The header/landing page for standards needs to be defined in t...
1
Benjamin EngelAug 17, 2023
remove draft from title and correct date.
/opendrive-group/o...enumerations.html
34
Mohammed HabibAug 17, 2023
Each annex have to be marked normative or informative
see ASM-91
/opendrive-group/o...st_of_tables.html
29
Mohammed HabibAug 17, 2023
Revise table names in respective sections, this is likely a ch...
/opendrive-group/o...06_04_header.html
24
Mohammed HabibAug 17, 2023
Why is this in a source code block? It is normal text with som...
/opendrive-group/o...ompatibility.html
21
Mohammed HabibAug 17, 2023
It is good that the full list is available in the annex, but t...
/opendrive-group/o...story_V1-8-0.html
20
Mohammed HabibAug 17, 2023
This block mixes source code and text. Suggest to revise the f...
/opendrive-group/o...elines/index.html
12
Benjamin EngelAug 17, 2023
Definition of chapter structure for non-specification type doc...
8
Benjamin EngelAug 17, 2023
Extend editorial guide with mandatory sections for other docum...
/opendrive-group/o..._crg_surface.html
6
Benjamin EngelAug 17, 2023
Add section in editorial guide - guidance on admonitions (note...
/opendrive-group/o...05_elevation.html
5
Benjamin EngelAug 17, 2023
Define formatting rules for equations & explanations.
13 of 12
39
Thomas BleherAug 23, 2023
In 1.7, there was a requirement to have at least one non-center lane with a non-zero width. I think this is a sensible requirement. Why was it dropped?
Status
Assigned to
Arne Düselder
Comment left on /opendrive-group/opendrive-... Chrome 115 · Linux x86_64 Desktop · 3458x1964
Labels
clarify details
technical
Arne Düselder Aug 28, 2023
As with comment 38, we need to clarify if this is a leftover from the intended model approach for the junction reference line.
Johannes SchmitzSep 1, 2023
Thomas Bleher This was dropped to enable the junction boundary construction. Also the discussion showed that some suppliers where violating this already anyway. If this creates a problem for you a more sophisticated rule could be added as a mitigation.
Thomas BleherSep 14, 2023
Johannes Schmitz: what is the semantic meaning of a road without any lanes? I think the standard should first and foremost describe the semantic meaning. To me a road without lanes doesn't make any sense. What should a tool do with it? How would you e.g. translate that to OSI?
Our tools currently explicitly reject roads without lanes.
If tools generate roads without lanes, isn't that a sign that some other construct is missing from the standard?
Johannes SchmitzSep 14, 2023
Thomas Bleher I agree that this is a little bit workaroundish but the idea in OpenDRIVE, I think already before I joined in the concept project phase was to describe junction boundaries based on road boundaries of connecting roads. If one accepts this as the modelling approach (instead of defining a separate "boundary geometry" XML element which I would be fine with and would probably be cleaner), then the problem arises that sometimes on one side of a junction there is no suitable connecting road available to close the boundary but there still exists an asphalt surface. To cover this edge case and close the boundary the idea is to use a "pseudo connecting road" or I also like to call them "non-connecting" junction roads. You can also attach a lane marking to the center lane of this road if necessary if the junction has boundary lines painted on it I think. In that case you might also want to define a border lane to have some extra asphalt outside of the marking I suppose. I think there are also people using laneless roads as a way to place objects since there is no other more generic concept for that (it would be good to decouple the geometry from the road and be able to use it for a variety of things, especially object outlines come to mind which are currently only polylines) I understand that this is maybe not 100% sactisfactory but I have tried multiple times for this and other topics to introduce some cleaner and/or more sophisticated solutions but it turns out that in OpenDRIVE legacy is king and the approach of least resistance usually wins.. Feel free to more actively participate in 1.9 so that we maybe have 2 OEM votes instead of 1 and can push for slightly cleaner decisions against tool vendors who want to. optimize for minimal changes :D We will still be outnumbered but maybe less extreme so.
Arne Düselder Sep 22, 2023
Add Note to explain why the old requirement was dropped
Johannes MeichssnerSep 27, 2023
Proposed note:
In older {THIS_STANDARD} versions a road required at least one lane with a width greater zero.
As roads can now be used for junction boundaries and do not need an extra lane, this rule has been removed.