Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [sumo-user] Crashing vehicles intentionally

Totally understood.

Thank you Mr. Erdmann.

On Wed, Apr 15, 2020 at 3:02 PM Jakob Erdmann <namdre.sumo@xxxxxxxxx> wrote:
You will need to have actionStepLength > tau to create unsafe driving in most models. The difference can be very small as long as you make sure that vehicles follow each other closely.
Note that you can configure whether you want to flag "physical" collisions or minGap violations ( --collision.mingap-factor). If vehicles are configured with minGap > 0 and you only detect physical collisions then the margin of error must be larger to overcome the minGap.

Am Mi., 15. Apr. 2020 um 13:11 Uhr schrieb Angelos Michalopoulos <michalopoulos@xxxxxxxx>:
Ok,

Thank you very much. This works indeed now.

In this case though the parameter changed is the tau as far as I see.
To create a similar behaviour by changing the reaction time, I am experimenting with action step length as suggested in:

(tau is not reaction time, I am aware of that.)

Would you say that there is a minimum difference between step-length and actionsteplength to have a behaviour similar to the one of the example?

Thank you very much.


On Fri, Apr 3, 2020 at 7:43 PM Jakob Erdmann <namdre.sumo@xxxxxxxxx> wrote:
You can download at test case where vehicles with minGap zero collide at https://sumo.dlr.de/extractTest.php by entering 'sumo/cf_model/tau/step1.0_tau0.5/krauss' in the test path field.

Am Fr., 3. Apr. 2020 um 12:47 Uhr schrieb Angelos Michalopoulos <michalopoulos@xxxxxxxx>:
Thank you for your answer Mr. Erdmann.

I can generate collisions with varying mingap distances and maxspeed conditions. However I can not do it when the distance is set to zero and have a physical collision.

I have tried increasing the speed as well and changing the model to IDM,W99 and change their parameters as well to create unsafe reaction conditions.

So my problem is not in the warning,this is issued, I can't actually create the crash in the simulation. I think I do not find the correct correlation in the parameters.

As far as I understand the safety checks can only be disabled in TraCI so my problem could be that I use sumo. But I am not trying to generate the collision at a specific time. My understanding may be incorrect and I would be happy to be wrong in this case.

Thank you!

On Fri, 3 Apr 2020, 11:56 Jakob Erdmann, <namdre.sumo@xxxxxxxxx> wrote:

Am Fr., 3. Apr. 2020 um 10:28 Uhr schrieb Angelos Michalopoulos <michalopoulos@xxxxxxxx>:
Hello everyone!
I would like to ask a question regarding collisions.

I am trying to create collisions, but with mingap==0.

I can create collisions but the vehicle(s) always avoid the physical collision in the scenario.

I am working on a simple scenario with four vehicles in a highway of one lane. The length is two km. Every car departs on max speed (tried from 30 to 60 m/s) and the first stops after a km suddenly. I have tried changing some parameters like tau, but I was unable to do what I am trying to do and I couldn't find specifically what I am looking for.

Thank you very much!

Regards,

Angelos
_______________________________________________
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
_______________________________________________
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
_______________________________________________
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