Skip to main content
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [sumo-user] Strange driverbehavior with the sublanemodel
Hello,
you will get better results if you set a higher sublane resolution value (e.g. 0.8).
Note, that this does not restrict the lateral placement of vehicles which is always continuous. Instead it reduced the choice set of drivers.

Looking at your scenario I get the impression that
a) there is a bug in the handling of blocked maneuvers that results in vehicles becoming stuck rather than moving back to their preferred alignment (https://github.com/eclipse/sumo/issues/6084)
b) there may be issues when simulating with low lateral resolution values - something that is not part of our regular tests

regards,
Jakob






Am Fr., 13. Sept. 2019 um 09:52 Uhr schrieb Wlodarczak, Tessa <tessa.wlodarczak@xxxxxxxxxxxxxxxxxxxxxxxxxxx>:

Hello Sumo-users,

 

I have a question concerning the Sublane-model. I got a lot of problems with the cars decentering and it does not look valid to me.

My testcases include an flow of 5000 cars/h (blue cars, (laneChange= SL) and everything on default) on a straight road with 5 lanes (Sumofiles in the zip). I put the steplength to 0.5 and the resolution to 0.1.  I was trying to define an defensive drivermodel that’s going within the flow. Picture 4 shows the case with this parameterization (red car):

 

<vType id="vType_0" minGap="3.00" maxSpeed="60" speedFactor="0.8" minGapLat="0.15" laneChangeModel="SL2015" tau="1" lcCooperative="1" lcSpeedGain="0.5" lcKeepRight="0.5" lcOpposite="0.5" lcAssertive="0.5" />

 

The ego (red car) starts on the left lane and starts a lanechange but gets stuck on the roadmarking there, and also several blue cars from the flow get stuck or drive on or close to the roadmarking for a long distance. Even though I did not change the lateral alignment and the default is supposed to be center (also setting the parameter to center does not change anything).

 

Picture 5 shows the ego with the same parameterization just “lcKeepRight=”5””. I don’t get why the two cars (red and blue) are going on the roadmarking for 100ds of seconds until the blue passed on the right.

 

What can I do to create a more valid testcase? What kind of parameters do I have to change so the cars don’t behave this strange?

 

Thanks and greetings,

Tessa

 

_______________________________________________
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://www.eclipse.org/mailman/listinfo/sumo-user

Back to the top