How long does it take SatNavs to know about road changes

[IvanV]

If a local authority built a new road, or made a road one-way, or closed a road to motorised vehicles, etc, how long would it take SatNavs in people's cars to know about it? I appreciate the answer will be "it varies". But I'm interested in the practical range of variation, and how long it would take before most people would see it in their SatNav.

The reason I am asking is that the local authority tried something out in a road very near me, which is in response to SatNavs inappropriately directing people down a narrow lane. It is very narrow, with passing places. When there is trouble on a certain section of the M25, enough SatNavs direct people to use this road to cause total gridlock. It only takes two cars facing each other between two passing places, with the passing places full behind them, and no one can move.

But most of the time the road is fine. And for people like me, if the road was closed, then the village 2 miles to the south of me becomes 5 miles away.

So the local authority brought in an 18-month experimental traffic order in July, to close the road to motorised vehicles, except for access. But they didn't signpost it - that was the experiment. The idea is that local people would know it and still use it. But SatNavs would think the road is closed, and direct people elsewhere. In Sept/Oct they put in measuring equipment, and found that the traffic situation hadn't changed. So with only 4 months of the experiment, they think it has already failed.

But I wonder if they have waited long enough. Certainly if I go on google maps today, it thinks the road is open. I don't have a SatNav so I don't know what that would think.

There are in fact two issues here. One is, how long does it take SatNav companies to update their databases for road changes notified to them, and download that into people's SatNavs in cars?

The other is, would a SatNav would ever think that a road that is closed, if it is closed only by regulation notified to SatNav companies, but not actually signposted? I know that people drive around with cars with AI in them that can read signposts. I hired a car last year that could do that. Does that information end up with SatNav companies? If so, then maybe SatNavs wouldn't take account of virtual road closure that isn't actually signposted.

[Martin Y]

I suspect a great number of cars' own navigation maps only get updated if the dealer does it at a service rather than over the air. Whether they learn about road closures as if they were traffic updates I don't know.

I had Mrs Y's car yesterday and got very lost trying to use its navigation to get home (I had stopped at Corley services on the M6 and all I needed it to do was help me turn around and go back the way I'd come to get back South). Made a right pig's ear of that. Her car has maps and GPS but no 4G connectivity so no traffic info. At one point I was on some newish looking roundabouts near the NEC and all it could tell me was that I was "Off Road". The car's 4 years old so its maps probably are too.

In my own car I'd use Google Maps, but you say that doesn't mention the closure. I wonder if they track users? Maybe they marked it as closed but updated their info when driver tracking showed the road is not actually closed. They might think if everyone else is being told it's closed they can send their own users that way and it won't be a problem.

[IvanV]

It occurs to me that many people navigate cars these days by plugging their phone into the car console. Last year, we hired a car that did that (we hired 2 cars last year, this was more basic than the one that could read road signs). So what matters for a lot of people is what get on their phones. Google Maps doesn't show the closure, and a local correspondent tells me Apple Maps doesn't either. So no one using those apps, which is likely many people using their phones, will see the closure, atmo.

Then there are explicit SatNavs. I read that in a fairly recent development, there are some that automatically update when you park them outside your house and it connects to your wifi. Otherwise you have to plug them in to something and explicitly ask them to update. And getting the SatNav provider to update their maps is also a slow process.

When the council proposed this experimental traffic order, I thought it was a clever idea. I'm now coming to think of it as a fool's errand, it was never going to work.

But I do now have what I hope is an improved proposal for the council. Make the road closed to motorised vehicles except for permit holders. There is a parallel with road-side parking which can be over-used, and so a permit system can apply. Residents will get a free permit. People who live nearby can get apply for a permit for a charge, which would at least cover the admin costs. And people who don't use it often enough to pay for a permit will not be allowed in.

Any thoughts on that?

[Gfamily]

You can report road closures on Google Maps - it may take several goes, and since it's not an complete road closure it may not be accepted,
Out of interest, how does it show on the https://one.network/ website? I've found this invaluable for checking whether there are roadworks in an area.

My brother had an early built in Satnav on his Mercedes estate back in the early 2000s. The main problem was that it was used to continental style postcodes, so would only accept 6 characters, which isn't much help when you're trying to enter a postcode with 7 characters. They were about 2 hours late for lunch -

Our last 2 Satnavs have had a wireless receiver in the charger cable that let us know about delays and road closures on our route. Our most recent one doesn't have that, but it does have a phone app that can send notifications via Bluetooth to the unit. In that regard it's not quite as good as the previous models, but it has an advantage of allowing me to send destinations to the Satnav before we get into the car.

[jaap]

It is well known that Google uses active tracking of phones to measure congestion on the roads and displays that on their maps. It would not surprise me that if the road was marked closed, the mere fact that locals use the road while carrying an Android phone or using Google maps automatically causes the maps to be updated again with the road reopened.

[bolo]

It is well known that Google uses active tracking of phones to measure congestion on the roads and displays that on their maps. It would not surprise me that if the road was marked closed, the mere fact that locals use the road while carrying an Android phone or using Google maps automatically causes the maps to be updated again with the road reopened.

This, and I'm not sure the "permit holders only" option would do much. Google Maps doesn't seem to have any understanding of roads that are available only to certain vehicles. Near me there are several roads that are only open to vehicles with at least two (or in some cases three) occupants, at least during rush hour. Maps has no way of dealing with this.

[IvanV]

It is well known that Google uses active tracking of phones to measure congestion on the roads and displays that on their maps. It would not surprise me that if the road was marked closed, the mere fact that locals use the road while carrying an Android phone or using Google maps automatically causes the maps to be updated again with the road reopened.

This, and I'm not sure the "permit holders only" option would do much. Google Maps doesn't seem to have any understanding of roads that are available only to certain vehicles. Near me there are several roads that are only open to vehicles with at least two (or in some cases three) occupants, at least during rush hour. Maps has no way of dealing with this.

Thank you for confirming that google etc does use active tracking of phones for this kind of thing.

I have just discovered from neighbours that the road was shown as closed on Google/Apple maps for about a fortnight after the regulation came in. But then, it ceased to do so, presumably because the active tracking of phones indicated plenty of traffic along there, inconsistent with the regulation.

Even if it was signed as no motorised vehicles except for access, there would still be some traffic along the road. I would trust that google maps was clever enough to realise the difference between the level of traffic on such a restricted road, and a fully open road. But, I don't know.

So the question becomes, to what extent is this actually enforced, and to what extent would people take no notice of a "no motorised vehicles except for access", especially if they are not locals and directed by their SatNavs?

If it was additionally open to permit holders, again the question becomes one of enforcement. If you go into Hillingdon, there is the extensive private estate of Moor Park. This is closed except for residents and for access. At some entrances, it has a gate and you need to be registered for the gate to open for you. But some entrances are not gated. They have cameras reading people's number plates, which presumably deters abuse to some degree. I use it on my bicycle because it is a substantial short cut for some journeys - though more especially because it avoids cycling up an enormous hill, than saving very much linear distance.

[FlammableFlower]

Google Maps can take time to update to road closures (especially unplanned ones) - we had some serious flooding the other week and several roads got cut off. It didn't help that there are diversions in place due to a major A-road having work done on it and is closed for 6-months - these diversions are down substantially smaller roads through villages. Couple that with the flooding and it caused mayhem. A colleague who is effected by this road closure said that the first day of the floods it took him 2 hours to make a normally 25 minute journey.

So the second day he attempted an alternative route, except Google Maps kept telling him to go via the flood. Oh yes, there's no delay... but it's not because traffic isn't in a jam, that's because the bridge is under several feet of water.

[IvanV]

In researching this, I have learned that google increasingly knows road signs. This change has happened very fast, and I'm not sure how complete its data now is.

So the question has become, did Google show the road as open because of the traffic level, which they certainly knew, or because they knew that there wasn't a road sign, which they might have known after a little while if, eg, they were gleaning that info from smart cars that can read road signs.

If a road sign is now put up, will Google now show the road is closed because it knows the road sign, or will it show it as open because the traffic is inconsistent with the road sign? Which signal will predominate? The traffic may well be inconsistent with the road sign, because locals will know that it isn't enforced, and so still use the road because otherwise they'd have a 3 or 4 mile detour, depending on their origin and destination. Such signs are in general not actively enforced outside dense urban areas. And actively enforcing would force the locals to make a 3 or 4 mile detour, which is not in the public interest. I estimated the detriment of an enforced road closure as worth several million pounds per year. I had to guess the traffic level, but I think I did so conservatively, the council know it but didn't publish that data in their documents.

I now realise there may be a couple of good technological solutions. But they cost money. In the first one you can potentially make a charge to cover the cost:

A number-plate controlled barrier - these are now common, eg in car parks, and increasingly used to control access to roads in traffic-sensitive areas. I suggest that locals can apply for a permit for a modest fee, to cover admin, etc. And free to residents.

A variable road closure sign, that closes the road when the traffic level gets too high. People would obey that one, because it would act as a warning of getting stuck in gridlock for a long time. That's possibly quite tricky to calibrate.

So I have sent all this into the council, and see what they think. The official has asked to have a meeting with me, but only after the consultation deadline.

[dyqik]

I now realise there may be a couple of good technological solutions. But they cost money. In the first one you can potentially make a charge to cover the cost:

A number-plate controlled barrier - these are now common, eg in car parks, and increasingly used to control access to roads in traffic-sensitive areas. I suggest that locals can apply for a permit for a modest fee, to cover admin, etc. And free to residents.

A variable road closure sign, that closes the road when the traffic level gets too high. People would obey that one, because it would act as a warning of getting stuck in gridlock for a long time. That's possibly quite tricky to calibrate.

So I have sent all this into the council, and see what they think. The official has asked to have a meeting with me, but only after the consultation deadline.

The first solution here fails to deal with a large number of edge cases: e.g. agency carers who have to attend a disabled person's house in response to an assistance call, tradespeople visiting job sites, and occasional visitors to residents.

Fundamentally, roads are rights of way available equally to all users of a particular class of vehicle. Closing them to the general public is making them private roads, and they should lose all public money for upkeep at that point.

[Grumble]

Also deliveries. The number of deliveries nowadays is very high.
Surely the easiest thing to do is make it one-way?

[Gfamily]

An alternative would be to use a variant of the average speed camera (which records the time you enter and leave a stretch of road, and sends a ticket if it's less than the time it would take at the speed limit that applies).
The variant would be that it would send a ticket if the time was less than the speed limit time plus x minutes which would allow people access who are intending to make a stop along the road.

[IvanV]

I now realise there may be a couple of good technological solutions. But they cost money. In the first one you can potentially make a charge to cover the cost:

A number-plate controlled barrier - these are now common, eg in car parks, and increasingly used to control access to roads in traffic-sensitive areas. I suggest that locals can apply for a permit for a modest fee, to cover admin, etc. And free to residents.

A variable road closure sign, that closes the road when the traffic level gets too high. People would obey that one, because it would act as a warning of getting stuck in gridlock for a long time. That's possibly quite tricky to calibrate.

So I have sent all this into the council, and see what they think. The official has asked to have a meeting with me, but only after the consultation deadline.

The first solution here fails to deal with a large number of edge cases: e.g. agency carers who have to attend a disabled person's house in response to an assistance call, tradespeople visiting job sites, and occasional visitors to residents.

I'm proposing one (two way) barrier at one end of the road, which makes it into a cul-de-sac for those that don't have permits. Access is available to all at one end of the road. Regular users can apply for a permit, so can use it as a through route. For various practical reasons, not worth going into, it would have to be a barrier near one end of the road - one specific end of the road - not in the middle. So it would be one cul-de-sac, not two. All the properties are well away from the junctions at both ends of the roads, so the barrier can be a little distance into the road so any queuing to use the barrier is, in normal traffic conditions, off the connecting roads. You'd have to ensure traffic queuing one way for the barrier can get past traffic queuing the other way for it. Suitably located traffic lights would be needed for this. But this is no different from the present issue of traffic travelling each way getting past each other.

The road is too narrow, in relevant places, to stop a vehicle on the road to make a delivery or otherwise visit a property. So anyone doing that has to enter a property, and so can turn their vehicle around as the property owner themselves would, and come out of the road the same way they went in. There is also a field entrance with a short drive, after all the properties, but before the end of the road where the barrier would have to be, where even a construction truck can turn, if it was unable to turn within a property.

Fundamentally, roads are rights of way available equally to all users of a particular class of vehicle. Closing them to the general public is making them private roads, and they should lose all public money for upkeep at that point.

The point on turning a road into a private road maintained at public expense was well made in an earlier consultation before the experimental traffic order. And it is a good point, that I thought about for quite some time. I realised that in fact this is a fairly common situation. It applies to all kinds of cul-de-sacs, and other situations where a road is relatively little use for through traffic. So people get their road maintained at public expense, but experience little or no through or casual traffic. Discouraging through traffic from all kinds of residential areas and other traffic sensitive areas has long been done, though its recent large extension in dense urban areas is controversial. The difference in the present case is that the road saves people up to 4 miles on a journey, whereas in most cases there is an alternative route more suited to through traffic nearby, with nothing like a journey extension of that magnitude.

What I'm proposing keeps the road much more open to the general public than the council's proposal. The general public can use the road as a through route if they get a permit. They can always use it as a cul-de-sac. Whereas making the road no motorised vehicles except for access, as the council proposes, really makes it into a private road - if observed by the wider public.

An ex-DfT person said to me that it is becoming increasingly common to use number-plate controlled barriers to restrict entry to traffic sensitive areas. So I'm not the first to invent this, and it is an established solution.

Surely the easiest thing to do is make it one-way?

That would be another easy "solution", but it would still have substantial detriments. I'm not sure why it hasn't been mentioned as an option by the council, at least in the materials I have been able to access. This and the consultation and previous consultation have been about specific proposals, without mention of alternatives, and so the council hasn't said why they have been rejected, although I understand WS Atkins did a report for them.

In terms of making people drive 3 or 4 miles round, they'd only have to do it half as often, if the alternative was a widely-observed no entry at both ends. But now residents of the lane would also suffer the journey extension, which might be one reason those people in particular might be upset by it. That might be the reason it has been rejected. Though I am acquainted with a (former) resident of the road who strongly argued for it to be a one-way street (during the period he lived there).

It is much more likely that a one-way street would be widely observed, because you don't want to meet traffic coming the other way in a very narrow one-way street.

But it would still be open in one direction, as far as Google is concerned, and so we'd likely still get the massive diversion onto the road during M25, etc, issues, albeit only in one direction. At least the road wouldn't gridlock due to such traffic meeting traffic coming the other way and be unable to get past it. It would just become heavily congested with a lot of queuing, like any other inappropriate road M25 traffic was trying to use during problems. I don't know to what extent that should be considered as unacceptable.

So I agree with you that, if the council's proposal does in fact stop general through traffic using the road - which I don't know, and I don't think anyone can definitively say - then a one-way street could well be better. This is why I think the council's proposal should be experimental, not final. A one-way street could be a better solution. But a number plate controlled barrier could, in that case, I think, be an even better solution.

I'm pleased to get these reactions from you to force me to argue it, and see if you can spot any further holes in my arguments. Or whether I'm being unduly utopian about this.

[IvanV]

An alternative would be to use a variant of the average speed camera (which records the time you enter and leave a stretch of road, and sends a ticket if it's less than the time it would take at the speed limit that applies).
The variant would be that it would send a ticket if the time was less than the speed limit time plus x minutes which would allow people access who are intending to make a stop along the road.

Speed isn't a problem. Though there could be a point to measuring how long it takes people to get through. If you get stuck in the road for 20 minutes because of a gridlock you didn't know about when you entered the road - which in general you won't - it's not your fault. You didn't know. But it could could potentially be made known to you that there is a problem, if there was a variable sign closing the road when traffic detectors discovered that cars were not getting through in a reasonable time. Then you would know that it would be foolish to disobey the sign. 4 miles round will be quicker than joining the gridlock.