-
-
Notifications
You must be signed in to change notification settings - Fork 985
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Import Waypoints from GPX File into Navigation #20100
Comments
Could you share a specific GPX file for us to review? |
"attaching the track to the road" does not work for long routes, 100 - 300 kilometers. It is to slow to be useful. And to be honest I have never waited for it to complete because of how slow it is. Maybe this is fixed in the newest version? Any benchmarks for how must faster the new algorithm is? I just heard about the new algorithm yesterday. Here is a very short GPX file route that I created to test how OsmAnd handles multiple waypoints. When you go to navigation only the beginning and end show in the navigation list, leaving out all waypoints between start and end. Maybe if the waypoints were not excluded in the navigation list and a trip recalculation could be triggered based on the navigation list would work for cleaning up the turn-by-turn instructions? Here is the longer trip that would not provide proper turn-by-turn instructions. Since this trip was only Autobahn most of the way only a beginning and end waypoint was needed. Most of my routes have waypoints spread through out the route for route shaping. This trip had the Autobahn misdirection which caused a wrong turn. Please let me know if you would like anything else. |
The Furkot export track files.
Trjaect: Furkot_Test_Route-2.gpx
Traject: 2024_Use_Hotels-_3.gpx
OsmAnd imports this as regular simple "naked" track. As the various routers generate trajects along different roads an aternavite along the A5. |
Let me dig into that, maybe I did something wrong with the export? |
What I notice is that the track point definition in the Furkot version is much smaller and thus does not follow the roads very detailed and precise, which is what you should normally expect with a track. A road matching tool does probaly not handle that very well. The OsmAnd dev's can probably tell more. |
When I created that route I think it was in low resolution. I will create a new file and try to upload it into Cruiser App. Does that app work with multiple waypoints? |
I give up trying to create an importable gpx file. So if the routing service I am using cannot create importable gpx files that will provide turn-by-turn instructions then what web based routing service is a user suppose to use? I don't want to create routes on a small tablet screen and https://osmand.net/map is not ready. |
What you need is definitely possible. Reference and examples find here: Cruiser app. BRouter web and RouteYou web (Osrm routing- Set export as Locus gpx file) also generate the simple to use 100% reliable plug and play gpx navtracks. Plus the Cruiser app additionally imports and navigates the Kurviger web (GH routing) designs by the .kurviger files. Instantly follow an imported navigation track 100% truthfull inclusive correct TBT. There are sure (motor)bikers in your area who will help you to get started. Exchanges between apps unfortunately remains a pain point which is nevertheless quite easily optimisable with a Gpx navigation track |
Thank you for the reply! I will look into these tools to see how I can use them for planning tours! I like the routing service I am currently using since it allows planning a trip with multiple days in one go and breaks them up into multiple days for the navigation app. It also makes suggestions where the hotel stops would fall on the route. And allows to add time to see an attraction like 15 minutes for a photo stop or several hours for seeing a cave, etc. so the day does not become overly long. I would like to stay with OsmAnd since it has most of the features I desire. Although the other navigation software you mentioned looks interesting too. Between the routing and navigation service I will likely have to give up some features to make this all work together with the least amount of effort. I notice when the gpx file import fails it is usually at roundabouts, freeway exits and intersections. And I have yet to test the new attach to roads algorithm. I read here that the beta has a speed bump so maybe it is usable now? I thought of another test, will be back shortly. |
Fine understood. I see it this way. Each individual app definitely has features and tools that someone likes more or less. I don't see the range of apps as competitors either, but they can complement each other perfectly. What is mainly missing is a simple uniform file exchange method so that already added valuable elements such as TBT together with the via and shaping point info are not lost in exchanges. However, the exchange situation is currently as confusing unfortunately as it once was in Babylon. |
I have noticed that to be true! |
Since gps files are not standard and since importing them to create routes that provide turn-by-turn instructions is hit or miss, mostly miss in my experience. I believe having the option to load the route list with only entries from the gps file would be a doable workaround. It is unlikely all of the gps files are cleaned up so they work with OsmAnd. |
@yuriiurshuliak please summarize with screenshots I'm not sure I caught how exactly it would be easier to deal with GPX waypoints. |
@yuriiurshuliak Or if there is a way to trigger a route recalculation based on OsmAnd router for the whole route after import that would be great too. And thank you for looking into this! |
I tested the road matching further anyway by using OsmAnd web. Your original Furkot track demo version: Until both roundabouts correct, then jumbled.
I removed that nasty trackglitch from the original file. Find the new file in attachment. Import into OsmAnd web and the TBT are correct. |
Describe the idea (required)
When importing a GPX file the import only shows the exact route. The imported file does not show waypoints between the start and end waypoints and does not provide proper turn-by-turn navigation. I propose a new GPX file import process type. When importing the file OsmAnd prompts the user for which import type the user wants:
Choice one would import the GPX files the same as OsmAnd currently works.
Choice two would add the the first 'wpt' entry in the GPX file into the navigation list as if the user was building a new navigation. Then the second and third and etc. 'wpt' entry in the GPX file into the navigation list. This would continue until the last 'wpt' was added to the navigation list. At this point the waypoint import process would be complete. The user could then edit the navigation list if desired and / or save the navigation for future use. Or just start navigating the route.
This would provide all of the waypoints with turn-by-turn navigation.
Many users of OsmAnd have complained about the waypoints in OsmAnd. This would fix the user's problems with missing waypoints and the lack of proper turn-by-turn instructions.
Tell us about the expected behaviour (required)
I think above explains it all. Please let me know if there are questions.
Tell us about alternatives you've considered (required)
I have tried importing numerous GPX files of various formats to obtain proper turn-by-turn instructions and to load the waypoints without success.
Context (optional)
The current turn-by-turn instructions provided after importing a GPX file are not useful when driving on the Autobahn. For example, OsmAnd shows a right hand turn at the intersection of the Autobahn and does not prompt the user to get off the exit and does not draw the route using the exit. It is very confusing to know which way to go when there are two or more Autobahn intersections and exits to consider since there is only a right hand turn icon splashed down in the middle of the intersections. This caused me to miss my Autobahn exit causing me to drive 35 minutes out of my way to get back on route, more than once.
The text was updated successfully, but these errors were encountered: