Specialists from the University of Glasgow believe self-driving cars "need to learn the language of cyclists", with their research suggesting such improvements are necessary to help autonomous vehicles safely share the roads with those riding bicycles.
In a paper titled 'Keep it Real: Investigating Driver-Cyclist Interaction in Real-World Traffic', which will be published later in 2023 and was today reported by The Herald newspaper, researchers looked to unpick the relationship between cyclists and automated vehicles, saying there had been "comparatively little" research into how self-driving technology can keep cyclists safe.
Professor Stephen Brewster of the university's School of Computing Science said there had been "a lot of research in recent years on building safety features into autonomous vehicles to help keep pedestrians safe", something that needs to be repeated with cyclists.
> "These are completely safe autonomous vehicles": Cyclist spots driverless car using cycle lane
"Cars and bikes share the same spaces on the roads, which can be dangerous – between 2015 and 2020, 84 per cent of fatal bike accidents involved a motor vehicle, and there were more than 11,000 collisions," he said.
"There has been a lot of research in recent years on building safety features into autonomous vehicles to help keep pedestrians safe, but comparatively little on how automated vehicles can safely share the road with cyclists.
"That's a cause for concern as automated vehicles become more commonplace on the roads. While pedestrians tend to meet automated vehicles in highly controlled situations like road crossings, cyclists ride alongside cars for prolonged periods and rely on two-way interactions with drivers to determine each other's intentions.
"It's a much more complicated set of behaviours, which makes it a big challenge for future generations of automated vehicles to tackle. Currently, self-driving cars offer very little direct feedback to cyclists to help them make critically important decisions like whether it's safe to overtake or to switch lanes. Adding any guesswork to the delicate negotiations between car and bike has the potential to make the roads less safe."
> Tesla car in Full Self-Driving Beta almost rams cyclist
Brewster's team studied the ways drivers and cyclists directly and indirectly communicate in real-world situations. From the research they have formed recommendations for future generations of automated vehicles.
The researchers suggest the vehicles' intentions could be displayed on their exteriors, for example displaying animations signalling intention to speed up, slow down, give way or manoeuvre.
At the other end of the relationship they suggest cyclists could wear 'smart glasses' communicating the vehicle's intentions to them, for example when coloured LEDs on the car light up to signal right of way is up for negotiation a vibration could be sent to the glasses as a non-verbal message.
> Tesla using Full Self-Driving Beta crashes into cycle lane bollard...weeks after Elon Musk's zero collisions claim
The paper's co-author, Ammar Al-Taie, said he hopes the research will inform autonomous vehicle designers, encouraging them to develop "new ways that self-driving cars can work safely alongside cyclists by speaking their language".
"Just like spoken languages, communication between cyclists and drivers varies from country to country. We're very conscious that this paper focuses specifically on UK roads – any future developments will need to take into account the differences in drivers' and cyclists' interactions across the world."
The research will be presented, at the ACM Conference on Human factors in Computing Systems, in Germany next week.
Add new comment
105 comments
Own door? Given vehicles seem to be getting bigger sounds like self-driving apartments would be the way to go. Everyone's got problems with their WiFi? Just send the flats across town! Town centres would be delighted by sudden surges in passing trade. People could do "appartment-pool karaoke" videos with a real choir! I'm sure some would find it as enjoyable as a cruise ship. Better than Windows Spotlight! "oh look, it's Bognor Regis out there! I think I'll have an afternoon out."
Should be easy politically, just establish a supporter or two in every apparent block - hey presto, you can get the vote out!
Are you trying to re-invent trains?
They are more of a reinvention of buses.
I think of autonomous cars like miniature buses without fixed time tables or routes.
If we can get average car occupancy to 2(!) then we can almost halve the number of cars on our roads. Add to that the huge amount of street space freed up from a lack of parked vehicles and you have a far more pleasant environment.
Maybe we could call them minibuses.
I was thinking smaller than that so maybe microbuses would be more accurate?
Jeepney / Songthaew?
(I suspect the current ones aren't exactly "green transport" - unless they're painted that way...)
That would certainly be a major benefit. Depending on what governments subsidised / prioritise I guess we'll initially see a mixture of a few of these things - possibly at the "luxury / novelty" end / maybe in the biggest and densest urban areas, a continuation of the likes of Uber and maybe a growth in car share schemes - where you get your "own" car. People are currently accustomed to that. (These are well established in some places).
OTOH I'm not certain we'll see much change in the UK driving environment at all any time soon.
The three leading cities for autonomous vehicles (outside of China) are all in the US which is about as car centric a culture as you can get.
I would be very surprised if they weren't a common sight in UK cities by the end of this decade.
I expect they'll initially displace taxis, then buses and second cars and eventually almost all passenger vehicles.
I can't disagree with your final paragraph, but I just don't see why self-driving vehicles will suddenly make sharing attractive (but that may just be my lack of imagination).
I can see cases where the reverse might happen. Going out for some drinks? No need for one person to be the designated driver and give everyone else a lift; you can all arrive in your own self-driving vehicle.
If AVs become commonplace, then I think it'll lead to increased traffic. Not only will there be the journeys that people currently make, but there'll be the journeys for the vehicles to get to and from the people and charging points.
We will all be locked down in 15 minute neighbour cells by then - no need for any vehicles!
It will be a lot cheaper to share and, if you have a private compartment, there will be almost no downside to sharing.
In that situation I can't see why sharing wouldn't flourish.
In your example that may lead to an increased number of cars but each group member could share part or all of their journeys to the pub and back with people heading in the same direction so overall there'd still be a decrease compared to the status quo.
If we're going to have private compartments, then wouldn't it be better to split apart the vehicle so that it takes up less space? It'd also allow the compartments to travel in different directions. Maybe the ideal solution would be a vehicle per person?
I've just realised that the picture shows some high body-fat people, so maybe a better bet would be to include an exercise device so that people can exercise whilst being transported somewhere?
Now, all we have to do is come up with a design of a cheap vehicle that can transport a person wherever they want to go and also provides them with some healthy exercise whilst doing it. Ideally it'd be a vehicle that doesn't pollute as well.
You might be on to something there!
Maybe a vehicle, I don't know, powered by its passenger?
Is such a thing even possible?
There may be some scope for single person vehicles but it will always be considerably cheaper to share a vehicle so I think most people will opt for that with private compartment etc.
So, what we need are vehicles that can carry a large number of people at the same time for the ultimate in cheap transport. Maybe have a fleet of such vehicles that travel common, busy routes and pick-up/drop-off people at known points. To make it even more efficient, we could designate specific lanes for those shared vehicles so that they don't get held up so much by the non-shared vehicles.
I'm guessing you're being facetious but if not you're not really comparing like for like.
A bus isn't a door to door service. You don't have any privacy either. With an autonomous car you get 90% of the environmental and cost benefit of a bus with none of the downsides.
Yes, I was being facetious.
I'm not convinced about the environmental/cost benefit of having an extremely heavy (batteries innit?) vehicle traversing all the minor roads in an attempt to drop people off at their front doors. It'd be better to keep the heavy shared EV on the main roads and have people walk from their LTN to the heavy, shared EV-stop.
Ultimately, if we stick with car shaped vehicles then we'll continue to experience car shaped problems such as congestion and increased pollution (tyres and brakes) due to the extra weight of EVs. What we should be doing is getting as many journeys as possible onto e-scooters, bikes (electric or acoustic) and shoes. By using a smaller vehicle, the batteries can be vastly reduced and of course, a large part of EV batteries are used to carry the EV batteries around.
Congestion can be solved by increasing occupancy and all the other problems you mention will be decreased as well. Brake pad usage in an autonomous EV should be minimal as all planned deceleration can be done using regenerative braking.
It will be politically impossible to get most people to abandon their cars in favour of walking to the bus stop to take a crowded bus in the general direction of their destination and then walk some more.
That option already exists and the vast majority of people choose not to take it.
Small shared autonomous public transport brings almost all of the benefits of large buses but without the barriers that dissuade people from using large buses.
The latest EV batteries announced have nearly double the energy density of the current market leaders so EV weight should drop significantly by the time these vehicles are commonplace in the UK.
However, if you're aiming to use AVs for door-to-door trips, then that's going to cause congestion as you have a large shared vehicle going down smaller streets, stopping for a while and then moving on.
I'm not convinced that having a driverless bus really changes the geometry of the streets and vehicles.
EV batteries may get better, but there's going to be a major problem with trying to get enough lithium to replace all the ICE vehicles. The better solution is to scale down the vehicles so that they don't need to use so many batteries - that can and is being done currently.
They've also developed lithium free batteries (using sodium instead) so that also shouldn't hold us back. As people get used to EVs I do think a market will develop for smaller batteries but currently people want as big a range as possible.
Dropping people at their doors will be no different to how the roads are used now with the exception that there won't be anywhere near as many parked cars so congestion on residential streets will be dramatically improved.
There's a lot of work to be done with sodium-ion batteries in terms of energy density, so it seems unlikely that lithium-ion will be replaced any time soon.
Getting rid of all the parked vehicles would certainly be great, but it relies on people not wanting to own their own transport.
Sodium ion is where LFP was just a few years ago so it's certainly not implausible that it will achieve the required density to be the sole battery in an EV. Even with the lower density it has useful enough characteristics to be used alongside Li-ion which would still reduce the overall amount of lithium required.
I honestly think that once autonomous cars are widely available private car ownership will plummet. The average car only does 6000 miles a year, at the bottom end of that bell curve there are a huge number of cars which barely move, they'll be first to go.
sounded low to me so I did some google fu, and first result was this
which of course left me no wiser but it gets better still
all from the same page, baffling.
another source gave the following
which I found a little more credible
The latest MOT data is for 2021 and shows significantly less than 6000 on average.
Given that 2021 was still COVIDed slightly, it seems reasonable to use 6000.
highly desireable, but of course the real reason why residential streets have become so congested with parked cars, is that adults are no longer able to afford to leave the family home, cost of housing is the biggest problem we have, it affects many other things.
but the landowning classes benefit, so I doubt much will change. And of course if anyone want to byuild new houses, the boomers are out with their banners "save our village" Maybe I'll get some made up "no houses for our grandchildren" and put them up next to them.
So who is buying the houses then?
Genuine question - at least round me in Edinburgh there's a major expansion going on - both driven by council and also lots of private developments. (Another concern is - we're building on former industrial (good) / farming (hmm) sites here - but what jobs are these people going to do? We also don't appear to be building new amenities or even cycle paths...)
Pages