trial run took place with Stagecoach’s Co-Design Panel, a group of local bus users that the company consults on service and equipment projects. According to one rider, the experience was, by and large, just like catching a regular bus. “I wasn’t worried at all about it,” said Fleur, adding “You wouldn’t know the difference between this and a normal bus from the driving.” Given the concerns around the sometimes-erratic nature of certain “self-driving” systems, that’s almost a glowing endorsement. The Co-Design Panel recommended the “bus captain” concept to ease passengers that may have questions or concerns about using an autonomous bus, as well as developing materials to communicate how the bus will work to the broader public.

In much the same way that self-driving cars promise to revolutionize personal transport, self-driving buses aim to do the same for public transit. Overall, the basic technologies used for self-driving cars and self-driving buses are by and large the same. However, it bears noting the stakes are higher – there’s a lot more people on board a bus that will suffer the results of any poor driving decisions, and a bus running over a car does a lot more damage. Companies are also juicy targets for litigation, and transit drivers care strongly about the safety of their passengers. While the owner of a modern car may choose to take their life into their own hands by trusting a self-driving system, transit operators and their drivers can’t be so cavalier. Their decision to trust such a system isn’t taken lightly, and that is likely why we’ve seen self-driving buses take longer to show up on the road.

In any case, the CAVForth project in Scotland will be watched closely by many in the UK and around the world. There’s nothing like the pressure of sticking to a tight bus schedule, day in, day out. If the buses can handle the AB1 bus route without relying on human drivers, it will mark a historic day in the history of self-driving technology. It will also leave the companies behind the project with a highly-valuable product on their hands – the technology to create self-driving bus fleets in cities around the world.

70 thoughts on “Driverless Buses Take To The Road In Scotland

  1. > transit operators and their drivers can’t be so cavalier.

    You’d think. Around here, when the bus driver puts the blinkers on they’re going to pull up right NOW. Doesn’t matter that you’re already past the first quarter of the bus, they’re f***ing going because the driver ain’t looking and they don’t care. People have gotten used to the fact that the bus drivers are mad, and will keep a wide berth whenever possible. That works in the bus driver’s favor, because if they start to yield for the cars, they’ll never get back into traffic and they’ll be late – so they drive aggressively on purpose and let the company handle any complaints.

    A self-driving bus could in principle just drive like an ass-hat and nobody would notice the difference because they’re already doing that.

    1. Hmmm haven’t commercially driven a bus, eh? Tight schedule in crap weather with really pissed in your face sober passengers, the drugged ones are worse….yahoos in cars not paying any attention to anything but the cellphone conversation. All that and more for just above minimum wage. Sure after 10 years you get paid more if you’re union, maybe.

      I’m waiting for trials on the A9 north of loch Lamond

        1. They could pay a few pennies more so all the drivers weren’t crazies and druggies. But government work so what are we going to do?
          There will always be air thief nieces/nephews of powerful people that need to be employed. Police departments and courthouses only have so many spots, so bus/ambulance drivers they are.

          I sometimes get a slight moral twinge selling clean urine to Otto. It goes away…joke, I have no personal access to ‘clean’ urine.

      1. I’m sure the crazy folk don’t make it as idyllic as it could be, but 11 quid an hour for driving a vehicle about all day seems pretty reasonable. I’ve never met a bus driver who didn’t seem absolutely sick of life though, apart from an airport link bus driver who took every opportunity to ask me where I was going and advise me which stop to get off at. He was a champ. The previous one asked me if I wanted a ticket, when I said “no thanks” (what do I do with it, I don’t need a receipt, I thought it was a genuine question) she printed it, looked hugely infuriated and shouted “I’ll take your f***ing ticket then” and ripped it out of the machine. My only response was “you asked if I wanted it, I said no and you printed it anyway?”. In my mind, it’s usually the bus drivers that are crazy.

    2. So… driverless busses need 2 humans on board instead of just one? That’s how they got them past the unions!

      I’m not convinced about removing drivers; if they put a conductor on, maybe, but I can see that being a very low skilled job compared to driver, and the crap drivers have to deal with to keep the people on busses safe in the evenings is an issue.

      Ultimately, if they can’t do AI controlled trains, why are we attempting busses? Trains don’t even need steering, and don’t have to deal with other cars and pedestrians.

      Also, does the Cambridge trial not count as “full size”? They look pretty big even if their capacity wasn’t high, and I think Cambridge roads are about the worst I’ve seen anywhere in the uk.

      1. It has always amused me that planes (3-axis) were the first thing to have autopilot, then we started trying to build auto-pilot cars (2-axis), but we still cannot keep trains (1-axis) from falling off the tracks…

        1. There are driverless train systems out there, and have been for decades. Its just somehow never caught on and got rid of all those stupidly overpaid level of expensive and unionised people that drive trains (at least in the UK its a really really well paid job)…

          Which is kinda stupid, as anything happening in visual range on a train on the move is too late for the train to do anything about anyway, all that inertia on low friction tracks… So the safety of the system in general use is all down to the signalling.

          1. The driverless systems never really caught on because you always need somebody on-board to sort out the mess when things go wrong – otherwise you’re stuck there for ages before they get someone over – so that person might as well be the driver. Saves the cost and complexity of the automation.

          2. Besides, automation for the sake of automation is never a good idea. There’s enough people as it is with nothing better to do than beg money for making youtube videos, so if some of them could do something useful like driving buses and trains, they wouldn’t need to do the useless and everyone would be better off.

          3. Driving a train in the UK is adequately paid – starting at £30k and eventually getting up to £60k, i.e. enough that you can hope to buy a home one day without help from parents. It’s a matter of perspective whether this, or the fact that other ordinary “careers” don’t come close, is stupid.

            Driverless operation is possible, it’s just more complicated and expensive than most BBC news viewers guess – the fact that management doesn’t want it is a good clue as to whether drivers are actually overpaid.

          4. @bobtato
            When you compare to say junior doctor, or quite a few other much more highly skilled jobs folks have racked up heaps of debt to study for its a stupid amount… That house prices round here are even more stupid is a whole other problem, if everyone earning under say 100k could get a bump with heavy weighting towards the lower paid jobs it would be good for making houses affordable (for about a month before the lack of supply just pushes house prices up even further), but its got to be everyone there and really has no merits on the relationship between the real value of the workforce – heck pilots don’t get paid as much as train drivers, and many of those folks don’t get to go home every day!

            @Dude you need other folks on the trains often anyway to catch the folks without tickets, help the disabled person on and off, etc – but on a driverless train you only need that ‘conductor’ that deals directly with the passengers. Docklands Light railway for instance has been ticking along quite nicely for decades now.

          5. This x1000.

            Nothing wrong with fighting for workers rights however when technology goes “dark”, and driverless-trains are curiously absent from discussion, then either it is so cost-prohibitive or something simmering on the back-burner.

            The cost-prohibitive part might be true but trains should and are the easiest mode of transport to fully automate (i.e. highly controlled environment, limited D.o.F etc.).

            It is easy to forget that autopilot-equipped aircraft are rated for modes so although they technically have more D.o.F than a vehicle they typically fly in regulated patterns and the captain always retains full authority and responsibility for checking the aircraft.

            Driver-less cars seem to have skipped some key points in attempts to get to market. It all strikes me as dangerously close to the old joke about the person who crashed their motorhome after they put it into cruise control and went for a cup of tea/coffee/comfort break…

          6. >but on a driverless train you only need that ‘conductor’ that deals directly with the passengers.

            Yes, assuming nothing goes wrong and/or technical personnel can arrive in minutes from system failure. When you have a train going 100 miles in the middle of nowhere, you need someone who can troubleshoot and solve technical problems and knows how the train works. That would be the engineer who drives the train.

            Having a driver who can solve issues makes the difference of being 15 minutes late and jamming the whole track for the day. The automation doesn’t make these workers go away – you still need people at the back office monitoring the automated system, and people on dispatch to rush out to unstuck the system, which hardly saves any money.

          7. >Yes, assuming nothing goes wrong
            On the docklands for instance the ‘conductor’ role was initially at least effectively a backup driver, the train can (or at least the early ones could) be manually controlled if it needed to be. Though as far as I am aware that has never actually happened in service, and this is for a train system that has been running driverless for 30 odd years…

            If the actual drive system goes wrong doesn’t matter if you have a driver or not the train isn’t going anywhere as that is a problem for the engineer. But when all the controls to these things have been computerised fly by wire forever and the ‘conductor’ can poke the computer to do stuff if its ever required actually running the train has to be cheaper, and hasn’t added much if anything in the way of points of failure – one shaved off here in bored inattentive humans making mistakes, one added there perhaps as reliance on a network connection type thing all the way down. So judging by how reliable Docklands seems to have been despite being pretty ancient I’d say it seems like it works out rather more reliable and cheaper to run… At least if you can get manage to get rid of the union staff and actually implement such a system on a historicly human driven line without a riot.

          8. Paying a driver 1000 people a day at UK ticket prices is not a major issue.

            In fact drivers are so cheap compared to installing driverless systems, that unless a line is designed for driverless from the beginning, it’s simply not worth even trying.

            But the right wing media will love to claim that drivers are overpaid, despite the responsibilities they have to their passengers, and also for being the person in control if something bad happens, for which a self-driving system will be absolutely useless.

        2. People way overestimate how smart plane autopilot systems are. They’re “keep the plane level, on course, at the right altitude and at the right speed” control systems, with nowadays some smart updates to course, speed and altitude based on the navigation settings built in. They are in no way a replacement for the pilot and will happily fly a plane into terrain, into other aircraft, straight into a cyclone, etc etc. Even “autoland” systems aren’t nearly as smart as people think and require a lot of set-up to get right and lots of ground support equipment to make them work.

          In short aircraft autopilots rely on a LOT of external infrastructure not to prang into anything and on the skies being comparatively empty.
          Rail automation has been tried and it works for some routes, and doesn’t work for others. Rail is “easy” to automate and turns out it’s really difficult. Road/car automation is on a whole different level of challenge. It would be comparatively easy if we had empty roads where ALL traffic was automated and intercommunicating but that’s not a world that will ever exist. This is also the reason I laugh at all the “flying taxi” boondoggles that say they’ll automate all the problems of heavy traffic density away to create “safety”. It’s a utopia we can’t even manage in 2.5D with everything being in more or less the same plane, let alone if we start adding in 3D traffic in all directions that all has to be managed. Speaking as a (glider)pilot, flying is easy (if there’s no other air traffic around). Flying in a gaggle of other planes (like 10 or 15 planes in the same thermal) is mentally straining, but everyone is more or less going in the same direction. Flying a traffic pattern with 10 other planes, some of them gliders, some of them motorized, heading for 2 parallel runways, all with different rates of decent and thus flight paths, is absolutely exhausting and something to avoid if at all possible

  2. First I’ve heard of this, work in Edinburgh and regularly drive the bridges, one think of note here is that busses have exclusive use of the old road bridge, and their own lane seperate from regular traffic for a good chunk of the route if it goes how I expect it does. So it’s less likely to encounter the kind of driving one does on the public roads. So question is, is this system good because it’s good, or because it’s given the red carpet treatment.

    1. >because it’s given the red carpet treatment.

      Here’s one thing that puzzles me: buses are said to reduce road congestion and the the need for road infrastructure because they take up less of the road for the same number of passengers. But – when the city builds dedicated bus lanes all over the place – that’s not actually true, is it? That’s less efficient use of the infrastructure just to accommodate the bus routes and makes the whole point moot.

      If the bus lane was open to all traffic, well, you’d have twice the space for cars and those who drive the bus still wouldn’t drive, because the main reason they’re on the bus – with what the fare prices are and how slow and sporadic the service is anyhow – is because they’re subsidized to be there: school children, students, pensioners… people who simply don’t own cars.

      1. A well-used bus lane can move about 4 buses a minute, 80 people per bus, that’s 320 people a minute. There is no way two lanes of car traffic can move 320 people a minute, with 1.3ppl/car on average. One lane would fit about 20 cars, is about 30 people a minute. Less than a tenth.

          1. Single route probably not, but that one buslane likely serve many routes for a while as they branch in and out, all running something between 1 and 10 bus an hour. Doesn’t take that many routes running at short interval to get to counting buses passing in a min rather than mins between buses.

          2. Foldi:
            Still unlikely to justify a lane _anywhere_ with the possible exception of right next to the depot/garage. The bastards are just deliberately making driving worse.

          3. HaHa:
            Lots more places than just those – geography and historic building frequently pushes roads into more hourglass branched type shapes where lots of routes that have nothing to do with each other will be funnelled though common points.

            And as in many places they are also treated as cycle lane so its something of a benefit to them as well as all the traffic that would through blindness or maliciousness end up hitting the cyclist if they shared some parts of the route… Once you have that hit and run or immediately apologetic driver while waiting on the paramedic you end up with way more and very widespread congestion…

            Not going to say everywhere should be on, but there are many places it should actually help traffic flow even for the personal cars, and more than enough others where its a major benefit for the public transport networks reliability and has little impact on other road users, so overall is still a good thing.

        1. That’s not the point. The bus doesn’t really remove cars from the road in the first place – it mainly moves people who can’t drive for various reasons. Closing off the bus lane for the rest of the traffic just causes more congestion.

          1. Untrue. Check out The Netherlands, Germany, Switzerland, and a host of others. If you live in one of the cities around Amsterdam, but work in Amsterdam, it’s almost always either just as fast, or faster to go by bus and/or train, than to go by car.
            This is *because* of dedicated public transport lanes. The tram line is always shared by buses here, not a minute goes by without a bus, taxi or tram zooming past.
            If everyone who uses the bus and tram here would drive a car, that would mean insta-gridlock.

          2. False. Paradoxically it’s quite often the case that adding more lanes actually CAUSES congestion. Very often having more lanes just widens the traffic jam but does nothing to shorten travel times. Bus lanes are not usually added nilly-willy and thought is (usually) put in to where and when to put them in. Removing lanes from some roads can actually ease congestion by forcing traffic over better flowing routes. Busses are also a major obstacle in slow moving traffic and removing them from the rest of traffic can ease congestion for other road vehicles too, meaning that even with the bus lane removed from other traffic, this still moves better than it would with 2 lanes available.

            See also: The Braess paradox https://en.wikipedia.org/wiki/Braess%27s_paradox
            and: https://www.strongtowns.org/journal/2019/4/18/mr-go

          3. > This is *because* of dedicated public transport lanes.

            The only reason to have dedicated bus lanes is when there are so many stops that it would obstruct other traffic. We have a 5 mile long bus lane with no stops, along an interconnecting road that changes a double lane motorway into a single lane road which becomes congested while the bus lane remains empty. We wouldn’t have the rush hour gridlock that makes people (and buses) late if both lanes were available for use – it’s just a deliberate choice to inconvenience driving cars.

      2. Bus lane are also usually it seems a cycle lane and taxi lane – and when even the taxi is more often than not going to be better efficiency wise than private cars – no more driving around for 20 mins looking for a space, getting lost when the GPS is out of date or throws a wobbly in the concrete canyons, no more awkward multi story moments where the parked car and incoming seeking car clog the whole works up for several mins, and direct destination to destination travel. The taxi won’t always be more efficient of course, as how you use them also matters – if you are summoning a taxi by phone to the arse end of nowhere for a pickup to then take you equally out into the nothing else here club miles away for instance…

        I don’t think every location you could put a bus lane should have one, there is definitely going to be cases where the whole road network would flow better if all traffic was using all the road, etc. But in the right place a buslane is a good thing for all the traffic involved – like the cases where it allows the bus to easily get through the junction or roundabout that is frequently too narrowed by congestion to easily navigate such a giant vehicle without. Having a bus (or 5) stuck unable to really make progress just jams everything else around them up then.

          1. Won’t get me arguing against separate cycle paths – but its still a great deal nicer to be in the bus lane where you don’t have to be nearly as paranoid about the hordes of blind or actively malicious drivers – so much less for the cyclist to keep track of.

            And personal car’s being prioritised is equally privileged to the richer folks who can actually afford to buy the car, maintain it, get whatever the local equivalent of the MOT is, pay for the fuel, any taxes related to it, and to park it wherever you are going (that one around here can easily cost more for a paltry few hours than the taxi would) – running a car can easily end up more expensive than using a taxi frequently, and in the bigger commuter cities this is taxi from the train station for that last mile – a cost that can’t easily be compared anyway.

          2. Bike lanes need cars running in them to sweep out the broken glass.

            That or cyclists will have to cut down on TourdeFrog LARPing and use sensible street tires. Like the Dutch do. I bet a typical American Lycra clad shellhead would be laughed at by the Dutch as they fix their third tire in a km (while some junkie makes off with their CF frame).

          3. I don’t think that is an accurate portrayal of taxi use, in the UK at least. I don’t know a single rich person that uses a taxi unless they absolutely *have* to, or are on a business meeting that they flew to. Mostly I see people getting off trains using it, and people who can’t afford a car using them for their weekly shop at the supermarket.

  3. >Buses on the AB1 route will stop at every bus stop

    Which is one of the annoying feature about buses. At the busiest hour, the bus goes slower than my bicycle because it has to stop every half a mile – but off-peak it goes twice as fast because it skips the empty stops.

    A bus needs an average ridership between about 5-7 passengers to break even vs. cars in terms of efficiency. This it achieves mostly by getting packed full during the busy part of the day, to offset the fact that it’s traveling almost empty 80% of the time otherwise. Making the bus stop & go when there’s few passengers on-board, when the stops are empty as well, is a huge efficiency penalty that makes operating a bus line pointless.

  4. So Scotland’s the land of stereotypes?

    I only was there once, summer of 1965. Really hot in Edinburgh. I think I found that money there. And watched some of Road to Utopia.

    Apparently my ancestors there were in the slave trade.

        1. How long before someone hacks one and takes it for a joyride. I wonder what happens when one of those crashes and kills someone.
          Do you blame the owner or the software company.

    1. There building trams in Edinburgh. To go on the roads the busses already go on. Because the trams are on rails and move real slow it would seem to be the ideal candidate for automation. But no, apparently the busses want automated. The bus which seems to cover a route the train from Fife seems to already cover. All mocking aside Edinburgh’s public transport is more organised than Glasgow, and there all so expensive it’s cheaper to drive.

    2. Yes, trains, the least convenient form of transport. Goes nowhere near where you want it to go, costs as much as driving in most cases, and requires expensive dedicated infrastructure and hauls empty carriages off-peak. Good call.

      I’m not against trains, but they almost never go where I want, require other forms of transport at either end and during peak/offpeak times suffer the usual public transport problem – packed to a sweaty uncomfy standing mess/empty and environmentally awful (which averages out OK, but still is nonsensical).

          1. Ah, another person who has never lived in Japan.

            That is invariably the last train of the night. Everyone has been out drinking (because they don’t have to drive) and they know exactly what the train will be like. A small price to pay for not dying in a head-on collision with another drunk driver.

            Trains are objectively the best.

        1. I’m afraid you seem to be blind to the difference between a society where trains were deprioritised throughout the 20th century and other infrastructure grew up around it, vs one where they have promoted it from the beginning as the primary transport means. They’re not the same starting point and just adding more doesn’t work. And regardless, I can’t take my sports equipment on a train so I’m always going to own a car, and then it’s always cheaper and faster to drive if you’ve already factored in the fixed costs for other uses. Even in the Netherlands, it’s very rare that it’s cheaper or faster for me to take a train than drive because it’s a 3km walk/cycle to a train, by the time I’ve walked it I can have driven to the destination.

      1. Steel wheels on a steel track = low friction. Electric drive allows for regenerative braking. Electric drive also has the advantage of potentially zero emissions to power the thing. If the train/tram doesn’t go where you want, blame the cartel of auto makers and oil producers. Los Angeles is the classic example – they ripped out the trams (“streetcars”) in the 1950s to build road. What followed was more than half a century of smog and gridlock. I hear their new metro is (slowly) undoing some of the damage. https://en.wikipedia.org/wiki/General_Motors_streetcar_conspiracy

  5. I don’t get it. They have the not-really-driver AND the conductor onboard, and still they can’t look out the window and see if anyone is hailing the bus?

    Besides, I think that detecting a hailing person is a small, and quite “easy” CV task compared to the whole self-driving thing.

  6. ““You wouldn’t know the difference between this and a normal bus from the driving.” Given the concerns around the sometimes-erratic nature of certain “self-driving” systems, that’s almost a glowing endorsement.”

    This is not a glowing endorsement. This is an acceptance that bus drivers are almost universally awful and, in Scotland at least, the buses are generally knackered and the roads they run on are knackered. They use the brakes like a binary system, they corner like I would in a sports car causing you to merge seats with a neighbour and the drivers generally grunt at you like an ape. The number of times I’ve stood at a bus stop and had to call a taxi because the bus didn’t come (with no warning on their app) is getting to be a joke.

    Add that to the fact that the Scottish road network is basically a set of cart tracks now through decades of underinvestment – I lost a camera that was stored in my hand luggage, not from the flight, but from the bus journey home vibrating a screw loose internally.

    Would I use an autonomous bus – sure. It’s driving probably couldn’t be worse.

  7. A few thoughts, having worked with driverless and assisted cars and trucks before:

    Need for (a) driver and (b) conducter:
    This is obviously (a) a safety and regulatory thing and (b) to ease the people into using it and being able to ask questions and be involved. Both of these are good in an introduction phase. If it pases this phase and prooves safe enough people will be used to going wihtout a dedicated driver (as they are with e.g. Docklands Light Railway). Regulations and safety rules that can accept a remote takeover or someone close by to go to the bus fast in cases needed is maybe a bit more difficult to realise but crutial to have real running cost benefits which would in turn allow for more remote/frequent/nightly public services. Otherwise it is basically an assitant that can minimise human error, *maybe* allowing for less skilled / cheaper personal.
    Another thing is ticket control, but I am a fan of free public transport anyway and if needed you could do sporadic controls.

    Bus vs Car automation:
    Buses usually go on more defined routes, so there is way less edge-cases to handle making the system easier to verify. The complexity otherwise is roughly the same if you want to automate a bus vs a car. But as the cost of a bus is higher anyway it is more reasonable to afford higher quality sensors. 4 Lidars are easier to justify on a bus than on a car, where you, just from a cost perspective, want to use as cheap sensors as possible (see Tesla’s relience on cameras)

    Hailing:
    Especially at times where not every stop needs to be serviced it would be nice and morew efficient to have another system than stoping everywhere. A local button in additon to an app-call would be nice, though every need for infrastructure (even just power and a radio) might be prohibitive in remote areas.

    1. It’s kinda like self-checkouts at stores. There’s still many edge cases like discount stickers, age restricted products, malfunctions, people messing around, people not knowing how to use the thing… and these problems don’t go away over time. There will always be more idiots, more people trying to sneak alcohol past the register and steal stuff etc. so the self-checkout will never be fully automated. There will always be a cashier standing there doing nothing most of the time.

      1. The key is then to not have one person per instance (bus / busstop / self-checkout) as before but one per “group” that is reasonably available if needed.

        Thinking of the self-checkouts, maybe it’s also nice to be able to signal in advance that help is needed so the flow is better. If I could signal that I need an age check because I am buying alcohol so that someone is ready when I’m done would be so much nicer than waiting til somone realises the yellow blinking light after I’m done scanning…

      2. “There will always be a cashier standing there doing nothing most of the time.”

        You’re not a regular Walmart customer.
        Their self checkout “associates” are very busy, often I have to wait until they have finished resetting a different register, before they can reset the one I’m using.

  8. I’m much much much more concerned for the safety of the people outside self-driving buses than the ones inside of them. (But that assumes an absence of deep cliffs in the routes I have to say, and Scotland has some cliffs I’m sure, but probably not on the routes of these buses.)

  9. Fortunately, the people who write the algorithms and the software code to implement an autonomous train,bus, car, etc are renowned for their ability to predict every possible circumstance, in advance, and deal with it.
    Computers and their software are infalible .. everyone knows that!

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.