[PATCH] Alpha code for pedestrian routing in plazas
Hello, It was previously mentioned on this list that pedestrian routing did not work when encountering pedestrian plazas and squares (highway=pedestrian & area=yes). Attached is an alpha patch inspired by (read: shamelessly copied from) Mark's cycleway=opposite addition. The coding adds a pedestrian way around the perimeter of the polygon. Routing appears to work in Mapsource, but I have not yet hat a chance to test this on a device. Note that Mark's cycleway=opposite code must be present for this patch to compile and work. Your comments would be appreciated. Cheers.
Hi Clinton, On Fri, May 29, 2009 at 12:31 PM, Clinton Gladstone <clinton.gladstone@googlemail.com> wrote:
Hello,
It was previously mentioned on this list that pedestrian routing did not work when encountering pedestrian plazas and squares (highway=pedestrian & area=yes).
Attached is an alpha patch inspired by (read: shamelessly copied from) Mark's cycleway=opposite addition. The coding adds a pedestrian way around the perimeter of the polygon. Routing appears to work in Mapsource, but I have not yet hat a chance to test this on a device.
Good idea with this! What do you think about creating pedestrian ways that connect to each of the streets that touch the plaza instead of just creating ways around the outside? There isn't a pedestrian plaza near me anymore so I can't test this patch. Sorry. Cheers, Ben
Good idea with this! What do you think about creating pedestrian ways that connect to each of the streets that touch the plaza instead of just creating ways around the outside?
That was my first thought too when I read Clintons post. But... what it the pedestrian area is shaped as a banana? Then you could get ways displayed which lead straight through a building. Does anybody have an Idea to solve this? Another aspect: The number of ways virtual through the pedestrian area should be limited a it increases quadratically. Despite of this - Clintons idea sounds very good! As far it's possible, it should be optimized. Markus
On May 29, 2009, at 10:18 PM, gypakk@gmx.eu wrote:
Good idea with this! What do you think about creating pedestrian ways that connect to each of the streets that touch the plaza instead of just creating ways around the outside?
That was my first thought too when I read Clintons post. But... what it the pedestrian area is shaped as a banana? Then you could get ways displayed which lead straight through a building. Does anybody have an Idea to solve this?
Yes, these are things I have thought about. I believe this is a general routing problem with OSM, and not just one affecting Garmin maps created by mkgmap: this also applies to other routing programs which attempt to provide routes suitable for pedestrians. Does anyone know how the other OSM routing programs (such as travelling salesman, etc.) solve this? (If at all) I have also noticed that the success of pedestrian routing through squares seems to be largely dependent on the person doing the mapping. For example, most of the pedestrian squares in Rome appear to be nicely linked to other streets; Siena, on the other hand, has its Piazza del Campo drawn as an independent polygon. Somehow I suspect that for pedestrian routing to work reliably, in any program, mappers must manually enter connecting ways. What do you think? Cheers.
Hi Clinton, I think, streets and pedestrian squares need to be linked to allow routing. If there is no link, we should assume that this is intended as there is no connection reality. The more OSM related routings systems are being used, the more people will see to edit their data more precisely.
Does anyone know how the other OSM routing programs (such as travelling salesman, etc.) solve this? (If at all) I didn't try it with TS yet, but www.openrouteservice.org leads you along the borders of pedestrian areas. This seems to be not the best solution but a practicable one.
Of course, to connect the junctions to each other by additional edges would be better. But an algorithm which is able to cope with the "banana" problem has to be developed first. Markus -------- Original-Nachricht --------
Datum: Sun, 31 May 2009 23:43:08 +0200 Von: Clinton Gladstone <clinton.gladstone@googlemail.com> An: Development list for mkgmap <mkgmap-dev@lists.mkgmap.org.uk> Betreff: Re: [mkgmap-dev] [PATCH] Alpha code for pedestrian routing in plazas
On May 29, 2009, at 10:18 PM, gypakk@gmx.eu wrote:
Good idea with this! What do you think about creating pedestrian ways that connect to each of the streets that touch the plaza instead of just creating ways around the outside?
That was my first thought too when I read Clintons post. But... what it the pedestrian area is shaped as a banana? Then you could get ways displayed which lead straight through a building. Does anybody have an Idea to solve this?
Yes, these are things I have thought about. I believe this is a general routing problem with OSM, and not just one affecting Garmin maps created by mkgmap: this also applies to other routing programs which attempt to provide routes suitable for pedestrians.
Does anyone know how the other OSM routing programs (such as travelling salesman, etc.) solve this? (If at all)
I have also noticed that the success of pedestrian routing through squares seems to be largely dependent on the person doing the mapping. For example, most of the pedestrian squares in Rome appear to be nicely linked to other streets; Siena, on the other hand, has its Piazza del Campo drawn as an independent polygon.
Somehow I suspect that for pedestrian routing to work reliably, in any program, mappers must manually enter connecting ways.
What do you think?
Cheers. _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
On Jun 1, 2009, at 8:41 PM, gypakk@gmx.eu wrote:
I think, streets and pedestrian squares need to be linked to allow routing. If there is no link, we should assume that this is intended as there is no connection reality. The more OSM related routings systems are being used, the more people will see to edit their data more precisely.
I agree with this, and think this is most likely the best way to handle this. I have noticed, for example, that there are several pedestrian areas which are completely surrounded by buildings. Trying to create connecting ways could result in ways that go through buildings or other impassible objects. Therefore, as you suggest, we can only rely on existing links to the squares. This has the side-effect of making the coding much easier. ;-)
Does anyone know how the other OSM routing programs (such as travelling salesman, etc.) solve this? (If at all) I didn't try it with TS yet, but www.openrouteservice.org leads you along the borders of pedestrian areas. This seems to be not the best solution but a practicable one.
This is also the same way that routing with my patch will work (provided there are linking ways), so at least that is consistent. Thanks for your comments. Cheers.
participants (3)
-
Ben Konrath -
Clinton Gladstone -
gypakk＠gmx.eu