Re: [sumo-user] monitoring headway distances in simulation

[Jakob Erdmann <namdre.sumo@xxxxxxxxxxxxxx>]

Hello Gerald,

- in a --netstate-dump file the vehicles are writen for each lane in downstream order so you can trivially determine leader-follower relationshops. Figuring out leaders across multiple (short) edges is considerably more involved

- a vehicle that changes lanes continously is considered to have changed once its centerline (aligned with the driving direction) has moved onto the new lane. A vehicle may also occupy another lane laterally, partially (called 'shadow lane' in sumo-gui) and it may have leader/follower relationships with vehicles on that shadow lane as well (as soon as there is lateral overlap). There is currently no convenient method to retrieve all these leaders (i.e 2 bicycles in front on the current lane and another vehicle partially in front on the shadow lane)

- traci supports vehicle.getLeader() which returns an ID and a distance but does not support the sublane case where there are multiple leaders. I suppose a new method getSublaneLeaders is needed here.

regards,

Jakob

2018-05-07 11:18 GMT+02:00 Richter Gerald <Gerald.Richter@xxxxxxxxx>:

Hi,

what is the best way to evaluate the headway distances over the course
of a simulation run?
That supposes post-execution FCD is available, now I want to do my
calculations.

* Do I have to store the lane info for each vehicle and group the FCD
and search for the leading vehicle on the current lane?

* How is the situation on lane-change with/without sub-lanes, meaning:
when does a leading vehicle start/stop to be a leading vehicle?

* Is there another way to get this data online (during the simulation run)?


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