Reversal of priority on pelican crossings

Discussion about street lighting, road signs, traffic signals - and all other street furniture - goes here.

Moderator: Site Management Team

yen_powell
Member
Posts: 179
Joined: Wed Feb 01, 2017 16:17

Re: Reversal of priority on pelican crossings

Post by yen_powell »

I once had a letter from a man objecting to me converting 3 zebra crossings into Pelican crossings. His argument was that I had taken his rights as a pedestrian away because where as once he could step out on the crossing and the vehicles should legally stop, now he had to press a button and wait till the red light stopped them. He demanded that we rig them so that they remained on a red signal/green man normally and any driver wishing to pass would have to wind down their window and press a button and wait to get a green light/red man before proceeding.

I laughed back then, but it seems he was years ahead of his time.
Craig91
Member
Posts: 11
Joined: Mon Apr 05, 2021 10:00

Re: Reversal of priority on pelican crossings

Post by Craig91 »

traffic-light-man wrote: Fri May 28, 2021 18:56 As I've mentioned recently in another similar thread, Liverpool has a dual Puffin that rests in pedestrian green, and it's been like that since around 2006 when it was installed.
Forgive me as I see you’ve mentioned this before, but where is said crossing?
User avatar
traffic-light-man
Member
Posts: 4728
Joined: Fri Sep 01, 2006 18:45
Location: Liverpool, UK

Re: Reversal of priority on pelican crossings

Post by traffic-light-man »

Craig91 wrote: Fri May 28, 2021 23:01 Forgive me as I see you’ve mentioned this before, but where is said crossing?
It's across the eastern approach/exit to the Liverpool One Bus Station, so perhaps not quite as remarkable given it's buses only. It's also laid out like a dual puffin, but in actual fact, both sides run together.

Edit: I've just uploaded a quick video of it, apologies for the severe flickering from the older Microsense kit. It would perhaps be better demonstrated when there's aren't any buses to keep it ticking around!

Simon
AndyB
SABRE Developer
Posts: 11056
Joined: Tue Dec 26, 2006 21:58
Location: Belfast N Ireland
Contact:

Re: Reversal of priority on pelican crossings

Post by AndyB »

WHBM wrote: Fri May 28, 2021 20:41
Stevie D wrote: Fri May 28, 2021 17:54 The Highway Code is not law, it is guidance. It is not illegal to cross when a red man is showing.
I think I knew that...

Compliance does not mean illegal. But one is expected to comply with the Highway Code.
If you cross against a red man and get hit by a car, it's probably not the driver's fault you crossed dangerously unless there was another reason why their driving was dangerous, but insurance purposes are the limit of it unless you are in Northern Ireland - yes, jaywalking is still on the books in NI, but is rarely invoked as the penalty of a hospital stay is usually enough and there are stronger penalties for drunkenness in a public place and AOABH (ie pushing someone off a kerb), the usual main associated offences.
AndyB
SABRE Developer
Posts: 11056
Joined: Tue Dec 26, 2006 21:58
Location: Belfast N Ireland
Contact:

Re: Reversal of priority on pelican crossings

Post by AndyB »

However, there are two main postscripts to that post.

1. If you press the button and then cross in accordance with the Green Cross Code, ie you don't force a driver to brake sharply or take evasive action, the police won't take any interest
2. If it is possible or even routine, to safely cross the road using the Green Cross Code between pressing the button and the road traffic lights turning red, there is something badly wrong with the programming - the lights should respond to the traffic clearing far more quickly.
WHBM
Member
Posts: 9706
Joined: Thu Nov 30, 2006 18:01
Location: London

Re: Reversal of priority on pelican crossings

Post by WHBM »

AndyB wrote: Sat May 29, 2021 13:48 If it is possible or even routine, to safely cross the road using the Green Cross Code between pressing the button and the road traffic lights turning red, there is something badly wrong with the programming - the lights should respond to the traffic clearing far more quickly.
Unfortunately TfL have gone the other way, and converted a significant number of signals to Fixed Cycle, on the grounds that bicycle detection does not work, so the signals must work round the stages in case there is a bicycle there (although of course normally there isn't).

It will be interesting to see how the Pedestrian Team at TfL, who (according to last night's local TV news) are behind this, square up to the Bicycle Team there who find the lights stuck on red for them.
User avatar
Stevie D
Member
Posts: 8000
Joined: Wed Dec 20, 2006 17:19
Location: Yorkshire

Re: Reversal of priority on pelican crossings

Post by Stevie D »

yen_powell wrote: Fri May 28, 2021 22:18 I once had a letter from a man objecting to me converting 3 zebra crossings into Pelican crossings. His argument was that I had taken his rights as a pedestrian away because where as once he could step out on the crossing and the vehicles should legally stop, now he had to press a button and wait till the red light stopped them. He demanded that we rig them so that they remained on a red signal/green man normally and any driver wishing to pass would have to wind down their window and press a button and wait to get a green light/red man before proceeding.

I laughed back then, but it seems he was years ahead of his time.
It's the attitude rather than the idea. The idea of having pedestrian priority as the default and then giving green time to cars when needed is a perfectly sensible one (but can you imagine the furore if drivers had to pull up to a stop, push a button and then wait for their turn? 🤔 (and no, I'm aware that it isn't completely analogous because of the aforementioned discussion about it not being illegal for a pedestrian to cross without waiting for a green man, but the same wouldn't apply to cars in reverse (although why cars would be reversing over the crossing I don't know))).

But when someone starts frothing and ranting on about "their rights" when they now might have to go to the supreme inconvenience of having to push a button and wait a few seconds while at the same time the crossing is made safer and more accessible, it shows that their logic and thought processes are laughable and so you assume that the idea is as well, even though in this case they may have lucked into a good suggestion but for the wrong reasons.
User avatar
traffic-light-man
Member
Posts: 4728
Joined: Fri Sep 01, 2006 18:45
Location: Liverpool, UK

Re: Reversal of priority on pelican crossings

Post by traffic-light-man »

AndyB wrote: Sat May 29, 2021 13:48 2. If it is possible or even routine, to safely cross the road using the Green Cross Code between pressing the button and the road traffic lights turning red, there is something badly wrong with the programming - the lights should respond to the traffic clearing far more quickly.
I generally find that even at a well validated MOVA site, the three seconds of leaving amber and the two (or so) seconds of the interstage is enough time to have crossed the road, assuming it didn't max out and saw the gap.

I think the issue is more along the lines that it's far too common for crossings to have a high vehicle max green, so by the time the crossing has gapped out, the crossing user has already seen the same gap and crossed anyway.

It's a bit of a simple statement, but I think it covers what a lot of people experience.
WHBM wrote: Sat May 29, 2021 14:51 Unfortunately TfL have gone the other way, and converted a significant number of signals to Fixed Cycle, on the grounds that bicycle detection does not work <snip>
It depends what's used, as there are options that most definitely work - but presumably they have their reasons.

There's also products in development in London that will, presumably, use cycle detection in order to be successful, so it'd be interesting to see what form of detection that uses.
Simon
WHBM
Member
Posts: 9706
Joined: Thu Nov 30, 2006 18:01
Location: London

Re: Reversal of priority on pelican crossings

Post by WHBM »

traffic-light-man wrote: Sat May 29, 2021 21:59 There's also products in development in London that will, presumably, use cycle detection in order to be successful, so it'd be interesting to see what form of detection that uses.
Well we've had traffic signals for a century and bicycles for a lot longer, so as for something still "in development" we shouldn't really be holding our breath.

Incidentally, the pneumatic pressure pad signal detectors which were retired a generation or more ago as old-hat used to handle bicycles (at a time when they were far more prevalent than now of course) no problem.
AndyB
SABRE Developer
Posts: 11056
Joined: Tue Dec 26, 2006 21:58
Location: Belfast N Ireland
Contact:

Re: Reversal of priority on pelican crossings

Post by AndyB »

Cycles aren't an issue if traffic lights rest on green and change to red on demand as at most pedestrian crossings.
traffic-light-man wrote: Sat May 29, 2021 21:59 I generally find that even at a well validated MOVA site, the three seconds of leaving amber and the two (or so) seconds of the interstage is enough time to have crossed the road, assuming it didn't max out and saw the gap.

I think the issue is more along the lines that it's far too common for crossings to have a high vehicle max green, so by the time the crossing has gapped out, the crossing user has already seen the same gap and crossed anyway.

It's a bit of a simple statement, but I think it covers what a lot of people experience.
There's something psychological about seeing the leaving the amber which means "Oh good, it's changing now, I can wait for this."

In my view, the amber should come up as the last vehicle reaches one car length past the last MOVA loop or the infrared detector. They clear the crossing as it turns red. No inconvenience to anyone.

You're correct about the max green. If it's not acceptable for motor vehicles to have to wait a long time at traffic lights for a cycle to complete, then it shouldn't be acceptable for pedestrians to have to wait that long.
Fenlander
Member
Posts: 7801
Joined: Fri Mar 26, 2010 21:54
Location: south Lincolnshire

Re: Reversal of priority on pelican crossings

Post by Fenlander »

AndyB wrote: Sat May 29, 2021 13:48 2. If it is possible or even routine, to safely cross the road using the Green Cross Code between pressing the button and the road traffic lights turning red, there is something badly wrong with the programming - the lights should respond to the traffic clearing far more quickly.
There’s a crossing in town here that functions like that. It’s halfway along between 2 sets of traffic lights that don’t have pedestrian stages so is the only safe place to cross. The problem is the time between the request and the lights changing is so long that the lights changing at the 2 junctions either side usually creates a gap for the pedestrian to cross first. Eventually the crossing goes free for the pedestrian but they’re long gone and the drivers just sit there wondering what’s going on.
The council won’t adjust the crossing as they say there’s bigger plans to link it with the 2 adjacent junctions, and they require linking to the level crossing just past them so it’s too big a job to do.
User avatar
traffic-light-man
Member
Posts: 4728
Joined: Fri Sep 01, 2006 18:45
Location: Liverpool, UK

Re: Reversal of priority on pelican crossings

Post by traffic-light-man »

WHBM wrote: Sat May 29, 2021 22:11
traffic-light-man wrote: Sat May 29, 2021 21:59 There's also products in development in London that will, presumably, use cycle detection in order to be successful, so it'd be interesting to see what form of detection that uses.
Well we've had traffic signals for a century and bicycles for a lot longer, so as for something still "in development" we shouldn't really be holding our breath.

Incidentally, the pneumatic pressure pad signal detectors which were retired a generation or more ago as old-hat used to handle bicycles (at a time when they were far more prevalent than now of course) no problem.
It isn't the detectors that's in development, it's an adaptive system that caters for all road users. A different product to the one I'm thinking of has been on trial in Manchester too, using video-based AI, which I believe is performing well.

Here's a quick snippet of me activating a surface-based radar as a pedestrian, which gives me faith that they have the ability to pick up cycles of all varieties and speeds. Admittedly, I do think this one is perhaps a little bit too close to the stop line, and in this instance there's also an above ground radar which generally does all of the leg work.

Yes, I can also see an argument of it erroneously detecting crossing pedestrians, but I think that's just something to live with as you'd expect that to only happen as a 'fluke' every once in a while, presuming it was sited suitably. After all, walk at any typical AGD with a reasonable pace and it'll detect you as a vehicle anyway, and we seem quite contempt with throwing those up everywhere.


AndyB wrote: Sun May 30, 2021 09:54 There's something psychological about seeing the leaving the amber which means "Oh good, it's changing now, I can wait for this."

In my view, the amber should come up as the last vehicle reaches one car length past the last MOVA loop or the infrared detector. They clear the crossing as it turns red. No inconvenience to anyone.
At a well set-up MOVA site, the amber should appear as the last vehicle crosses the stop line, similarly to as you describe. Unless it maxes out, of course.

VA sites, which are probably the most common, are generally the ones with the problems, because of the fact that they're essentially doing things in blocks of pre-determined times.

Of course, crossings running on something like SCOOT are even worse, because they 'have to' wait for a window of opportunity to be opened up by the computer before they can change, which is (currently) an entirely vehicle-centric system. It makes no odds as to what gaps in traffic actually exist locally to the crossing, it will just wait until it's told it can change if it needs to.
Simon
AndyB
SABRE Developer
Posts: 11056
Joined: Tue Dec 26, 2006 21:58
Location: Belfast N Ireland
Contact:

Re: Reversal of priority on pelican crossings

Post by AndyB »

There is a lot of money to be made by the engineer who can merge SCOOT and MOVA so that the controller of a junction so equipped makes decisions based on both where there is space to send the next load of vehicles and what's approaching a particiular stop line.
WHBM
Member
Posts: 9706
Joined: Thu Nov 30, 2006 18:01
Location: London

Re: Reversal of priority on pelican crossings

Post by WHBM »

AndyB wrote: Sun May 30, 2021 13:19 There is a lot of money to be made by the engineer who can merge SCOOT and MOVA so that the controller of a junction so equipped makes decisions based on both where there is space to send the next load of vehicles and what's approaching a particiular stop line.
I think you are quite right, that a controller approaching the efficiency/commonsense of someone standing there doing manual control (whether police or engineer) has yet to be invented. And that's one of the problems, you can't legislate for what's not on the market. It's a bit disappointing, after the substantial sums collected by Siemens and others over the decades, that they have not been able to come up with anything adequately practical.

I remember being told that in Edinburgh up to about 1950 there was a substantial tramway junction at the east end of Princes Street, a somewhat staggered junction with North Bridge and Leith Street, where the various routes turned all different ways, and there was a kerbside points controller who switched it notably efficiently as he saw the various trams turn up. Got replaced by traffic signals, at more cost than the "pointsman" wages for the next 20 years, which thereafter caused considerable delays that had never been experienced before, through to the end of trams some years later.
AndyB
SABRE Developer
Posts: 11056
Joined: Tue Dec 26, 2006 21:58
Location: Belfast N Ireland
Contact:

Re: Reversal of priority on pelican crossings

Post by AndyB »

I think my version looks like this:

MOVA detects vehicles on the approach loops on a leg with red and makes an initial decision to change.
SCOOT then reviews based on the exit loops of the junction (ie somewhere to put the vehicles), what SCOOT loops on the exit from the previous junctions are saying (ie will the queues risk becoming unacceptable if the lights don’t change), and finally the picture across the SCOOT system and knock-on effects before telling MOVA whether the lights may change just now or to wait until a review point expires and ask again.

Finally, after a certain timeout, MOVA decides it’s been stuck on red too long and changes anyway and hang the consequences - for the same reason, if the reason why SCOOT was saying no to the phase change request no longer applies, then MOVA is told it doesn’t have to wait.

It’s not particularly complicated in my head - the only real change is that instead of a fixed timing plan telling SCOOT to consider changing, MOVA does instead.
crb11
Member
Posts: 1630
Joined: Mon Feb 08, 2010 21:35
Location: Cambridge

Re: Reversal of priority on pelican crossings

Post by crb11 »

Stevie D wrote: Fri May 28, 2021 15:57
WHBM wrote: Fri May 28, 2021 11:51
Stevie D wrote: Fri May 28, 2021 11:28 My gut feeling is that there is also better compliance with pelican crossings
I am really surprised by that statement. Compliance BY PEDESTRIANS at Pelicans, other than when road traffic is a continuous stream, is appalling. Probably less than 10% wait in light traffic for their green.
There is no such concept as pedestrian compliance with pelican crossings, because there is nothing for them to comply with. Pedestrians are not required to wait for the green man, they are allowed to cross at any time (as long as it doesn't pose a danger to other road users, obvs).

I would also be amazed if so few pedestrians are waiting for a green man unless your observation sights are highly atypical. In most urban situations there is far too much traffic around for that many pedestrians to cross on a red man.
Round here (suburban Cambridge), I think 10% would be a fair figure for those choosing to wait when it would be otherwise safe to cross, and the bulk of those seem to be parents with small children and those who look as though they may be foreign students. Personally, unless there's a steady stream of traffic or small children around, I'll just cross normally and not press the button at all - means I get across quicker in general, and the road traffic isn't stopped. Call this appalling if you want.
[real name Colin]
User avatar
traffic-light-man
Member
Posts: 4728
Joined: Fri Sep 01, 2006 18:45
Location: Liverpool, UK

Re: Reversal of priority on pelican crossings

Post by traffic-light-man »

AndyB wrote: Sun May 30, 2021 15:02 I think my version looks like this:

MOVA detects vehicles on the approach loops on a leg with red and makes an initial decision to change.
SCOOT then reviews based on the exit loops of the junction (ie somewhere to put the vehicles), what SCOOT loops on the exit from the previous junctions are saying (ie will the queues risk becoming unacceptable if the lights don’t change), and finally the picture across the SCOOT system and knock-on effects before telling MOVA whether the lights may change just now or to wait until a review point expires and ask again.

Finally, after a certain timeout, MOVA decides it’s been stuck on red too long and changes anyway and hang the consequences - for the same reason, if the reason why SCOOT was saying no to the phase change request no longer applies, then MOVA is told it doesn’t have to wait.

It’s not particularly complicated in my head - the only real change is that instead of a fixed timing plan telling SCOOT to consider changing, MOVA does instead.
That's essentially what SCOOT does. It monitors the traffic flow between each node (junction, crossing, whatever), and if it sees stationary traffic on a link, it'll see congestion and try to reduce that as soon as possible. It's also possible to identify a loop which, if congested, must be causing exit blocking, so it then tries to reduce the amount of traffic heading for that exit.

SCOOT, by its very nature, applies a common cycle time to all the controllers within that SCOOT region in order to deal with the offset, hence it either forces stages or opens windows of opportunity to the controller, which, if the demand is there, it will serve (if it's set up as demand dependent). That's also where it can be incredibly sluggish off-peak, hence it's probably ideal to use the hybrid option and run MOVA off-peak, running linked MOVA for the more closely-spaced sites if required.

There is an option to only have SCOOT force some stages within a cycle (i.e., the 'main' road), allowing the controller to fall back to something else (i.e. MOVA) for the remainder of the cycle, but the risk ran there is that if the controller is still running it's minimum on the stage prior to the next SCOOT forced stage, and SCOOT tries to take over again, it'll not necessarily happen at the optimum time because SCOOT cannot override the controller's minimum timings. I'm not sure how commonly used that feature it, I imagine it's seldom used for that reason, but could be wrong.

It's probably worth checking out Dynniq's ImFlow - it's probably more along the lines of what you're mentioning. Of course, being a Dutch development, it's already quite common in the Netherlands but is still new to the UK. It also has the ability to prioritise certain road users, which comes back around to what we were discussing earlier upthread regarding adaptive systems that cater for more than just motor traffic.
Simon
User avatar
Stevie D
Member
Posts: 8000
Joined: Wed Dec 20, 2006 17:19
Location: Yorkshire

Re: Reversal of priority on pelican crossings

Post by Stevie D »

crb11 wrote: Mon May 31, 2021 11:20
Stevie D wrote: Fri May 28, 2021 15:57
WHBM wrote: Fri May 28, 2021 11:51
I am really surprised by that statement. Compliance BY PEDESTRIANS at Pelicans, other than when road traffic is a continuous stream, is appalling. Probably less than 10% wait in light traffic for their green.
There is no such concept as pedestrian compliance with pelican crossings, because there is nothing for them to comply with. Pedestrians are not required to wait for the green man, they are allowed to cross at any time (as long as it doesn't pose a danger to other road users, obvs).

I would also be amazed if so few pedestrians are waiting for a green man unless your observation sights are highly atypical. In most urban situations there is far too much traffic around for that many pedestrians to cross on a red man.
Round here (suburban Cambridge), I think 10% would be a fair figure for those choosing to wait when it would be otherwise safe to cross, and the bulk of those seem to be parents with small children and those who look as though they may be foreign students. Personally, unless there's a steady stream of traffic or small children around, I'll just cross normally and not press the button at all - means I get across quicker in general, and the road traffic isn't stopped. Call this appalling if you want.
Ah, I think I owe WHBM an apology, I skim read and saw the words "light", "traffic" and "green" and put them together as "green traffic light", completely ignoring the bit about "light traffic" ... yes, if there is no traffic coming then most people don't wait for the green man, which is fine as they don't need to. I'm not convinced that traffic being that light is a common enough occurrence at most signalised crossings that that is what we should base policy on.
crb11
Member
Posts: 1630
Joined: Mon Feb 08, 2010 21:35
Location: Cambridge

Re: Reversal of priority on pelican crossings

Post by crb11 »

Stevie D wrote: Mon May 31, 2021 20:10 I'm not convinced that traffic being that light is a common enough occurrence at most signalised crossings that that is what we should base policy on.
I would beg to differ: I would estimate there are in the order of 20 pedestrian crossings within a mile (by road) of my house, and no roads which are sufficiently busy outside peak periods (rush hour plus start/end of the school day) that an able-bodied person would need any help to cross them. (This is roughly the region bounded by Histon Road, Victoria Road and Milton Road, for those who know it.) This is a fairly typical suburban environment, so there are going to be thousands of similar cases across the country, and any policy should take them into account.
[real name Colin]
User avatar
the cheesecake man
Member
Posts: 2462
Joined: Mon Jan 07, 2019 13:21
Location: Sheffield

Re: Reversal of priority on pelican crossings

Post by the cheesecake man »

crb11 wrote: Mon May 31, 2021 11:20 Round here (suburban Cambridge), I think 10% would be a fair figure for those choosing to wait when it would be otherwise safe to cross, and the bulk of those seem to be parents with small children and those who look as though they may be foreign students. Personally, unless there's a steady stream of traffic or small children around, I'll just cross normally and not press the button at all - means I get across quicker in general, and the road traffic isn't stopped. Call this appalling if you want.
It's probably better than deliberately waiting for traffic to approach so I can press the button to stop it like I often did when crossing Queen's Road late at night. :D :rolleyes:
Post Reply