Skip to main content
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [sumo-user] Lateral offset 1.0 causes perma-blockage
Hello Jakob,

Upon further inspection, I set the lane change mode to 101000000000 (2560) which allows sublane changes even when overriding TraCI requests to force center lateral alignment of vehicles in lane. However, there still seemed to be some issues with strategic lane changing (Even if the lane change mode clearly did not allow it). By setting arrivalLane="random" instead of the default "current", there were times when the vehicle's lane change state looks like ['stay', 'strategic', 'TraCI'] even when the action in changeLaneRelative was "right" (-1). This caused the same effect where the vehicle was on the edge of a lane instead of the center. This seemed to disappear after setting arrivalLane to "current".

On the same line of questioning, it complicates the scenario, even more, when I require any decision to change lane by vehicles to take place 5 seconds after the end of the last lane-change maneuver by them. I tried looking into action step length, but that seems to be for a different purpose. So I ended up using a custom counter to determine at which steps the changeLaneRelative statement will actually be used. At the steps where I don't want the vehicles to make any lane changing decisions, I simply do nothing. Just to confirm, that would be the "correct" way to set this up?

Thank you.

Sincerely,
Hriday


On Tue, 3 May 2022 at 10:35, Hriday Sanghvi <sanghvih@xxxxxx> wrote:
Hello Jakob,

Without any TraCI intervention, vehicles seem to stay in the center of their lane.

But with TraCI intervention, even setting the lane change mode to default (1621), the problem seems to be the changeLaneRelative (https://sumo.dlr.de/pydoc/traci._vehicle.html#VehicleDomain-changeLaneRelative) command combined with <lateral-resolution value="3.2" /> in the SUMO config file. The width of the vehicle is the default 1.8 m. I have set the lateral resolution to 3.2 since the default lane width is 3.2 m, and I want it to consider the entire width of the lane as 1 sublane and not create divisions within the lane. At least, that's what I thought would make sure the vehicles stay in the center of their lanes. But that does not seem to be happening. 

I also tried setting the vehicle width to 3.2 m, but that also changed the time it took to complete the simulation and I don't understand why (clearly changing the width of the vehicle implied something else and the behaviour of lane-changing also changed). So maybe that has something to do with minGapLat being 0.6? I am unsure of what's going on. Please advise.

Thank you.

Sincerely,
Hriday


On Thu, 28 Apr 2022 at 10:23, Jakob Erdmann <namdre.sumo@xxxxxxxxx> wrote:
moving within a lane is the 'sublane' motivation. The problem should disappear with laneChangeMode to 1537

Am Do., 28. Apr. 2022 um 10:57 Uhr schrieb Hriday Sanghvi via sumo-user <sumo-user@xxxxxxxxxxx>:
Hello,

By some chance, there are certain vehicles that seem to stick to a side of the lane when using the sublane model (SL2015). In that case, they cannot get past the blockage on the other lane because they are too close to it. They also cannot perform a "lane change" to get away from the blocked lane because they are already on the unblocked lane (but on the far left edge of it closest to the blockage; please see attached screenshot).

The lane change mode is set to 513 (which means apart from collision avoidance and safety-gap enforcement, SUMO only controls strategic lane changes - which I assume happen at the end of the route during arrival time?)

The relevant parameters I've identified that control this lateral alignment

departPosLat, default: "center"
arrivalPosLat, no default (not set)
latAlignment, default "center"
minGapLat, default: 0.6
lcSublane, default: 1.0
lcPushy, default: 0
lcPushyGap, default: minGapLat
lcLaneDiscipline, default: 0

What combination of set parameters would ensure that the vehicle will always treat a lane change such that it maintains a central position at all times?

Thank you.

Sincerely,
Hriday
_______________________________________________
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user
_______________________________________________
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user

Back to the top