[sumo-user] Vehicles slowing at junctions on high priority road

[Kathy Jang]

Hi there,

I'm constructing a roundabout (image attached). When running on sumo-gui,
I'm seeing strange behavior where the vehicles on the higher priority edges
in the roundabout *appear* to be slowing down as they approach junctions,
which should not be the case.

I ran a control experiment with a single IDM vehicle in the roundabout, and
no other vehicles. When I plot the acceleration of this vehicle, the
standard deviation is close to 0 and there are no issued accelerations that
would suggest the vehicle is slowing down or doing anything other than
maintaining a constant velocity.

Could this be an issue with the GUI rendering, or something else?

All my best,

Kathy Jang
University of California, Berkeley | 2018
___
sumo-user mailing list
sumo-user@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/sumo-user

Re: [sumo-user] TL question

[Jakob Erdmann]

Hello,
there is currently no option for sharing programs among traffic lights.
Each  definition must include a traffic light id and you will have
to copy the program for each traffic light that shall be controlled by it.
regards,
Jakob

2018-08-23 19:07 GMT+02:00 mohsen hs :

> Hi Jose,
>
> TL can be defined in the additional.xml as stated here
> http://sumo.dlr.de/wiki/Simulation/Traffic_Lights#
> Defining_New_TLS-Programs
> 
>
> and additional.xml can be called as input in cfg file as stated here
> http://sumo.dlr.de/wiki/SUMO#Input
>
> Bests
> Mohsen
>
>
>
>
>
>
>
>
>
>
> 
> On Friday, August 24, 2018, 4:28:30 AM GMT+12, Jose Monreal Bailey <
> jmonr...@gmail.com> wrote:
>
>
> Hey everyone,
>
> I have a Traffic Light program defined. I want to use that program Id on
> all my Traffic lights.
> How can I reference that program id to the different Traffic LIghts in my
> network file?
> And where should I store this traffic light definition?
>
> Best regards,
> José
> ___
> sumo-user mailing list
> sumo-user@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/sumo-user
>
> ___
> sumo-user mailing list
> sumo-user@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/sumo-user
>
>
___
sumo-user mailing list
sumo-user@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/sumo-user

[sumo-user] Intelligent traffic light

[mohsen hs]

Dear sumo users, 

I am interested to make my traffic lights a bit more intelligent, so they would 
not have static phases (through runtime). Is there such a configuration in 
SUMO, or I should get the edge info and update the connected traffic lights 
through TraCI?
I have seen tlccyclecadaptation.py exists, but it needs the route as input, so 
I guess it cannot be used dynamically in the simulation.

Many thanksMohsen
___
sumo-user mailing list
sumo-user@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/sumo-user

[sumo-user] TL question

[Jose Monreal Bailey]

Hey everyone,

I have a Traffic Light program defined. I want to use that program Id on
all my Traffic lights.
How can I reference that program id to the different Traffic LIghts in my
network file?
And where should I store this traffic light definition?

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

Re: [sumo-user] Jammed Pedestrians

[Jose Monreal Bailey]

the Jammed is produced by the shape. So after adding vClass="pedestrian"
everything worked perfectly.

Best!

On Thu, 23 Aug 2018 at 14:44, Jose Monreal Bailey 
wrote:

> Hey everyone,
>
> I am getting Jammed pedestrians as you can see in the picture.
> I am running sumo master hash 7b378d18a18b4696bd84ea907adf10d6da8b4d97
> commit.
> How can I fix this? I generated the pedestrians trips with randomTrips.py
>
> Best regards,
> José
>
>
> [image: image.png]
>
___
sumo-user mailing list
sumo-user@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/sumo-user

Re: [sumo-user] Person walking

[Jose Monreal Bailey]

it was I was missing vClass="pedestrian" in my vType definition.

Thanks again !

On Thu, 23 Aug 2018 at 14:59, Jose Monreal Bailey 
wrote:

> I just figured out I was using vType instead of vClass.
>
> Thanks!
>
> On Thu, 23 Aug 2018 at 14:26, Jose Monreal Bailey 
> wrote:
>
>> Hi Jakob,
>>
>> Sorry, but this is the shape I am seeing when I set up the vType
>>
>> [image: image.png]
>>
>> And this is the shape without setting a vType
>>
>> [image: image.png]
>>
>> The latter one does look like a person.
>> That's why I said for the first a Hotdog, because at first I had it with
>> a orange colour.
>>
>> How can I get the yellow shape instead of the first one?
>>
>> Best regards,
>> José
>>
>> On Thu, 23 Aug 2018 at 13:25, Jakob Erdmann 
>> wrote:
>>
>>> The guiShape is only used when setting the person drawing mode to
>>> 'simple shapes' (the default is the triangle)in the sumo-gui visualisation
>>> settings dialog. This also happens when you select the preset 'real world'.
>>>
>>> The default guiShape shape is indeed the 'hotdog' as you have so aptly
>>> put it. It's supposed to suggest shoulders and a head. Look closely and you
>>> can even see the nose.
>>>
>>> 2018-08-23 13:37 GMT+02:00 Jose Monreal Bailey :
>>>
 I was able to create my pedestrian type, but when I use guiShape
 pedestrian I get a triangle, and in real world mode a hotdog shape image.
 What's the guiShape for pedestrians by default, how do I force a vType
 to use that shape?

 Best regards,
 José

 On Thu, 23 Aug 2018 at 12:08, Jose Monreal Bailey 
 wrote:

> Hey everyone,
>
> I want to define a pedestrian, similar as I do with vehicles.
> Creating a vType and then giving this type to the person or pedestrian.
> Is this valid or possible? Or what would be the best way to do it?
>
> Best regards,
> José
>

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


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

Re: [sumo-user] Person walking

[Jose Monreal Bailey]

I just figured out I was using vType instead of vClass.

Thanks!

On Thu, 23 Aug 2018 at 14:26, Jose Monreal Bailey 
wrote:

> Hi Jakob,
>
> Sorry, but this is the shape I am seeing when I set up the vType
>
> [image: image.png]
>
> And this is the shape without setting a vType
>
> [image: image.png]
>
> The latter one does look like a person.
> That's why I said for the first a Hotdog, because at first I had it with a
> orange colour.
>
> How can I get the yellow shape instead of the first one?
>
> Best regards,
> José
>
> On Thu, 23 Aug 2018 at 13:25, Jakob Erdmann  wrote:
>
>> The guiShape is only used when setting the person drawing mode to 'simple
>> shapes' (the default is the triangle)in the sumo-gui visualisation settings
>> dialog. This also happens when you select the preset 'real world'.
>>
>> The default guiShape shape is indeed the 'hotdog' as you have so aptly
>> put it. It's supposed to suggest shoulders and a head. Look closely and you
>> can even see the nose.
>>
>> 2018-08-23 13:37 GMT+02:00 Jose Monreal Bailey :
>>
>>> I was able to create my pedestrian type, but when I use guiShape
>>> pedestrian I get a triangle, and in real world mode a hotdog shape image.
>>> What's the guiShape for pedestrians by default, how do I force a vType
>>> to use that shape?
>>>
>>> Best regards,
>>> José
>>>
>>> On Thu, 23 Aug 2018 at 12:08, Jose Monreal Bailey 
>>> wrote:
>>>
 Hey everyone,

 I want to define a pedestrian, similar as I do with vehicles.
 Creating a vType and then giving this type to the person or pedestrian.
 Is this valid or possible? Or what would be the best way to do it?

 Best regards,
 José

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

Re: [sumo-user] Person walking

[Jose Monreal Bailey]

Hi Jakob,

Sorry, but this is the shape I am seeing when I set up the vType

[image: image.png]

And this is the shape without setting a vType

[image: image.png]

The latter one does look like a person.
That's why I said for the first a Hotdog, because at first I had it with a
orange colour.

How can I get the yellow shape instead of the first one?

Best regards,
José

On Thu, 23 Aug 2018 at 13:25, Jakob Erdmann  wrote:

> The guiShape is only used when setting the person drawing mode to 'simple
> shapes' (the default is the triangle)in the sumo-gui visualisation settings
> dialog. This also happens when you select the preset 'real world'.
>
> The default guiShape shape is indeed the 'hotdog' as you have so aptly put
> it. It's supposed to suggest shoulders and a head. Look closely and you can
> even see the nose.
>
> 2018-08-23 13:37 GMT+02:00 Jose Monreal Bailey :
>
>> I was able to create my pedestrian type, but when I use guiShape
>> pedestrian I get a triangle, and in real world mode a hotdog shape image.
>> What's the guiShape for pedestrians by default, how do I force a vType to
>> use that shape?
>>
>> Best regards,
>> José
>>
>> On Thu, 23 Aug 2018 at 12:08, Jose Monreal Bailey 
>> wrote:
>>
>>> Hey everyone,
>>>
>>> I want to define a pedestrian, similar as I do with vehicles.
>>> Creating a vType and then giving this type to the person or pedestrian.
>>> Is this valid or possible? Or what would be the best way to do it?
>>>
>>> Best regards,
>>> José
>>>
>>
>> ___
>> sumo-user mailing list
>> sumo-user@eclipse.org
>> To change your delivery options, retrieve your password, or unsubscribe
>> from this list, visit
>> https://dev.eclipse.org/mailman/listinfo/sumo-user
>>
>>
> ___
> sumo-user mailing list
> sumo-user@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/sumo-user
>
___
sumo-user mailing list
sumo-user@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/sumo-user

Re: [sumo-user] Person walking

[Jakob Erdmann]

The guiShape is only used when setting the person drawing mode to 'simple
shapes' (the default is the triangle)in the sumo-gui visualisation settings
dialog. This also happens when you select the preset 'real world'.

The default guiShape shape is indeed the 'hotdog' as you have so aptly put
it. It's supposed to suggest shoulders and a head. Look closely and you can
even see the nose.

2018-08-23 13:37 GMT+02:00 Jose Monreal Bailey :

> I was able to create my pedestrian type, but when I use guiShape
> pedestrian I get a triangle, and in real world mode a hotdog shape image.
> What's the guiShape for pedestrians by default, how do I force a vType to
> use that shape?
>
> Best regards,
> José
>
> On Thu, 23 Aug 2018 at 12:08, Jose Monreal Bailey 
> wrote:
>
>> Hey everyone,
>>
>> I want to define a pedestrian, similar as I do with vehicles.
>> Creating a vType and then giving this type to the person or pedestrian.
>> Is this valid or possible? Or what would be the best way to do it?
>>
>> Best regards,
>> José
>>
>
> ___
> sumo-user mailing list
> sumo-user@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/sumo-user
>
>
___
sumo-user mailing list
sumo-user@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/sumo-user

Re: [sumo-user] Person walking

[Jose Monreal Bailey]

I was able to create my pedestrian type, but when I use guiShape pedestrian
I get a triangle, and in real world mode a hotdog shape image.
What's the guiShape for pedestrians by default, how do I force a vType to
use that shape?

Best regards,
José

On Thu, 23 Aug 2018 at 12:08, Jose Monreal Bailey 
wrote:

> Hey everyone,
>
> I want to define a pedestrian, similar as I do with vehicles.
> Creating a vType and then giving this type to the person or pedestrian.
> Is this valid or possible? Or what would be the best way to do it?
>
> Best regards,
> José
>
___
sumo-user mailing list
sumo-user@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/sumo-user

Re: [sumo-user] SpeedFactor question

[Jakob Erdmann]

For vehicles, the factor is applied to the speed limit of edges/lanes
whereas for pedestrians it is applied to the maxSpeed of their type. (see
http://sumo.dlr.de/wiki/Definition_of_Vehicles,_Vehicle_Types,_and_Routes#Additional_remarks_on_speed_distributions
)

2018-08-23 12:28 GMT+02:00 Jose Monreal Bailey :

> Thanks for the clarification Jakob.
> And those are factors from the speed defined for the lane, pavement, or is
> it taken from the maxSpeed defined by the vehicle type?
> I want to define a speed factor for my pedestrians.
>
> Best regards,
> José
>
> On Thu, 23 Aug 2018 at 11:24, Jakob Erdmann  wrote:
>
>> the 'c' in normc stans for 'normal distribution with cut-off'. The
>> parameters 0.2 and 2 are the cut-off parameters and this means the sampled
>> speedFactor values will allways be in the range [0.2, 2].
>>
>> Tne range of 80%-120% comes from the fact that for a normal distribution,
>> 95% of the results are within 2*SD of the mean. And sd =0.1 causes 95% of
>> the factors to be within [0.8,1.2].
>>
>> 2018-08-23 12:18 GMT+02:00 Jose Monreal Bailey :
>>
>>> Hey everyone,
>>>
>>> I am trying to understand in the wiki
>>> 
>>> the following sentence:
>>>
>>> Using *speedFactor*="normc(1,0.1,0.2,2)" will result in a speed
>>> distribution where 95% of the vehicles drive between 80% and 120% of the
>>> legal speed limit.
>>>
>>> As stated before the parameters are (MEAN, SD, MIN, MAX), so for max if
>>> its 2, shouldn't it be twice ? 200%? and by MIN I am guessing that it
>>> reduces by that factor, right?
>>>
>>> Best regards,
>>> José
>>>
>>> ___
>>> sumo-user mailing list
>>> sumo-user@eclipse.org
>>> To change your delivery options, retrieve your password, or unsubscribe
>>> from this list, visit
>>> https://dev.eclipse.org/mailman/listinfo/sumo-user
>>>
>>>
>> ___
>> sumo-user mailing list
>> sumo-user@eclipse.org
>> To change your delivery options, retrieve your password, or unsubscribe
>> from this list, visit
>> https://dev.eclipse.org/mailman/listinfo/sumo-user
>>
>
> ___
> sumo-user mailing list
> sumo-user@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/sumo-user
>
>
___
sumo-user mailing list
sumo-user@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/sumo-user

Re: [sumo-user] SpeedFactor question

[Jose Monreal Bailey]

Thanks for the clarification Jakob.
And those are factors from the speed defined for the lane, pavement, or is
it taken from the maxSpeed defined by the vehicle type?
I want to define a speed factor for my pedestrians.

Best regards,
José

On Thu, 23 Aug 2018 at 11:24, Jakob Erdmann  wrote:

> the 'c' in normc stans for 'normal distribution with cut-off'. The
> parameters 0.2 and 2 are the cut-off parameters and this means the sampled
> speedFactor values will allways be in the range [0.2, 2].
>
> Tne range of 80%-120% comes from the fact that for a normal distribution,
> 95% of the results are within 2*SD of the mean. And sd =0.1 causes 95% of
> the factors to be within [0.8,1.2].
>
> 2018-08-23 12:18 GMT+02:00 Jose Monreal Bailey :
>
>> Hey everyone,
>>
>> I am trying to understand in the wiki
>> 
>> the following sentence:
>>
>> Using *speedFactor*="normc(1,0.1,0.2,2)" will result in a speed
>> distribution where 95% of the vehicles drive between 80% and 120% of the
>> legal speed limit.
>>
>> As stated before the parameters are (MEAN, SD, MIN, MAX), so for max if
>> its 2, shouldn't it be twice ? 200%? and by MIN I am guessing that it
>> reduces by that factor, right?
>>
>> Best regards,
>> José
>>
>> ___
>> sumo-user mailing list
>> sumo-user@eclipse.org
>> To change your delivery options, retrieve your password, or unsubscribe
>> from this list, visit
>> https://dev.eclipse.org/mailman/listinfo/sumo-user
>>
>>
> ___
> sumo-user mailing list
> sumo-user@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/sumo-user
>
___
sumo-user mailing list
sumo-user@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/sumo-user

Re: [sumo-user] SpeedFactor question

[Jakob Erdmann]

the 'c' in normc stans for 'normal distribution with cut-off'. The
parameters 0.2 and 2 are the cut-off parameters and this means the sampled
speedFactor values will allways be in the range [0.2, 2].

Tne range of 80%-120% comes from the fact that for a normal distribution,
95% of the results are within 2*SD of the mean. And sd =0.1 causes 95% of
the factors to be within [0.8,1.2].

2018-08-23 12:18 GMT+02:00 Jose Monreal Bailey :

> Hey everyone,
>
> I am trying to understand in the wiki
> 
> the following sentence:
>
> Using *speedFactor*="normc(1,0.1,0.2,2)" will result in a speed
> distribution where 95% of the vehicles drive between 80% and 120% of the
> legal speed limit.
>
> As stated before the parameters are (MEAN, SD, MIN, MAX), so for max if
> its 2, shouldn't it be twice ? 200%? and by MIN I am guessing that it
> reduces by that factor, right?
>
> Best regards,
> José
>
> ___
> sumo-user mailing list
> sumo-user@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/sumo-user
>
>
___
sumo-user mailing list
sumo-user@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/sumo-user

[sumo-user] SpeedFactor question

[Jose Monreal Bailey]

Hey everyone,

I am trying to understand in the wiki

the following sentence:

Using *speedFactor*="normc(1,0.1,0.2,2)" will result in a speed
distribution where 95% of the vehicles drive between 80% and 120% of the
legal speed limit.

As stated before the parameters are (MEAN, SD, MIN, MAX), so for max if its
2, shouldn't it be twice ? 200%? and by MIN I am guessing that it reduces
by that factor, right?

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

Re: [sumo-user] Metrics

[Jose Monreal Bailey]

They are. The problem is that for right most lane they can do it at phase
p. But to go straight is phase k. And vehicles going straight are also
using that lane. That lane is to go straight and right.


On Thu, 23 Aug 2018, 09:43 Jakob Erdmann,  wrote:

> If the vehicles are not allowed to go straight from the rightmost lane,
> then you can just remove the connection to avoid this behavior.
>
> 2018-08-23 10:41 GMT+02:00 Jose Monreal Bailey :
>
>> Hi Jakob,
>>
>> I have a 3 lane edge which 1 of the them is used to turn right (with left
>> side traffic).
>> This means I have a TL setup for that phase.
>> Sometimes I have vehicles using that lane, which are not going to turn,
>> they use it as they are going to go through to the next edge in front of
>> them.
>> So for this cases, I would like to know how many vehicles are impacted by
>> this reckless drivers.
>> I think I would have to check everytime at that point if there is a
>> vehicle blocking and start counting how many were impacted and for how long.
>>
>> Best regards,
>> José
>>
>> On Thu, 23 Aug 2018 at 09:34, Jakob Erdmann 
>> wrote:
>>
>>> Hello,
>>> there is no dedicated tool for this in the toolbox. You could parse
>>> lanes from from the netstate dump or fcd output and check for specific
>>> sequences of inappropriate lane use. Computing the inpact of this behavior
>>> is even harder as you have to pick out the appropriate interval for
>>> collecting vehicle timeLoss and filter out the background timeLoss that
>>> would have occured anyway.
>>>
>>> However, the problem description is not quite clear: Are the problematic
>>> vehicles themselves jammed by downstream traffic or is the traffic light
>>> red for them? The latter case would point to a unusual TLS setup.
>>>
>>> regards,
>>> Jakob
>>>
>>> 2018-08-22 16:15 GMT+02:00 Jose Monreal Bailey :
>>>
 Hey everyone,

 I noticed that sometimes vehicles in my scenarios, use the lane for
 turn and go through, when they want to go through, jamming all the vehicles
 behind that want to turn.

 Is there a way to get measurement?
 How many times that does happen, how many vehicles impacted by that,
 time, among others?

 Best regards,
 José

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


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

Re: [sumo-user] Metrics

[Jakob Erdmann]

If the vehicles are not allowed to go straight from the rightmost lane,
then you can just remove the connection to avoid this behavior.

2018-08-23 10:41 GMT+02:00 Jose Monreal Bailey :

> Hi Jakob,
>
> I have a 3 lane edge which 1 of the them is used to turn right (with left
> side traffic).
> This means I have a TL setup for that phase.
> Sometimes I have vehicles using that lane, which are not going to turn,
> they use it as they are going to go through to the next edge in front of
> them.
> So for this cases, I would like to know how many vehicles are impacted by
> this reckless drivers.
> I think I would have to check everytime at that point if there is a
> vehicle blocking and start counting how many were impacted and for how long.
>
> Best regards,
> José
>
> On Thu, 23 Aug 2018 at 09:34, Jakob Erdmann  wrote:
>
>> Hello,
>> there is no dedicated tool for this in the toolbox. You could parse lanes
>> from from the netstate dump or fcd output and check for specific sequences
>> of inappropriate lane use. Computing the inpact of this behavior is even
>> harder as you have to pick out the appropriate interval for collecting
>> vehicle timeLoss and filter out the background timeLoss that would have
>> occured anyway.
>>
>> However, the problem description is not quite clear: Are the problematic
>> vehicles themselves jammed by downstream traffic or is the traffic light
>> red for them? The latter case would point to a unusual TLS setup.
>>
>> regards,
>> Jakob
>>
>> 2018-08-22 16:15 GMT+02:00 Jose Monreal Bailey :
>>
>>> Hey everyone,
>>>
>>> I noticed that sometimes vehicles in my scenarios, use the lane for turn
>>> and go through, when they want to go through, jamming all the vehicles
>>> behind that want to turn.
>>>
>>> Is there a way to get measurement?
>>> How many times that does happen, how many vehicles impacted by that,
>>> time, among others?
>>>
>>> Best regards,
>>> José
>>>
>>> ___
>>> sumo-user mailing list
>>> sumo-user@eclipse.org
>>> To change your delivery options, retrieve your password, or unsubscribe
>>> from this list, visit
>>> https://dev.eclipse.org/mailman/listinfo/sumo-user
>>>
>>>
>> ___
>> sumo-user mailing list
>> sumo-user@eclipse.org
>> To change your delivery options, retrieve your password, or unsubscribe
>> from this list, visit
>> https://dev.eclipse.org/mailman/listinfo/sumo-user
>>
>
> ___
> sumo-user mailing list
> sumo-user@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/sumo-user
>
>
___
sumo-user mailing list
sumo-user@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/sumo-user

Re: [sumo-user] Metrics

[Jose Monreal Bailey]

Hi Jakob,

I have a 3 lane edge which 1 of the them is used to turn right (with left
side traffic).
This means I have a TL setup for that phase.
Sometimes I have vehicles using that lane, which are not going to turn,
they use it as they are going to go through to the next edge in front of
them.
So for this cases, I would like to know how many vehicles are impacted by
this reckless drivers.
I think I would have to check everytime at that point if there is a vehicle
blocking and start counting how many were impacted and for how long.

Best regards,
José

On Thu, 23 Aug 2018 at 09:34, Jakob Erdmann  wrote:

> Hello,
> there is no dedicated tool for this in the toolbox. You could parse lanes
> from from the netstate dump or fcd output and check for specific sequences
> of inappropriate lane use. Computing the inpact of this behavior is even
> harder as you have to pick out the appropriate interval for collecting
> vehicle timeLoss and filter out the background timeLoss that would have
> occured anyway.
>
> However, the problem description is not quite clear: Are the problematic
> vehicles themselves jammed by downstream traffic or is the traffic light
> red for them? The latter case would point to a unusual TLS setup.
>
> regards,
> Jakob
>
> 2018-08-22 16:15 GMT+02:00 Jose Monreal Bailey :
>
>> Hey everyone,
>>
>> I noticed that sometimes vehicles in my scenarios, use the lane for turn
>> and go through, when they want to go through, jamming all the vehicles
>> behind that want to turn.
>>
>> Is there a way to get measurement?
>> How many times that does happen, how many vehicles impacted by that,
>> time, among others?
>>
>> Best regards,
>> José
>>
>> ___
>> sumo-user mailing list
>> sumo-user@eclipse.org
>> To change your delivery options, retrieve your password, or unsubscribe
>> from this list, visit
>> https://dev.eclipse.org/mailman/listinfo/sumo-user
>>
>>
> ___
> sumo-user mailing list
> sumo-user@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/sumo-user
>
___
sumo-user mailing list
sumo-user@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/sumo-user

Re: [sumo-user] Pedestrian signal control

[sung hyun Jang]

Dear Jakob

Thank you for your reply!  I checked the development version that you
mentioned. I will try in the network.

Best regards,
 Sunghyun Jang

On Tue, 21 Aug 2018 at 08:47, Jakob Erdmann  wrote:

> Hello,
> It cannot be done in version 0.32.
> However,in the latest development version crossings support the attributes
> linkIndex and linkIndex2 which can be set seperately to control the forward
> and backward direction.
> regards,
> Jakob
>
> 2018-08-21 7:58 GMT+02:00 sung hyun Jang :
>
>> Dear Sumo User
>>
>> I have a question about pedestrian crossing and signal. First please look
>> into the attached picture.  As the attached picture, I created pedestrian
>> islands. I marked pedestrian signals (both ways) in the same pedestrian
>> crossing with numbers. Actually, I want to control the marked pedestrian
>> signals separately as signals are controlled in real pedestrian crossing.
>>
>> I know in SUMO default, pedestrian signals (each signal head) works
>> together in the same pedestrian crossing. So I wonder if  there is a way or
>> not.
>>
>> I am looking forward to your reply. Thank you for your time in advance.
>>
>> Best regards,
>>  Sunghyun Jang
>>
>>
>>
>> ___
>> sumo-user mailing list
>> sumo-user@eclipse.org
>> To change your delivery options, retrieve your password, or unsubscribe
>> from this list, visit
>> https://dev.eclipse.org/mailman/listinfo/sumo-user
>>
>>
> ___
> sumo-user mailing list
> sumo-user@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/sumo-user
>
___
sumo-user mailing list
sumo-user@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/sumo-user

Re: [sumo-user] Metrics

[Jakob Erdmann]

Hello,
there is no dedicated tool for this in the toolbox. You could parse lanes
from from the netstate dump or fcd output and check for specific sequences
of inappropriate lane use. Computing the inpact of this behavior is even
harder as you have to pick out the appropriate interval for collecting
vehicle timeLoss and filter out the background timeLoss that would have
occured anyway.

However, the problem description is not quite clear: Are the problematic
vehicles themselves jammed by downstream traffic or is the traffic light
red for them? The latter case would point to a unusual TLS setup.

regards,
Jakob

2018-08-22 16:15 GMT+02:00 Jose Monreal Bailey :

> Hey everyone,
>
> I noticed that sometimes vehicles in my scenarios, use the lane for turn
> and go through, when they want to go through, jamming all the vehicles
> behind that want to turn.
>
> Is there a way to get measurement?
> How many times that does happen, how many vehicles impacted by that, time,
> among others?
>
> Best regards,
> José
>
> ___
> sumo-user mailing list
> sumo-user@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/sumo-user
>
>
___
sumo-user mailing list
sumo-user@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/sumo-user

Re: [sumo-user] Question netconvert witn opendrive as Input

[Jakob Erdmann]

Hello,
the network contains a spiral definition with length=0. This wasn't
properly handled until now.
You can compile the latest repository version or download updated binaries
tomorrow at
http://sumo.dlr.de/wiki/Downloads#SUMO_-_Latest_Development_Version
regards,
Jakob

Am Do., 23. Aug. 2018 um 07:13 Uhr schrieb :

> Hello Jakob, hello all
>
>
>
> I’m trying to convert an Opendrive File to a Sumo Files using
> “netconvert”. I just get some warning but there is no error during the
> conversion. But unfortunately I don’t get any Output File. And there is
> also no reason while the Conversion is not successful. I don’t know if you
> can help me to know what’s wrong in the opendrive File.
>
>
>
> Thank you for your help and I hope to read you very soon. On the attached
> document you can find the opendrive file and the netconvert configuration
>
>
>
> Br
>
>
>
> Christian
>
>
> ___
> sumo-user mailing list
> sumo-user@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/sumo-user
>
___
sumo-user mailing list
sumo-user@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/sumo-user

Re: [sumo-user] Change Lane Information Retrieval with C++ API

[Jakob Kaths]

Great, thanks a lot. We’re looking forward to the release to adapt all the new 
functions!
 
Von: Jakob Erdmann  
Gesendet: Donnerstag, 23. August 2018 08:37
An: Jakob Kaths ; Sumo project User discussions 

Betreff: Re: [sumo-user] Change Lane Information Retrieval with C++ API
 
No Problem/ Release has been pushed back 1 week and is now scheduled for Sept 
4th: https://github.com/eclipse/sumo/milestones
 
regards,
Jakob
 
 
2018-08-23 7:48 GMT+02:00 Jakob Kaths :
Wow, thanks a lot. Yet another very fast solution from your side. We’ll try to 
get our intended idea working with that. Thanks again! I now, it could be one 
of the questions less well received, but do you have a plan for the next 
release date of SUMO? 
 
Best regards
 
Jakob
 
Von: Jakob Erdmann  
Gesendet: Mittwoch, 22. August 2018 14:54
An: Jakob Kaths ; Sumo project User discussions 

Betreff: Re: [sumo-user] Change Lane Information Retrieval with C++ API
 
Hello,
you can find the C++ TraCI client implementation of vehicle.getLaneChangeState 
in the development version (f4367fcf1a7e0231a1e30082001cafe3e9aaf492)
regards,
Jakob
 
2018-08-22 9:33 GMT+02:00 Jakob Kaths :
Hey everybody,
 
we’re trying to implement a change lane information retrieval with the C++ API. 
The goal is to identify, if a certain vehicle is currently able to change 
lanes. In Python the method couldChangeLane is already implemented, for C++ we 
thought, the compound bitset that is described on 
http://sumo.dlr.de/wiki/TraCI/Vehicle_Value_Retrieval can be retrieved. 
However, we cannot find the corresponding command in the C++ API. We’re using 
0.32 release with Traci V17, but also couldn’t find it in V18. Could you tell 
us if and how we can retrieve these values? Many thanks in advance.
 
Jakob

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

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

Re: [sumo-user] Change Lane Information Retrieval with C++ API

[Jakob Erdmann]

No Problem/ Release has been pushed back 1 week and is now scheduled for
Sept 4th: https://github.com/eclipse/sumo/milestones

regards,
Jakob


2018-08-23 7:48 GMT+02:00 Jakob Kaths :

> Wow, thanks a lot. Yet another very fast solution from your side. We’ll
> try to get our intended idea working with that. Thanks again! I now, it
> could be one of the questions less well received, but do you have a plan
> for the next release date of SUMO?
>
>
>
> Best regards
>
>
>
> Jakob
>
>
>
> *Von:* Jakob Erdmann 
> *Gesendet:* Mittwoch, 22. August 2018 14:54
> *An:* Jakob Kaths ; Sumo project User discussions <
> sumo-user@eclipse.org>
> *Betreff:* Re: [sumo-user] Change Lane Information Retrieval with C++ API
>
>
>
> Hello,
>
> you can find the C++ TraCI client implementation of
> vehicle.getLaneChangeState in the development version (
> f4367fcf1a7e0231a1e30082001cafe3e9aaf492)
>
> regards,
>
> Jakob
>
>
>
> 2018-08-22 9:33 GMT+02:00 Jakob Kaths :
>
> Hey everybody,
>
>
>
> we’re trying to implement a change lane information retrieval with the C++
> API. The goal is to identify, if a certain vehicle is currently able to
> change lanes. In Python the method couldChangeLane is already implemented,
> for C++ we thought, the compound bitset that is described on
> http://sumo.dlr.de/wiki/TraCI/Vehicle_Value_Retrieval can be retrieved.
> However, we cannot find the corresponding command in the C++ API. We’re
> using 0.32 release with Traci V17, but also couldn’t find it in V18. Could
> you tell us if and how we can retrieve these values? Many thanks in advance.
>
>
>
> Jakob
>
>
> ___
> sumo-user mailing list
> sumo-user@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/sumo-user
>
>
>
> ___
> sumo-user mailing list
> sumo-user@eclipse.org
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/sumo-user
>
>
___
sumo-user mailing list
sumo-user@eclipse.org
To change your delivery options, retrieve your password, or unsubscribe from 
this list, visit
https://dev.eclipse.org/mailman/listinfo/sumo-user