Researchers suggest cyclists could wear smart glasses to communicate with self-driving cars — automated vehicles "need to learn the language of cyclists"

Rich_cb wrote:

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.

What is it about driverless vehicles that makes this suddenly such a great option compared with sharing a cab?

Steve K wrote:

What is it about driverless vehicles that makes this suddenly such a great option compared with sharing a cab?

With AVs, we can maybe rethink what kind of form factor is best for transport.

Traditional cabs are designed around needing to have a driver and so it makes sense to try to transport multiple people with using only one driver. Autonomous cabs shouldn't need to replicate that design, so I think that single person vehicles will become popular as certainly there's a lot of single occupancy cars on the roads.

There's also the ownership of the vehicles to consider. If people are going to purchase and own a vehicle, then it makes sense to consider the maximum number of people you want to transport, even if that's a rare occurrence. If you're hiring a vehicle on a single trip basis, then you just want the minimum size available.

The cost and the privacy/safety of your own compartment.

Most people often share taxis with friends, most don't do so with strangers. In most cities you'll often be traveling in the same direction as large numbers of other people so there is scope to reduce both costs and congestion significantly.

Rich_cb wrote:

The cost and the privacy/safety of your own compartment. Most people often share taxis with friends, most don't do so with strangers. In most cities you'll often be traveling in the same direction as large numbers of other people so there is scope to reduce both costs and congestion significantly.

Is anyone actually designing/producing these individually compartmentalised vehicles you envisage?

The first commercial operation of a bespoke driverless taxi will begin later this year.

Both Cruise and Zoox have gone for a very similar design which can easily be compartmentalised.

I imagine they'll start with one space and then as the commercial operation increases they'll compartmentalise to maximise revenue per car.

Rich_cb wrote:

 If we can get average car occupancy to 2(!) .

an incredible amount of heavy lifting there for the superhero letters I and F

It's a huge leap I'd agree but I actually think it's achievable.

The difference on our streets would be profound.

Rich_cb wrote:

It's a huge leap I'd agree but I actually think it's achievable. The difference on our streets would be profound.

maybe the biggest improvement in our street from autonomous shared cars would be the removal of on street parking. If you don't have somewhere on your own property to store a car, then self driving share cars are your solution.

Allowing all the space annexed by drivers for long term storage to be used for either imrpoved pedestrian environment or less restricted thoroughfare would be great.

It would be transformational. I suspect that when/if widespread on street parking disappears it will be a bit like the indoor smoking ban, we'll wonder why on earth we put up with it previously.

Rich_cb wrote:

Why wouldn't people use that service?

Car sharing has great potential, even more for fully autonomous ones since the user bad driving risk is eliminated. But there are two very important reason that they will need lots of time to succeed.

The first is cost. A twenty year old car can work perfectly with a little care and you can purchase it used for very little money. Regarding car sharing running costs, I have seen examples of scooter sharing services that their running cost per mile is comparable to that of a car (excluding purchase cost/tax/insurance). If you scale it up to car scale, I believe it will be very expensive. This can only be "solved" if the car tax is raised so much that some will think twice before having a private car. But this is not a proper solution.

The second problem is that people like their stuff. Some people including me like to have their own bicycle, either because of special customizing either of sentimental value. The Dutch having shitty bicycle for commuting that don't care if stolen despite their wealth. This shows that this can be overcomed, but for me it seems a little difficult. The same applies to cars.

I think the Dutch example is good.

People often use their transport as a status symbol or extension of their personality but, given the right circumstances, will opt for something entirely practical.

I think autonomous cars will fall into this category, people will originally use them instead of taxis, then they'll ditch their second cars and use them as back up to their main vehicle, then they'll transition away from private vehicles altogether.

Owning a car is so expensive for most people they'll prefer to spend their money elsewhere once they have an alternative.

This "Dutch example" isn't 100% correct. I have a nice shiny new commuter bike, but also have my "beer fiets". A real heap of shit that I won't worry about if I can't remember where I left it.

Rich_cb wrote:

Owning a car is so expensive for most people they'll prefer to spend their money elsewhere once they have an alternative.

I think a lot of people are in denial about how much running their car really costs them.

marmotte27 wrote:

Individual motorized transport is an inefficient waste of public space, valuable ressources and energy, and we have to get rid of it asap. Does any of that change with autonomous cars? No.

Depends on how big the vehicle is. Mopeds can be quite efficient, though they can spew out a bit of pollution. The trick is to use the maximum percentage of the energy to move the person and minimise the energy to move the motor and vehicle. That's why I think e-scooters are a great solution for short trips around cities.

I think there's yet again a urban / low density divide here and I'm not sure about the latter so I'll ignore that*.  But for urban areas I think there's a great potential for synergy between the development of infra for mass cycling and e-scooters.  I suspect infra holds these back in the UK.  That would certainly be fixed by Dutch-quality infra with smooth pothole-free surfaces and fewer sudden kerbs to negociate.

In fact I wonder whether bikes will end up remaining a minor fraction of transport in UK urban areas?  E-scooters have the comfort / convenience (just stand there), I think are relatively cheap to power and fix some other private transport issues.  Security (just carry it), transporting the transport (just push / carry it), storage (minimal).  And maintenance would seem a positive also - they've not got many parts which stick out and I imagine like Dutch bikes and cars if they don't work you just take them to a shop (carry them)!

I'm not a huge fan of them - I think bikes are healthier and more sustainable.  I've seen the odd problematic interactions with pedestrians in the UK.  I'm not sure what the safety looks like especially if there are large numbers.  But it's quite easy to imagine shorter urban journeys being dominated by this mode - if there are flat surfaces not shared with large motor vehicles.

* Part of the issue is that cars accellerated the centralisation of jobs and amenities.  We've also fixed it for some to have the best of both e.g. reside in the quiet countryside but be hypermobile and access the amenities and money in the cities.

Hi Steve

I'm glad someone is looking at how autonomous cars are to communicate their intentions, not just to cyclists but to human drivers as well.

Driving or riding in any town requires subtle co-operation using visual cues including road speed, change of speed, steering angle, road position, hand signals, head movement etc. Also where possible helping other users like LGVs, buses and preventing deadlock with oncoming traffic e.g. outside schools.

Are we only now thinking about how an autonomous car might conduct this dance? Is there a plan, or will it be the usual mantra responsible for SUVs -  'Sell it and stuff other road users' ?

And for goodness sake, smart glasses are a non-starter. Do you really expect pedestrians, horse riders and young kids on bikes to use them?

I'd love to believe that our government with its excellent record of informed decsion-making (/s) will ban autonomous cars from towns and cities until workable answers are found.

hi jaysa - the communication is really interesting and often subtle. AVs need to understand the dance to fit in to human traffic. Right now, most work has looked at AVs and pedestrians at crossings. That is relatively easy compared to bikes, escooters, ... As i ride everyday, I wanted to make sure we cyclists werent forgotten about.

The smart glasses are one example that got picked up in the artice. To be clear, we dont expect riders to have anything; should be all on the AVs. But if you do have some tech, you could use it if it helps you. Might be a bike computer or a phone mounted on the handle bars. In some futures, augmented reality displays are predicted to be as common as phones, so we wanted to think how they might be used too. 

Congratulations on getting your paper published, I know how much work goes in to that kind of thing.

The eyes of the driver mean nothing.

It's the wheels you look at.

Actually, it really depends on the traffic scenario as to where people look. We got a bunch of cyclists to ride wearing eyetracking glasses so we could see where they looked. In controlled intersections they hardly looked at other vehicles at all, just the traffic lights. In bottleneck scenarios they stared straight into the eyes of the driver approaching them! There were loads of different behaviours that we saw.

My bicycles already carry visual light reflectors for the benefit of my visibility to other road users. If the AV manufacturers agreed a standard and made them freely available I could be persuaded to attach a discreet lidar / radar or other unpowered device which increases my visibility to the AV.

Ideally I expect that no AV system will be licensed for use on our roads until it meets stringent safety standards with respect to pedestrians, cyclists, horse riders and other vulnerable road users. However, I am generally supportive of the technology and appreciate that nothing arrives in a fully mature state.

A bit like sailing boats having a radar reflector fitted to the mast. But if a 10,000kg yacht needs a radar reflector the size of a 2 litre bottle, what would 100kg of bike and rider need?

Probably because it's been drummed into cyclists to make eye contact.

Pretty much it's a fallicy.

There are many on this site who can tell stories of making eye contact in an instant before being run over.

Or the motorist "looking right through them"

The wheels will tell you so much more, where the car is going, if it's moving.

There is one reason to make eye contact, it's so that if they do run you down and kill you, they will have something to remember when the wake up from their nightmares.

Except for the waste of all existing rear lights which will not have that benefit. 

All lights will be replaced at some point, no need for any to be wasted.

Just run the risk of being run over by an AV until your existing light dies?

If this were to ever become 'a thing' I imagine we'd have plenty of warning.

Given that the current market leaders in autonomy aren't relying on this sort of thing at all I doubt it will be required anytime soon.

We surveyed a whole load of cyclists in another paper to find out thoughts about technology use. One thing was that almost everyone carried their phone when riding, so you could potentially use that for communicating with other vehicles. But, as above we dont want to rely on that. The AV should have the tech required to be safe around riders.