-
Notifications
You must be signed in to change notification settings - Fork 85
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
[EPIC] Crossing the Lane/Road #1223
Comments
SegmentsMediansIssues: #89 , #192 , #503 , #504 , #243 , #1220 , #515 The game has no concept of medians or other "barriers" between lanes.
Suggestions:
Lane directionIssues: #89 (first image) We can't assume lanes in one direction are grouped together on one side of the road, or even the 'usual' side of the road:
Suggestions:
Lane sequenceIssues: #786 There doesn't appear to be control over the order in which lanes are defined (eg. from left to right); it's possible for assets to jumble them up. For example, in #786 (comment) ped lanes on opposite side of road are adjacent to each other in the data structure (lanes 7 & 8). This makes it painful to determine if going from lane A to lane C is going to cross a median or oncoming traffic. Suggestions:
|
NodesLane changes need to respect medians, where applicable.
Middle, BendIssues: #89 (OP)
U-turn, Stretch, Custom, Crossing
Example: A Queens Blvd style road could gap ouer medians between, but keep solid central median. That would allow traffic to path from outer lane group to inner lane group, but not cross to the other side of road (as central median would block). JunctionsIssues: #646 (Roads heading in OP)
Automation / UXIdeally, for nodes which facilitate "gappable" medians, it would be great if TM:PE (or external mod) could automatically set up lane connectors, etc. It would be spectacular if user could simply click medians (or thier outlines) to toggle them on/off, and have lane routing, etc., update accordingly. This approach would be especially desirable in cases where there might be multiple medians (eg. Queens Blvd) . Ped CrossingsWith the concept of lane groups and identifiable medians (which could have ped lanes added if not already present), we could possibly have staggered ped crossings. In other words, a ped crossing could be set to cross specific lane group(s), the cim would cross to a median, walk along it, to get to next ped crossing over the next lane group, etc. Junction Restrictions / Priority Signs / etcAgain, with lane groups, we could potentially apply junction restrictions, priority signs, traffic lights, etc., per lane group. Ability to define a lane as a filter lane (see also #515) could separate it in to its own lane group. Dynamic Lane SelectionWould need updating to stay within lane group. Given that road of same type should have same lane groupings from one segment to another, it should be feasible for DLS scan-ahead to handle that? |
Here's my take, from a player's perspective, not knowing the modding side so well (yet). Trying to detect medians on existing road assets, as it turns out, is a fool's errand. All the great feedback in #1220 showed me that. We could flag a lane as a median, but my gut tells me this would be too limiting, and might even be harder to implement than another option:
Related EnhancementsThe roads will not work the way people expect unless vehicle behavior can be altered appropriately. My suggestion is to add a "Lane Avoidance Level" property in Adaptive Roads.
Once this is implemented, asset makers would learn how to find the best number for the work they're doing. An interesting use case for this feature would be center turning lanes, which would allow traffic both directions and have a lane avoidance level of 100. Enhancements that would be cool but won't ever happenQueens Boulevard has stop signs and stoplights on the ramps from the center (express) roadways to the outer (local) roadways. So we're basically talking about a stop between lanes instead of between streets. This is obviously paradigm-breaking and probably not feasible in the slightest. But it would be cool. 😎 |
One more thought on this. If the implementation of a new lane movement control system is treated as its own project, then it really does simplify the median detection problem by allowing you to focus on that specifically. Knowing that the logic is there and does what it needs to, you can go in and start knocking out the specific scenarios piecemeal, carefully evaluating and testing each one. For example, it looks like the case of pedestrian or parking lanes between vehicle lanes is an easy one, and maybe the first where you'd just programmatically apply restrictions to the nearest vehicle lanes when the asset is loaded. Another case are vanilla assets and stable, popular workshop assets that are known to have medians. They could just be detected and altered at runtime right away. |
#1528 addresses some of what this issue covers. It is use case focused, rather than trying to be a general median solution. |
Epic Work Items
Related Issues
The following issues may be resolved or made obsolete by the implementation of this epic.
Summary
This is an(other) attempt to collate general issues relating to where road vehicles can change lanes or cross the road, notably in relation to medians and lane directions. These issues could also apply to rail, tram, monorail, etc., but the focus here is road.
Over the years loads of issues have amassed, many are duplicate or related. Purpose of this
Crossing the Lane/Road
issue is to try and get everything in one place, as coherently as possible, and then close all the other linked issues to declutter the issue tracker.The text was updated successfully, but these errors were encountered: