[mkgmap-dev] question about ignored no_u_turn restriction

Sun Aug 2 10:03:10 BST 2015

Hi Ben,

it seems the tests were even more complicated than I thought,
but in the end, today I found that I was wrong. The no_u_turn restriction
has an effect, the warning makes no sense.
This is what I tested:
Switch to car routing with "shortest distance" (shorter time seems to avoid u-turns),
and enable automatic recalculation of routes.
Enter a target which just requires a left turn after 100m. Ignore the left turn,
after a few seconds the device tells me to make a u-turn at the next crossing.
(tried this two or three times with the same result)

I've then used JOSM to add a no_u_turn restriction for this crossing , created a new
map (with a different id) and tried again. And, ta-ta, the device shows different 
hints for the same situation (no u-turn at this crossing). 
I thought I tried the same last time, but I must have made an error somewhere,
maybe some kind of caching in my Oregon 600.

So I've removed the check with r3626, the restriction is no longer ignored.

Gerd

Date: Sat, 1 Aug 2015 22:11:58 +0200
From: ben at bagu.org
To: mkgmap-dev at lists.mkgmap.org.uk
Subject: Re: [mkgmap-dev] question about ignored no_u_turn restriction

Hi Gerd,

For point 2, what makes you think that the restriction has no effect on Garmin routing? You said that it wasn't easy to test but did you mean that it's not possible to test this? If we don't know for sure or if it's too hard to test, changing the warning message to an informational would be good. You can always add a link to this thread in a comment in case we discover more information about the effect on the Garmin routing.

Thanks again for your help.

Ben

On Wed, Jul 29, 2015 at 11:09 AM, Gerd Petermann <gpetermann_muenchen at hotmail.com> wrote:

Hi Ben,

my results so far:
1) The img format allows to store this kind of no-u-turn, mkgmap also doesn't seem to have a problem
with it, I just have to comment the check.
2) I still think that the restriction has no effect on Garmin routing

I think I should change the message to ... no_u_turn with equal 'from' and 'to' way and via node has no effect, is ignored
and make it informational.

OK?

Gerd

Date: Tue, 28 Jul 2015 17:38:34 +0200
From: ben at bagu.org
To: mkgmap-dev at lists.mkgmap.org.uk
Subject: Re: [mkgmap-dev] question about ignored no_u_turn restriction

Hi Gerd,

I guess I was a little to 'quick' with that example. ;-) Thanks for looking into this.

Ben

On Tue, Jul 28, 2015 at 5:22 PM, Gerd Petermann <gpetermann_muenchen at hotmail.com> wrote:

Hi Ben,

okay, I did not yet think about the case that the device knows in which direction you are currently driving
when it re-calculates a route. I did only consider the case that you are planning while standing still.
If I remember correctly, this case will produce a message like "please make a u-turn", 
I never saw that it tells me to make a u-turn at a specific place.

It's not easy to test if Garmin uses this restriction (when written by mkgmap), I have to think 
about it.
@Steve: Do you see these restrictions in Garmin maps?

Gerd
PS: Your example is not good, as Quick Avenue is a (wrong) oneway, but that shouldn't matter here ;-)

Date: Tue, 28 Jul 2015 16:50:30 +0200
From: ben at bagu.org
To: mkgmap-dev at lists.mkgmap.org.uk
Subject: Re: [mkgmap-dev] question about ignored no_u_turn restriction

Hi Gerd,

In this example, if you're travelling southbound on North Harlem Avenue and you miss your right turn onto Quick Avenue, the GPS might suggest that you make a u-turn at the North Harlem Avenue / Ontario Street intersection so that you can double back get back to your original route. The traffic rule is saying that you're not allowed to make a u-turn at this intersection and the OSM data is capturing the traffic rule correctly. 

I've never been to this area so I can't actually test to see if the Garmin routing algorithm would try to do this u-turn. This is just an example that I'm using to try to figure if why the restriction is being ignored. I don't actually know the details Garmin routing algorithm but I have been routed on such u-turns in the past when I've missed a turn.

Given that this is a legitimate type of 'no u-turn' restriction, if the Garmin map format or routing algorithm can't deal with it, there should be a message that says so in the warning. In this case, maybe an info message makes more sense - but you should probably decide how you want to present it since you wrote it. :-) The other side of this is that the Garmin map format or routing algorithm can actually handle this specific type of 'no u-turn' restriction, in which case it would be nice to have the restriction included in the maps.

Thanks for your help and all your hard work on mkgmap!

Ben

On Tue, Jul 21, 2015 at 10:44 AM, Gerd Petermann <gpetermann_muenchen at hotmail.com> wrote:

Hi Ben,

unfortunately I did not add a comment to that part of the source which would explain
why I coded it, but I think the reason is that the restriction has no effect on route calculation.
I can't think of any case where the Garmin algo would route you along that u-turn.
Do you have an example that proves this assumption to be wrong?

If not, I can change the msg to say something like "has no effect", or please suggest a better text.

Gerd

Date: Mon, 6 Jul 2015 13:34:12 +0200
From: ben at bagu.org
To: mkgmap-dev at lists.mkgmap.org.uk
Subject: Re: [mkgmap-dev] question about ignored no_u_turn restriction

Just to follow up ... Does anybody know concretely that the Garmin format cannot handle and u-turn restriction with the same from and two way?

Thanks, Ben

On Sun, Jul 5, 2015 at 9:20 PM, Ben Konrath <ben at bagu.org> wrote:
Hi Anor,

Thanks for the tip but it seems that your suggestion breaks the OSM rule of manipulating the map for specific renderers (the renderer here being mkgmap). If the Garmin format truly doesn't support the u-turn restriction with the same to and from way at an intersection, we should come up with another solution.

Ben

On Sun, Jul 5, 2015 at 4:01 PM, A. Carlos <anorcarlos at hotmail.com> wrote:

   Ben

 There I draw two routes, one in each direction, since there is a false median, then with 2-way, it's easy for a restriction

_______________________________________________________________________________________________________

Anor                                                              Concórdia SC   

Date: Sun, 5 Jul 2015 15:53:50 +0200
From: ben at bagu.org
To: mkgmap-dev at lists.mkgmap.org.uk
Subject: Re: [mkgmap-dev] question about ignored no_u_turn restriction

Hi Thorsten,

Thanks for your reply. This type of restriction is probably a country specific thing. In Ontario Canada, you can make u-turns at intersections (regardless of the road is separation) if there is no sign indicating that you can't make the u-turn. Here's the information directly from the Ontario Ministry of Transportation:

http://www.mto.gov.on.ca/english/dandv/driver/handbook/section2.6.11.shtml

I realize that the link I sent previously is in the US but I suspect that it's the same policy there which is why the restriction is tagged like it is.

Since the tagging seems to be valid, does the Garmin format support this type of restriction? If not then it's probably a good idea to indicate this in the warning message. Maybe the message should also be changed to an info message if there's no problem with the data. Does anybody have other insights?

Thanks, Ben

On Sun, Jul 5, 2015 at 2:47 PM, Thorsten Kukuk <kukuk at suse.de> wrote:
On Sun, Jul 05, Ben Konrath wrote:

> Does anybody know why this particular restriction is being ignored?

Beside that this particular type of restrictions doesn't make

any sense to me, I would guess the GARMIN format does not support it.

If the to and from ways are the same, I have never seen a sign forbidding

u-turns. Only, if you have two ways, structural seperated. But then I

should tag the street that way and the restriction will not be ignored.

  Thorsten

--

Thorsten Kukuk, Senior Architect SLES & Common Code Base

SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nuernberg, Germany

GF: Felix Imendörffer, Jane Smithard, Dilip Upmanyu, Graham Norton, HRB 21284 (AG Nürnberg)

_______________________________________________

mkgmap-dev mailing list

mkgmap-dev at lists.mkgmap.org.uk

http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

_______________________________________________
mkgmap-dev mailing list
mkgmap-dev at lists.mkgmap.org.uk
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev 		 	   		  

_______________________________________________

mkgmap-dev mailing list

mkgmap-dev at lists.mkgmap.org.uk

http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

_______________________________________________
mkgmap-dev mailing list
mkgmap-dev at lists.mkgmap.org.uk
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev 		 	   		  

_______________________________________________

mkgmap-dev mailing list

mkgmap-dev at lists.mkgmap.org.uk

http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

_______________________________________________
mkgmap-dev mailing list
mkgmap-dev at lists.mkgmap.org.uk
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

_______________________________________________

mkgmap-dev mailing list

mkgmap-dev at lists.mkgmap.org.uk

http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

_______________________________________________
mkgmap-dev mailing list
mkgmap-dev at lists.mkgmap.org.uk
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev 		 	   		  

_______________________________________________

mkgmap-dev mailing list

mkgmap-dev at lists.mkgmap.org.uk

http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

_______________________________________________
mkgmap-dev mailing list
mkgmap-dev at lists.mkgmap.org.uk
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.mkgmap.org.uk/pipermail/mkgmap-dev/attachments/20150802/269fc4bc/attachment-0001.html>