Currently I'm working on the routing feature of my garmin maps and run into some problems. I was looking for detailed information concerning routing (incl. best pratice) but couldn't find such a document. Question: Could someone point me to such a documentation or describe "the best practice"? Thanks and regards - Klaus -- View this message in context: http://gis.638310.n2.nabble.com/Routing-Documentation-and-Best-Practice-tp67... Sent from the Mkgmap Development mailing list archive at Nabble.com.
On Mon, Aug 22, 2011 at 01:41:04AM -0700, toc-rox wrote:
Currently I'm working on the routing feature of my garmin maps and run into some problems. I was looking for detailed information concerning routing (incl. best pratice) but couldn't find such a document.
Question: Could someone point me to such a documentation or describe "the best practice"?
Routing is supposed to "just work" since more than two years now. Can you describe what you have been doing so far and in which way the routing is failing? Where do you get the map extract, how do you split and compile it? The exact commands would be nice. Best regards, Marko
Since Garmin has stopped the development of Mapsource in favour of Basecamp, it's maybe necessary to have a closer look at the routing scripts. In the latest releases of Basecamp parameters like bicycle=no, foot=no etc don't work anymore as expected. This leads to unwanted routing like cycling on steps and footways to find the shortest route (in Mapsource and on the GPS the route is behaving correctly according to the rules). This might be a bug in Basecamp but it might be also a new feature where mkgmap can't cope with. I'm afraid that if Garmin is updating the firmware of their units according to the newer Basecamp procedures routing will also fail on the GPS.
On 22/08/2011 10:54, Minko wrote:
I'm afraid that if Garmin is updating the firmware of their units according to the newer Basecamp procedures routing will also fail on the GPS.
The new eTrex series is due out "real soon now" - it'll be interesting to see what if anything they've changed. Cheers, Andy
Thanks for the response - I have something like this: highway = path & foot = designated {set highway = fussweg} highway = footway & foot = designated {set highway = fussweg} ... # Routing (nur Fussgaenger zulassen) highway = fussweg {add access = no; add foot = yes} highway = fussweg [0x16 road_class=0 road_speed=0 resolution 24 continue] # Bruecken / Tunnel highway = fussweg & bridge = yes [0x01040b resolution 24 continue] highway = fussweg & tunnel = yes [0x01040c resolution 24 continue] # Zoom-Out -1 highway = fussweg [0x010e05 resolution 23] My problem (tested on BaseCamp OS X and Dakota-20 Firmware 4.30) is: Motorcars and bicycles aren't avoided from routing. Peculiar: Disabling "carpool" has influence on the routing. Extract is from geofabrik, mkgmap release is r2012, splitter release is r179. This works but is imho a "poor" workaround and is NOT what i want: highway = fussweg {add toll = yes} @Minko: I think you exactly describe the problem. -- View this message in context: http://gis.638310.n2.nabble.com/Routing-Documentation-and-Best-Practice-tp67... Sent from the Mkgmap Development mailing list archive at Nabble.com.
On Mon, Aug 22, 2011 at 03:08:42AM -0700, toc-rox wrote:
Thanks for the response - I have something like this:
highway = path & foot = designated {set highway = fussweg} highway = footway & foot = designated {set highway = fussweg} ... # Routing (nur Fussgaenger zulassen) highway = fussweg {add access = no; add foot = yes} highway = fussweg [0x16 road_class=0 road_speed=0 resolution 24 continue]
# Bruecken / Tunnel highway = fussweg & bridge = yes [0x01040b resolution 24 continue] highway = fussweg & tunnel = yes [0x01040c resolution 24 continue]
# Zoom-Out -1 highway = fussweg [0x010e05 resolution 23]
My problem (tested on BaseCamp OS X and Dakota-20 Firmware 4.30) is: Motorcars and bicycles aren't avoided from routing. Peculiar: Disabling "carpool" has influence on the routing.
What is the calculated route like? Is this way the first or last portion of the route, or is this the only way to the destination? If yes, then this could be a Garmin feature that has been mentioned quite a few times in the past. If the routing directions said "get out of the vehicle and walk", it would be less confusing. Minko's explanation looks plausible too. I don't think it will be long until Garmin gets obsoleted by some Android devices. The Edge series could almost have a replacement in the upcoming SonyEricsson Xperia Active, if the software options are satisfactory. The hardware sounds good: support for ANT+Sport sensors and a rugged, replaceable case. This is just an impression I got without having actually used any Android devices yet. Best regards, Marko
The effect on BaseCamp OS X (3.2.1) Map with routing (red line = foot only; blue line = foot + bicycle): http://gis.638310.n2.nabble.com/file/n6710993/Map.png Map with unexpected motorcar routing: http://gis.638310.n2.nabble.com/file/n6710993/Unexpected_Routing.png Map with expected routing: http://gis.638310.n2.nabble.com/file/n6710993/Expected_Routing.png -- View this message in context: http://gis.638310.n2.nabble.com/Routing-Documentation-and-Best-Practice-tp67... Sent from the Mkgmap Development mailing list archive at Nabble.com.
I think Klaus deals with the same problems that my cycle map users have pointed me when using the latest releases of Basecamp. See http://forum.openstreetmap.org/viewtopic.php?id=13384&p=2 (in Dutch but you can use google translate) Not only my maps are affected but in fact all OSM/mkgmap and some other maps compiled with cgpsmapper (like the Onroute Motor and cycling maps). We have posted this problem on the Garmin forum, see http://forums.garmin.com/showthread.php?t=21681&page=5 I'm trying to figure out what Basecamp has changed and noticed that the road selection slider has become more powerfull. On my bikemap I've made cycleways the major highway by giving them the highest road_class (4) and a high road_speed (3, not 7 because the navigation warnings came way too early otherwise). When pushing the road selection slider towards the highways to the right, routing is forced now on cycleways which sometimes works much better than in Mapsource. A negative aspect is that the access=no bicycle=no foot=no etc parameters are ignored in the latest basecamp releases (3.2.1/3.2.2) so I'm working on some fixes in the style sheerts. One work arounnd is moving all the bicycle=no tagged roads to non routable lines, but this doesnt work for multi-purpose maps. Another negative aspect is that some users found out that avoidance of carpool lanes breaks the routing (like Klaus expererienced too). I have tried to avoid steps by giving them a mkgmap:unpaved=1 tag but Basecamp seems to ignore this too... I'll have to do more testing on this and I'm looking forward to new ideas. Minko
It seems that the "avoid" features (toll, carpool, ferry, unpaved, ...) are still working as expected. Maybe a workaround ... Klaus -- View this message in context: http://gis.638310.n2.nabble.com/Routing-Documentation-and-Best-Practice-tp67... Sent from the Mkgmap Development mailing list archive at Nabble.com.
I've tried it to avoid steps in the bicycle routing but this doesn't always work as expected: toll roads - yes ferry - yes unpaved - no (does work on track roads) carpool - no (plus unwanted side effects) Klaus wrote: It seems that the "avoid" features (toll, carpool, ferry, unpaved, ...) are still working as expected. Maybe a workaround ...
Toll roads and ferries seems to have the highest priority. Both have something to do with money - sounds reasonable. Klaus -- View this message in context: http://gis.638310.n2.nabble.com/Routing-Documentation-and-Best-Practice-tp67... Sent from the Mkgmap Development mailing list archive at Nabble.com.
Am 22.08.2011 14:40, schrieb Minko:
I've tried it to avoid steps in the bicycle routing but this doesn't always work as expected: toll roads - yes ferry - yes
unpaved - no (does work on track roads)
What is the tracktype? The rule is handling grade2 - grade6 as unpaved. highway=* & (surface=cobblestone | surface=compacted | surface=dirt | surface=earth | surface=grass | surface=grass_paver | surface=gravel | surface=grit | surface=ground | surface=mud | surface=pebblestone | surface=sand | surface=unpaved | mtb:scale=* | tracktype ~ 'grade[2-6]' | smoothness ~ '.*(bad|horrible|impassable)' | sac_scale ~ '.*(mountain|alpine)_hiking' | sport=via_ferrata) { add mkgmap:unpaved=1 } BTW: I would not see "cobblestone" as unpaved :-)
carpool - no (plus unwanted side effects)
Yes, because of the side effects this should not be used. Chris
Hi Chris, Finally I managed to get the avoidance of steps working with mkgmap:unpaved=1 I had to move it to the front of the action block. This rule didn't work for me in Basecamp (altough it worked in Mapsource!): highway=steps {add access = no; add foot = yes; set mkgmap:unpaved=1} [0x13 road_class=0 road_speed=0 resolution 24] Probably because the access rules were ignored by Basecamp, but I dont have a clue how. Finally this worked out: highway=steps {set mkgmap:unpaved=1; add access = no; add foot = yes} [0x13 road_class=0 road_speed=0 resolution 24] With this rule routing will avoid steps if the unpaved roads are avoided. If this box is not check marked, the route takes the steps no matter what the road selection is. In Mapsource this never happens. I'm also thinking to put mkgmap:unpaved=1 on footways, paths and pedestrian highways (if tagged with bicycle=no) to avoid those roads, even they are paved.
Could someone describe the "unwanted side effects" of the "avoid carpool roads" setting ? Thanks - Klaus -- View this message in context: http://gis.638310.n2.nabble.com/Routing-Documentation-and-Best-Practice-tp67... Sent from the Mkgmap Development mailing list archive at Nabble.com.
participants (5)
-
Chris66 -
Marko Mäkelä -
Minko -
SomeoneElse -
toc-rox