Hello Lukas

It would be less obvious how to join edges with differing lane numbers.
- there would be two thresholds involved (edge length and lane-count
difference)
- the connections at the start, join and end would have to be recomputed.
- if the short edge has more lanes then it must be decided which of them is
removed (non-trivial if there are mode-specific lanes involved such as a
bus lane)
Hence I'm unsure if a solution is possible that would be generally useful
(and not just tailored to one specific project).

With a bit of python processing you could iterate over the network edges
(sumolib.redNet().getEdges()) and create a patch .edg.xml file to change
lane numbers according to your needs, then use the min-length option. (see
https://sumo.dlr.de/docs/Tutorials/ScenarioGuide.html#modifying_the_network)

regards,
Jakob


Am Fr., 20. März 2020 um 17:56 Uhr schrieb Lukas Ambuehl <
[email protected]>:

> Thanks Jakob for the quick follow-up.
> Would that be a potential enhancement?
> Similar to geometry.remove.width-tolerance, you could set a threshold?
> I am asking because I am using meso, and as discussed previously, short
> edges have adverse effects  (and this seems to propagate onto other edges
> as well).
>
> On 20 Mar 2020, at 17:36, Jakob Erdmann <[email protected]> wrote:
>
> Hello,
> geometry.remove will never join edges that differ in the number of lanes.
> The option geometry.remove.min-length only applies to attributes such as
> speed, permissions and width.
>
> regards,
> Jakob
>
> Am Fr., 20. März 2020 um 17:14 Uhr schrieb Lukas Ambuehl <
> [email protected]>:
>
>> Hi all
>>
>> I am trying to reduce the number of short edges in my network.
>>
>> I use the following netconvert options:
>>      <geometry.remove value="true"/>
>>      <edges.join value="true”/>
>>      <*geometry.remove.min-length* value=“30"/>
>>      <geometry.min-dist value=“20"/>
>>      <geometry.remove.width-tolerance value=“2"/>
>>      <geometry.max-segment-length value="200”/>
>>
>> I thought that *geometry.remove.min-length * would control for the max.
>> length of an edge being merged with another one and
>> geometry.remove.width-tolerance for the difference in numbers of lanes
>> allowed. However, when running this on an OSM derived network, some edges
>> really do not want to merge with others. See attached screenshot of an edge
>> < 30m, which doesn’t want to be merged. Does it somehow depend on the
>> junction that follows or on the length of the other edges?
>> Any ideas on how to go about this? (I can also provide a MWE if needed)
>>
>> Thanks
>>
>> Lukas
>>
>> <PastedGraphic-1.tiff>
>> _______________________________________________
>> sumo-user mailing list
>> [email protected]
>> To unsubscribe from this list, visit
>> https://www.eclipse.org/mailman/listinfo/sumo-user
>>
> _______________________________________________
> sumo-user mailing list
> [email protected]
> To unsubscribe from this list, visit
> https://www.eclipse.org/mailman/listinfo/sumo-user
>
>
> _______________________________________________
> sumo-user mailing list
> [email protected]
> To unsubscribe from this list, visit
> https://www.eclipse.org/mailman/listinfo/sumo-user
>
_______________________________________________
sumo-user mailing list
[email protected]
To unsubscribe from this list, visit 
https://www.eclipse.org/mailman/listinfo/sumo-user

Reply via email to