new pseudo-tag: estimated_traffic_class

[abre...@googlemail.com]

This is a cross-post to a post in the german OSM user group:

http://forum.openstreetmap.org/viewtopic.php?id=31044

Just a short explanation here in this group:

I put a new pseudo-tag "estimated_traffic_class" into the rd5-files (cd5 not yet) that aims to estimate the actual traffic load on roads, independent from the tertiary/secondary/primary classfication.

Pseudo tag means it does not exist in OSM, it's just injected into the rd5 files.

You can see this the tag via http://brouter.de/brouter-web/ if you use the "Data" tab in the lower left box. It's just a first shot and I calculated them only for 4 squared ( E5_N45, E10_N45, E5_N50, E10_N50 ).

No problem to calculate more squares, just drop me a note. Currently I need to provide a "seed point" on the motorway network per square, but just from the processing time, worldwide processing is possible for this new tag.

[Jozef Matejička]

In Praha it seems to give results I do not understand.

http://brouter.de/brouter-web/#zoom=13&lat=49.9426&lon=14.5477&layer=OpenStreetMap&lonlats=14.52693,49.960221|14.518175,49.965411&nogos=&profile=trekking&alternativeidx=0&format=geojson

This street is really busy one, because it is used by huge suburb to access Praha. There is also motorway access, so I hoped, it would show.

Also I found bug in Web, when clicking on data tab, the horizontal scroll bar is below the screen, when  vertical scroll bar is added.

[broutertry]

Am Samstag, 16. Mai 2015 14:45:39 UTC+2 schrieb Jozef Matejička:

In Praha it seems to give results I do not understand.

http://brouter.de/brouter-web/#zoom=13&lat=49.9426&lon=14.5477&layer=OpenStreetMap&lonlats=14.52693,49.960221|14.518175,49.965411&nogos=&profile=trekking&alternativeidx=0&format=geojson

This street is really busy one, because it is used by huge suburb to access Praha. There is also motorway access, so I hoped, it would show.

Like Arndt in  http://forum.openstreetmap.org/viewtopic.php?id=31044 explained, it do not work in big town in urban agglomeration.

Also I found bug in Web, when clicking on data tab, the horizontal scroll bar is below the screen, when  vertical scroll bar is added.

So, it works better but not correct using fullscreen in firefox. 

[Jozef Matejička]

Like Arndt in  http://forum.openstreetmap.org/viewtopic.php?id=31044 explained, it do not work in big town in urban agglomeration.  

Ah OK. Need to reread that post. (And improve my Deutsch) 

[abre...@googlemail.com]

Am Sonntag, 17. Mai 2015 18:04:30 UTC+2 schrieb Jozef Matejička:

Like Arndt in  http://forum.openstreetmap.org/viewtopic.php?id=31044 explained, it do not work in big town in urban agglomeration.  

Ah OK. Need to reread that post. (And improve my Deutsch) 

I was working on a second version of that traffic model that better models the traffic in urban centers. It's a more sophistiacted model that includes direction preferences, short- and long-distance traffic components and a strong dependency of the traffic source density from the distance to the motorway backbone. I did not manage to get it fully consistent last weekend, needs more QA, but when it's ready to show, I will document it here as well, so no need for any language courses.

[Jozef Matejička]

Ah, I would like to master German anyway. Just not there yet.

I am looking forward to inspect results for Praha and Slovakia.

[Poutnik]

Hi Arndt, could you add the square E15_N45, so I could test is in South Moravia region of the Czech Republic ?

Dne neděle 10. května 2015 13:22:08 UTC+2 abre...@googlemail.com napsal(a):

[Jozef Matejička]

I would also appreciate E15_N45.
Thank you very much,
Jozef

Dňa po, 9. nov 2015 10:51 Poutnik <poutni...@gmail.com> napísal(a):

--
You received this message because you are subscribed to the Google Groups "OSM Android bikerouting" group.
To unsubscribe from this group and stop receiving emails from it, send an email to osm-android-biker...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[abre...@googlemail.com]

Am Montag, 9. November 2015 10:51:10 UTC+1 schrieb Poutnik:

Hi Arndt, could you add the square E15_N45, so I could test is in South Moravia region of the Czech Republic ?

done. ( E15_N45 + E15_N50 )

I also updated the rendered images of the traffic-simulation at: http://brouter.de/brouter_bin/traffic_pics/

This is best browsed by the leaflet-tool that Norbert explained here: http://forum.openstreetmap.org/viewtopic.php?pid=515249#p515249

[Jozef Matejička]

Thank you very much!

On first look results are impressive.

Jozef

Dňa pi, 20. nov 2015 18:54  <abre...@googlemail.com> napísal(a):

[broutertry]

Does this pseudo-tag still exist. The tag is in the lookups.dat, but I did not find a street with this. There were a lot in the Rhein-Main area two years ago.

[Poutnik]

Dne 20/04/2017 v 22:20 'broutertry' via OSM Android bikerouting napsal(a):

> Does this pseudo-tag still exist. The tag is in the lookups.dat, but I
> did not find a street with this. There were a lot in the Rhein-Main
> area two years ago.

It does. I notice it frequently in recent Czech Republic routing on
Brouter-web.

--
Poutnik ( The Wanderer )

My Brouter profiles
https://github.com/poutnikl/Brouter-profiles/wiki

[Poutnik]

Dne 20/04/2017 v 22:20 'broutertry' via OSM Android bikerouting napsal(a):

> Does this pseudo-tag still exist. The tag is in the lookups.dat, but I
> did not find a street with this. There were a lot in the Rhein-Main
> area two years ago.
>

Hmm, you may be right.
Few days ago they were in waytags, but it seems they disappeared.
Not sure if by intention or due generator error.

Perhaps Arndt realized this experimental feature is not as useful as
supposed.

[Arndt Brenschede]

On Thursday, April 20, 2017 at 10:49:51 PM UTC+2, Poutnik wrote:

> Does this pseudo-tag still exist. The tag is in the lookups.dat, but I
> did not find a street with this. There were a lot in the Rhein-Main
> area two years ago.

The change was that in Version 1.4.5 (see http://brouter.de/brouter/revisions.html )

 "filtering out unused way tags to increase profile cache efficiency"

so the tags are still there, but you generally see only tags that you use in the profile.

You can change back to the old behaviour by setting (in global context):

  assign processUnusedTags = true

but you should do that only for development

[Andrew38 Wizit]

Is it possible to assign a higher cost for sections of primary roads that lead to motorways? (Osmand does not always tag these as motorway_link as only a small portion of the road is really the motor_way link).

[Poutnik Fornntp]

Not on the profile level, as profiles cannot determine if a primary leads to a motorway.

It could be done by

- internally during OSM data preprocessing to raising the traffic class for such road segments

- raising penalty for high traffic classes for primaries, but this would not be specific for primaries connected to motorways.

Dne 6. října 2021 4:10:10 Andrew38 Wizit <wiz...@live.co.uk> napsal:

[Poutnik Fornntp]

It is meant as preprocessing before rd5 file creation on the Brouter server side.

Dne 6. října 2021 6:34:10 Poutnik Fornntp <poutni...@gmail.com> napsal: