The entry of vehicles in the network is not synchronous between OMNET and SUMO

[Rayane Sibaï]

Hello,
I am trying to test an algorithm for the communication between vehicles in VANET. I have moved with OMNET and sumo and all is well. I defined the value of step-length value = "0.1" in both OMNET and SUMO. But the problem is that each vehicle enters and leave the network in OMNET in different time from that in sumo. For example, a vehicle A enters the network in OMNET at time = 35 sec , while it enter the network in sumo at time = 45 sec.
Someone has an idea?

[Christoph Sommer]

Dear Rayane Sibaï,

which version of Veins are you using?


Best,

Christoph

Rayane Sibaï wrote:
> Hello,
> I am trying to test an algorithm for the communication between vehicles
> in VANET. I have moved with OMNET and sumo and all is well. I defined

> the value of step-length value = "0.1" in both OMNET and suSUMO. But the

> problem is that each vehicle enters and leave the network in OMNET in
> different time from that in sumo. For example, a vehicle A enters the
> network in OMNET at time = 35 sec , while it enter the network in sumo
> at time = 45 sec.
> Someone has an idea?

--
Dr. Christoph Sommer
Distributed Embedded Systems Group
University of Paderborn, Germany
http://www.ccs-labs.org/~sommer/

[Rayane Sibaï]

Veins 3a3 with omnet 4.2.2

--
You received this message because you are subscribed to a topic in the Google Groups "omnetpp" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/omnetpp/947SWQkhHHU/unsubscribe.
To unsubscribe from this group and all its topics, send an email to omnetpp+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Christoph Sommer]

Dear Rayane Sibaï,

this sounds almost impossible. Did you modify Veins in any way? As soon
as (due to misconfiguration or program error) the SUMO and OMNeT++
timesteps diverged, an exception should be raised by code in [1].

Are you absolutely sure that you are looking at the same vehicle?

If so, I would be happy to investigate if you could send me a minimal
simulation that exhibits the described error so I can check for myself.

Best,

Christoph


[1]
https://github.com/sommer/veins/blob/veins-3a3/src/modules/mobility/traci/TraCIScenarioManager.cc#L664

>> omnetpp+u...@googlegroups.com.

[Rayane Sibaï]

Dear Christoph,

Thx for your reply, yes I am, I looked at several vehicles, and it is the same problem. And to be sure, I looked at the trace file generated by sumo and those generated by OMNeT, there is a difference. And forthe error, I have not received such an error; the simulation excecute and terminates normally.

And there is another problem : the speed of vehicles in erlangen.rou.xml is between 80 and 120 m / sec, but in the simulation, the speed of any vehicle is between 0 and 50 m / sec!

note : I have made my own network (roads, trajectories and vehicles)

On Friday, November 21, 2014 5:44:48 AM UTC-8, Rayane Sibaï wrote:

Hello,
I am trying to test an algorithm for the communication between vehicles in VANET. I have moved with OMNET and sumo and all is well. I defined the value of step-length value = "0.1" in both OMNET and SUMO. But the problem is that each vehicle enters and leave the network in OMNET in different time from that in sumo. For example, a vehicle A enters the network in OMNET at time = 35 sec , while it enter the network in sumo at time = 45 sec.
Someone has an idea?

[Christoph Sommer]

Rayane Sibaï wrote:
> Christoph Sommer wrote:

>> Rayane Sibaï wrote:
>>>
>>> the problem is that each vehicle enters and leave the network in OMNET in different time from that in sumo. For example, a
>>> vehicle A enters the network in OMNET at time = 35 sec , while it enter the network in sumo at time = 45 sec.
>>

>> this sounds almost impossible. Did you modify Veins in any way?

>> Are you absolutely sure that you are looking at the same vehicle?
>

> Thx for your reply, yes I am, I looked at several vehicles, and it is
> the same problem.

I have run your simulation in Veins (version 3a3) and have failed to
reproduce your problems: all time steps are perfectly synchronized.

I have uploaded a screen capture of the running simulation:

http://www.ccs-labs.org/~sommer/intra/dbb1e90d8df5732c3a5c31a5b6a8a418-screen.webm

> And there is another problem : the speed of vehicles in erlangen.rou.xml
> is between 80 and 120 m / sec, but in the simulation, the speed of any
> vehicle is between 0 and 50 m / sec!

I have also looked at the vehicle speeds: both simulators report the
exact same vehicle speeds.

Also note that your vehicles will likely be unable at the 120 m/s you
configured. This is more than one third the speed of sound -- more than
the speed limit of most cities.


Best,

Christoph

[Rayane Sibaï]

Dear Christoph, thank you very much,

What you try is to run the simuation in OMNeT+sumo at the same time, but my problem is that there is a difference between the result of the simulation in OMNeT+sumo  and sumo alone.

(If I enter sumo and I run the simulation and I generates the trace file(result 1), and then I run the simulation in OMNeT+sumo and I look at the results of the simulation (result 2), I find that there is a difference of a few seconds between these two results .

On Friday, November 21, 2014 3:44:48 PM UTC+2, Rayane Sibaï wrote:

Hello,
I am trying to test an algorithm for the communication between vehicles in VANET. I have moved with OMNET and sumo and all is well. I defined the value of step-length value = "0.1" in both OMNET and SUMO. But the problem is that each vehicle enters and leave the network in OMNET in different time from that in sumo. For example, a vehicle A enters the network in OMNET at time = 35 sec , while it enter the network in sumo at time = 45 sec.
Someone has an idea?