Problem with adjust-turn-headings

Hi, I have found a situation where --adjust-turn-headings creates a non useful turn hint. Compile the file: http://files.mkgmap.org.uk/detail/207 with parameters --route and --adjust-turn-headings (maybe also --x-no-mergeroads for easier debugging). Try a route from Frankfurter Straße to Eilper Straße. Mapsource and the GPS give a hint that you have to turn right on the junction of Frankfurter Straße and Eilper Straße although the street is straight. Logging of the RouteNode class gives the hint: Adjusting turn heading from -52.342655 to -99.54512 at junction of (http://www.openstreetmap.org/browse/way/31171248) and (http://www.openstreetmap.org/browse/way/144387754) at http://www.openstreetmap.org/?mlat=51.349647&mlon=7.484266&zoom=17 For me the initial heading -52.34... looks weird. Gerd, do you have any ideas? WanMil

Hi WanMil, I think the algo should not use/need the finalHeading value. You may look at the algo in RoadNetwork.getAngle(). Gerd WanMil wrote
Hi,
I have found a situation where --adjust-turn-headings creates a non useful turn hint. Compile the file: http://files.mkgmap.org.uk/detail/207 with parameters --route and --adjust-turn-headings (maybe also --x-no-mergeroads for easier debugging). Try a route from Frankfurter Straße to Eilper Straße. Mapsource and the GPS give a hint that you have to turn right on the junction of Frankfurter Straße and Eilper Straße although the street is straight.
Logging of the RouteNode class gives the hint: Adjusting turn heading from -52.342655 to -99.54512 at junction of (http://www.openstreetmap.org/browse/way/31171248) and (http://www.openstreetmap.org/browse/way/144387754) at http://www.openstreetmap.org/?mlat=51.349647&mlon=7.484266&zoom=17
For me the initial heading -52.34... looks weird. Gerd, do you have any ideas?
WanMil _______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
-- View this message in context: http://gis.19327.n5.nabble.com/Problem-with-adjust-turn-headings-tp5805814p5... Sent from the Mkgmap Development mailing list archive at Nabble.com.

Hi WanMil, forget the 1dt sentence. I have no idea why the algo changes anything in this situation. Gerd GerdP wrote
Hi WanMil,
I think the algo should not use/need the finalHeading value. You may look at the algo in RoadNetwork.getAngle().
Gerd WanMil wrote
Hi,
I have found a situation where --adjust-turn-headings creates a non useful turn hint. Compile the file: http://files.mkgmap.org.uk/detail/207 with parameters --route and --adjust-turn-headings (maybe also --x-no-mergeroads for easier debugging). Try a route from Frankfurter Straße to Eilper Straße. Mapsource and the GPS give a hint that you have to turn right on the junction of Frankfurter Straße and Eilper Straße although the street is straight.
Logging of the RouteNode class gives the hint: Adjusting turn heading from -52.342655 to -99.54512 at junction of (http://www.openstreetmap.org/browse/way/31171248) and (http://www.openstreetmap.org/browse/way/144387754) at http://www.openstreetmap.org/?mlat=51.349647&mlon=7.484266&zoom=17
For me the initial heading -52.34... looks weird. Gerd, do you have any ideas?
WanMil _______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
-- View this message in context: http://gis.19327.n5.nabble.com/Problem-with-adjust-turn-headings-tp5805814p5... Sent from the Mkgmap Development mailing list archive at Nabble.com.

Hi Gerd, mmh, I don't understand what happens in the tweezeArcs method. Can you explain in some sentences? WanMil
Hi WanMil,
forget the 1dt sentence. I have no idea why the algo changes anything in this situation.
Gerd
GerdP wrote
Hi WanMil,
I think the algo should not use/need the finalHeading value. You may look at the algo in RoadNetwork.getAngle().
Gerd WanMil wrote
Hi,
I have found a situation where --adjust-turn-headings creates a non useful turn hint. Compile the file: http://files.mkgmap.org.uk/detail/207 with parameters --route and --adjust-turn-headings (maybe also --x-no-mergeroads for easier debugging). Try a route from Frankfurter Straße to Eilper Straße. Mapsource and the GPS give a hint that you have to turn right on the junction of Frankfurter Straße and Eilper Straße although the street is straight.
Logging of the RouteNode class gives the hint: Adjusting turn heading from -52.342655 to -99.54512 at junction of (http://www.openstreetmap.org/browse/way/31171248) and (http://www.openstreetmap.org/browse/way/144387754) at http://www.openstreetmap.org/?mlat=51.349647&mlon=7.484266&zoom=17
For me the initial heading -52.34... looks weird. Gerd, do you have any ideas?
WanMil _______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
-- View this message in context: http://gis.19327.n5.nabble.com/Problem-with-adjust-turn-headings-tp5805814p5... Sent from the Mkgmap Development mailing list archive at Nabble.com. _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

Hi WanMil, sorry, I read the comments multiple times, but I also don't know what it does in detail. It changes the "initial heading" value of an arc if some criteria are met, this initial heading value is written to the img file and Garmin uses it to calculate the angle between two arcs. This angle has an influence on routing (sharp angles are avoided) and it is used to create hints like "turn left" or "keep left". I did never try to find out if the used threshold values are good. I am pretty sure that the algo is not aware that one OSM way may be used for multiple routable ways, so if you have a closer look, please check this as well. Gerd WanMil wrote
Hi Gerd,
mmh, I don't understand what happens in the tweezeArcs method. Can you explain in some sentences?
WanMil
Hi WanMil,
forget the 1dt sentence. I have no idea why the algo changes anything in this situation.
Gerd
GerdP wrote
Hi WanMil,
I think the algo should not use/need the finalHeading value. You may look at the algo in RoadNetwork.getAngle().
Gerd WanMil wrote
Hi,
I have found a situation where --adjust-turn-headings creates a non useful turn hint. Compile the file: http://files.mkgmap.org.uk/detail/207 with parameters --route and --adjust-turn-headings (maybe also --x-no-mergeroads for easier debugging). Try a route from Frankfurter Straße to Eilper Straße. Mapsource and the GPS give a hint that you have to turn right on the junction of Frankfurter Straße and Eilper Straße although the street is straight.
Logging of the RouteNode class gives the hint: Adjusting turn heading from -52.342655 to -99.54512 at junction of (http://www.openstreetmap.org/browse/way/31171248) and (http://www.openstreetmap.org/browse/way/144387754) at http://www.openstreetmap.org/?mlat=51.349647&mlon=7.484266&zoom=17
For me the initial heading -52.34... looks weird. Gerd, do you have any ideas?
WanMil _______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
-- View this message in context: http://gis.19327.n5.nabble.com/Problem-with-adjust-turn-headings-tp5805814p5... Sent from the Mkgmap Development mailing list archive at Nabble.com. _______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
_______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
-- View this message in context: http://gis.19327.n5.nabble.com/Problem-with-adjust-turn-headings-tp5805814p5... Sent from the Mkgmap Development mailing list archive at Nabble.com.
participants (2)
-
GerdP
-
WanMil