Adding a check for endPos < 0 into MSVehicle.cpp planMoveInternal solves
the issue and doesn't impact SUMO test outputs - obviously I don't know enough
to know whether this will have wider impacts! (setting to 0 or NUMERICAL_EPS
does not work.)
if (stop.parkingarea != nullptr) {
// leave enough space so parking vehicles can exit
endPos = stop.parkingarea->getLastFreePosWithReservation(t,
*this);
if (endPos < 0) endPos = POSITION_EPS;
}
cheers
div
‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
On Sunday, 21 July 2019 20:05, The div <[email protected]> wrote:
> Scenario:
> 1. Current (today) github build on 64bit windows 10
> (& current win 10 release v1_2_0+0765-3423b2ab8c)
> 2. Parking area rerouter configured
> 3. Parking area close to junction (startPos="2.6")
> 4. The first vehicle reaching the junction after the parking area is full
> gets rerouted as expected
> 5. The next vehicle simply stops on the junction and waits for a space in the
> parking area.
>
> Is this behaviour an expected artefact of the simulation?
>
> Scenario can be demonstrated using the build test:
> sumo\extended\rerouter\parking\congestion_alternatives :
>
> If parking_c is set to startPos="2.601" or less then traffic is stopped
> once the parking area is full.
> If this is set to startPos="2.6011" traffic flows normally. I have seen
> equivalent behaviour on other layouts with positions of 3.4/3.4011 but can't
> identify the layout difference causing this.
>
> In trying to pin this down it appears to be related to the vehicle speed in
> the last step being set fractionally low such that the vehicle does not quite
> get to the end of the junction. (myState.myPos < myLane->getLength() stays
> true with vehicle stopped) I had to set debug output to 4 decimal places to
> clarify this.
>
> Unfortunately I don't understand the code well enough to work out where the
> incorrect computation, (if that is what it is), is happening. Happy to
> continue looking if anyone can give me a pointer as to where I might look.
>
> cheers
> div_______________________________________________
sumo-user mailing list
[email protected]
To change your delivery options, retrieve your password, or unsubscribe from
this list, visit
https://www.eclipse.org/mailman/listinfo/sumo-user
