Please check whether the problems persists when changing
carFollowModel="W99" to "Krauss".
Either way, the video suggests that there is a bug somewhere and I'd ask
you to provide input files for reproducing the jumpy behavior.

Am Fr., 18. Feb. 2022 um 16:43 Uhr schrieb Ahmad Abdallah via sumo-user <
[email protected]>:

> Thanks for your response, Jakob.
>
> The problem with reducing lcAssertive to smaller values as it seems to me
> is that in some cases merging vehicles coming from on-ramps won't find a
> long enough gap to merge before the end of the acceleration lane, and this
> creates a repetitive stopping and bottleneck instances downstream of
> merging segments. I noticed that even if I set high values for lcStrategic
> this problem keeps happening. I thought this is probably linked to the min
> gap parameters, but I'm using the default *tau*(which I believe is low
> enough) and here are my parameters for the car following and lane
> changing models. Do you see any other sources of trouble?
>
> [image: image.png]
>
> One other problem that I'm facing is lateral instability. I am not sure if
> this is the correct term, but basically in some occasions vehicles tend to
> perform a series of lane changes within a few seconds time frame,
> especially upstream of the location of the major bottleneck location
> (interchange) in the freeway I'm simulating . It looks as if these
> vehicles(drivers) are hesitant and cannot decide which lane to drive in. I
> think in most cases the cause of those abrupt lane changes is the
> LcSpeedGain parameter. However, this parameter has proved vital in getting
> vehicles to shift to lanes with less traffic density when being in a more
> congested lane. I'm attaching a short recording which displays this
> problem. Any thoughts?
>
>
> Best regards,
> Ahmad Abdallah.
>
>
> On Thu, Feb 17, 2022 at 11:33 AM Jakob Erdmann <[email protected]>
> wrote:
>
>> lcAssertive 30 is quite likely to cause this. It reduces the acceptable
>> gap size by a factor of 30 and thus causes lane-changing into short gaps
>> that are very unsafe.
>> I'd recommend keeping lcAssertive well below 2 (though even a value of
>> 1.1 may cause collision in exceptional circumstances).
>>
>> Am Do., 17. Feb. 2022 um 16:09 Uhr schrieb Ahmad Abdallah via sumo-user <
>> [email protected]>:
>>
>>> Hello,
>>>
>>>
>>>
>>> While running a simulation in SUMO, I am facing a problem which is that
>>> vehicles are colliding and running over each other when changing lanes, as
>>> shown in the attached picture. Do you have any idea how I can solve this
>>> problem? Could it be due to high LcAssertive value? I am currently using 30.
>>>
>>>
>>>
>>> To give more context, I just disabled the teleporting when colliding
>>> option. So now I only get warnings when vehicles collide and no teleporting
>>> is done except in instances of gridlocks.
>>>
>>>
>>> [image: Sumo_Snippet.jpg]
>>>
>>> Best regards,
>>> Ahmad Abdallah.
>>> _______________________________________________
>>> sumo-user mailing list
>>> [email protected]
>>> To unsubscribe from this list, visit
>>> https://www.eclipse.org/mailman/listinfo/sumo-user
>>>
>> _______________________________________________
>> sumo-user mailing list
>> [email protected]
>> To unsubscribe from this list, visit
>> https://www.eclipse.org/mailman/listinfo/sumo-user
>>
> _______________________________________________
> sumo-user mailing list
> [email protected]
> To unsubscribe from this list, visit
> https://www.eclipse.org/mailman/listinfo/sumo-user
>
_______________________________________________
sumo-user mailing list
[email protected]
To unsubscribe from this list, visit 
https://www.eclipse.org/mailman/listinfo/sumo-user

Reply via email to