You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@sis.apache.org by Martin Desruisseaux <ma...@geomatys.fr> on 2014/02/09 17:28:39 UTC
Question about GSoC 2013: use case for OGC's moving features?
Hello Nadeem Anjum (hopping he is still on the list)
The OGC's Moving Features standard working group is looking for use
cases. I saw some proposals about transportations in a Japanese city,
hurricane trajectory, or a soccer match composed of 23 moving features.
If my memory serve me right, the Google Summer of Code 2013 project was
about a multi-agent model applied to a security or law enforcement
problem, is that right? If so, I guess that those agents could be
considered as moving features. Should we propose this GSoC project as a
use case for the OGC working group?
Martin
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Martin Desruisseaux <ma...@geomatys.fr>.
Le 13/02/14 15:06, Martin Desruisseaux a écrit :
> Note: the goal is not to describe a particular work, but to describe
> the general scenario (...snip...)
I should add: I think that it is fine to reference your work in the
bibliography. It is just that OGC being a standardization organisation,
the main document needs to be implementation neutral. But I think that
it is perfectly fine to mention some implementations in the bibliography.
Martin
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Martin Desruisseaux <ma...@geomatys.fr>.
Hello Nadeem
Le 13/02/14 14:23, Nadeem Anjum a écrit :
> A Google doc sounds good. I have started one, which can be accessed at:
> https://docs.google.com/document/d/1CEOXvb6jFZKBSTM160-4lXFAoFgjROpUC46VlbmGi9g/edit
Cool, thanks! Could you write some lines about your work? Other uses
cases are about one page long, sometime with figures.
Note: the goal is not to describe a particular work, but to describe the
general scenario where some kind of data or information are likely to be
useful to policemen (for example). The emphasis is on which objects or
features are moving in this scenario, what are the characteristics of
their displacements, which data are available and what are the
characteristic of those data (sampling rate, accuracy, etc.).
Looking at the other use cases, the pattern seems to be:
* Description of the problem that we try to solve.
* Objects that could be represented as a moving features. Examples:
o Pedestrians. Speed: 4km/h. State: stay or walking.
o Vehicles. Speed: 60 km/h.
o Vessels. Details: enter/leave a particular area. Speed: 11 knots.
o Vessels. Details: suspicious manoeuvres. Speed: rapidly varying.
A description of data sources. Example:
* Smart phones. Localization method: GPS/Wi-Fi positioning. Sampling
rate: 1 point per second. Accuracy: 100 metres.
* Automatic Identification System (AIS). Sampling rate: 1 point every
6 minutes. Accuracy: 100 metres.
An explanation of constraints that a standard would need to meet in
order to be useful to your scenario. Example:
* Aviation: because of the large amount of data, an efficient binary
format is necessary. Verbose XML can not work.
Thanks!
Martin
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Martin Desruisseaux <ma...@geomatys.fr>.
Hello Nadeem
Its look good. I have transferred the information to the Moving Features
group.
Thanks for all your work,
Martin
Le 20/02/14 19:29, Nadeem Anjum a écrit :
> Hi Martin,
>
> The snapshot can be found here: http://cse.iitkgp.ac.in/~nanjum/500iter.jpg
>
> The meaning of color codes can be found here:
> http://cse.iitkgp.ac.in/~nanjum/color_codes.jpg
>
> Thanks,
> Nadeem.
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Nadeem Anjum <na...@gmail.com>.
Hi Martin,
The snapshot can be found here: http://cse.iitkgp.ac.in/~nanjum/500iter.jpg
The meaning of color codes can be found here:
http://cse.iitkgp.ac.in/~nanjum/color_codes.jpg
Thanks,
Nadeem.
On Thu, Feb 20, 2014 at 7:14 PM, Martin Desruisseaux <
martin.desruisseaux@geomatys.fr> wrote:
> Hello Nadeem
>
> Thanks for the correction and the screenshot. But the screenshot has
> apparently been filtered out by the mailing list. Could you put it in the
> google document, or elsewhere on the web with a link to it?
>
> Martin
>
>
> Le 20/02/14 14:01, Nadeem Anjum a écrit :
>
> Hi Martin,
>>
>> It looks good.I have done some minor modification as below. I have also
>> attached a screenshot of the simulation:
>> ____________________________________________________________________
>> Relationship with moving features:
>>
>> * The simulation needs a city description with roads and buildings of
>> interest. LeafletJS or CityGML could be used here.
>> * The simulation needs to represent the agent's trajectory in the
>> above-cited city. Moving features could be used here.
>> * The trajectory is repeated weekly with random variations. We may
>> represent that by a "base line" moving feature. This base line would
>> not describe a real moving feature, but would be used as a template.
>> * Interactions between the environment (CityGML) and the moving
>> features may trig events. For example when <mf:Trajectory> value
>> become close to a CityGML feature or interest having a low "presence of
>> police" attribute, a crime may occur depending on the value of that
>> attribute and other attributes like
>> "absence of guardian", "profit", "risk".
>>
>
>
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Martin Desruisseaux <ma...@geomatys.fr>.
Hello Nadeem
Thanks for the correction and the screenshot. But the screenshot has
apparently been filtered out by the mailing list. Could you put it in
the google document, or elsewhere on the web with a link to it?
Martin
Le 20/02/14 14:01, Nadeem Anjum a écrit :
> Hi Martin,
>
> It looks good.I have done some minor modification as below. I have
> also attached a screenshot of the simulation:
> ____________________________________________________________________
> Relationship with moving features:
>
> * The simulation needs a city description with roads and buildings of
> interest. LeafletJS or CityGML could be used here.
> * The simulation needs to represent the agent's trajectory in the
> above-cited city. Moving features could be used here.
> * The trajectory is repeated weekly with random variations. We may
> represent that by a "base line" moving feature. This base line would
> not describe a real moving feature, but would be used as a template.
> * Interactions between the environment (CityGML) and the moving
> features may trig events. For example when <mf:Trajectory> value
> become close to a CityGML feature or interest having a low "presence of
> police" attribute, a crime may occur depending on the value of that
> attribute and other attributes like
> "absence of guardian", "profit", "risk".
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Nadeem Anjum <na...@gmail.com>.
Hi Martin,
It looks good.I have done some minor modification as below. I have also
attached a screenshot of the simulation:
____________________________________________________________________
Relationship with moving features:
* The simulation needs a city description with roads and buildings of
interest. LeafletJS or CityGML could be used here.
* The simulation needs to represent the agent's trajectory in the
above-cited city. Moving features could be used here.
* The trajectory is repeated weekly with random variations. We may
represent that by a "base line" moving feature. This base line would
not describe a real moving feature, but would be used as a template.
* Interactions between the environment (CityGML) and the moving
features may trig events. For example when <mf:Trajectory> value
become close to a CityGML feature or interest having a low "presence of
police" attribute, a crime may occur depending on the value of that
attribute and other attributes like
"absence of guardian", "profit", "risk".
_____________________________________________________________________
The screenshot can be interpreted as below:
- The ciminal agent moves between Home, Office, Bar, Lunch-A, Lunch-B,
Mall marked on the map as per a weekly schedule
- The result corresponds to the weekly schedule run for 500 times on the
city: Indianapolis.
- The small squares on the map correspond to crime spots.
- The color of each crime spot represents the number of times crime has
been committed at that spot, normalized on a scale of 0-100.
[image: Inline image 1] : 0
[image: Inline image 2] : 1-10
[image: Inline image 4] : 11-20
[image: Inline image 12] : 21-30
[image: Inline image 10] : 31-40
[image: Inline image 11] : 41-50
[image: Inline image 9] : 51-60
[image: Inline image 8] : 61-70
[image: Inline image 7] : 71-80
[image: Inline image 6] : 81-90
[image: Inline image 5] : 91-100
____________________________________________________________
Thanks,
Nadeem
>
> Relationship with moving features:
>
> * The simulation needs a city description with roads and buildings of
> interest. CityGML could be used here.
> * The simulation needs to represent the agent's trajectory in the
> above-cited city. Moving features could be used here.
> * The trajectory is repeated weekly with random variations. We may
> represent that by a "base line" moving feature. This base line would
> not describe a real moving feature, but would be used as a template.
> * Interactions between the environment (CityGML) and the moving
> features may trig events. For example when <mf:Trajectory> value
> become close to a CityGML feature or interest having a "presence of
> guardian" boolean attribute, different events may occur depending on
> the value of that boolean.
>
>
>
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Martin Desruisseaux <ma...@geomatys.fr>.
Hello Nadeem
The Moving Feature group asked us to provide a little bit more details
about how the use case relates to moving features. What about the
following? (please edit as you see fit)
Relationship with moving features:
* The simulation needs a city description with roads and buildings of
interest. CityGML could be used here.
* The simulation needs to represent the agent's trajectory in the
above-cited city. Moving features could be used here.
* The trajectory is repeated weekly with random variations. We may
represent that by a "base line" moving feature. This base line would
not describe a real moving feature, but would be used as a template.
* Interactions between the environment (CityGML) and the moving
features may trig events. For example when <mf:Trajectory> value
become close to a CityGML feature or interest having a "presence of
guardian" boolean attribute, different events may occur depending on
the value of that boolean.
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Martin Desruisseaux <ma...@geomatys.fr>.
Hello Nadeem
Thanks a lot, it sound very good to me! I forward the link to the moving
features group right away, with copy to you.
Thanks again!
Martin
Le 18/02/14 22:39, Nadeem Anjum a écrit :
> Hi Martin and all others,
>
> I am done with the write-up for the OGC's Moving Features standard working
> group. Please have a look at it here:
> https://docs.google.com/document/d/1CEOXvb6jFZKBSTM160-4lXFAoFgjROpUC46VlbmGi9g/edit
>
> Suggestions are welcome.
>
> Thanks,
> Nadeem
>
>
> On Thu, Feb 13, 2014 at 6:53 PM, Nadeem Anjum <na...@gmail.com>wrote:
>
>> Hi Martin,
>>
>> A Google doc sounds good. I have started one, which can be accessed at:
>> https://docs.google.com/document/d/1CEOXvb6jFZKBSTM160-4lXFAoFgjROpUC46VlbmGi9g/edit
>>
>> Thanks,
>> Nadeem
>>
>>
>>
>> On Wed, Feb 12, 2014 at 11:46 PM, Martin Desruisseaux <
>> martin.desruisseaux@geomatys.fr> wrote:
>>
>>> Le 12/02/14 17:35, Martin Desruisseaux a écrit :
>>>
>>> Le 12/02/14 16:06, Nadeem Anjum a écrit :
>>>>> It's great that they agreed on the use case. But am not really sure on
>>>>> how to edit the wiki.
>>>>>
>>>> I do not know neither. We will need to proceed in an other way. Do you
>>>> have some online material that you can edit? (on an other server, school,
>>>> university...)
>>>>
>>> Alternatively, we could start a google document. Do you have an account?
>>>
>>> Martin
>>>
>>>
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Nadeem Anjum <na...@gmail.com>.
Hi Martin and all others,
I am done with the write-up for the OGC's Moving Features standard working
group. Please have a look at it here:
https://docs.google.com/document/d/1CEOXvb6jFZKBSTM160-4lXFAoFgjROpUC46VlbmGi9g/edit
Suggestions are welcome.
Thanks,
Nadeem
On Thu, Feb 13, 2014 at 6:53 PM, Nadeem Anjum <na...@gmail.com>wrote:
> Hi Martin,
>
> A Google doc sounds good. I have started one, which can be accessed at:
> https://docs.google.com/document/d/1CEOXvb6jFZKBSTM160-4lXFAoFgjROpUC46VlbmGi9g/edit
>
> Thanks,
> Nadeem
>
>
>
> On Wed, Feb 12, 2014 at 11:46 PM, Martin Desruisseaux <
> martin.desruisseaux@geomatys.fr> wrote:
>
>> Le 12/02/14 17:35, Martin Desruisseaux a écrit :
>>
>> Le 12/02/14 16:06, Nadeem Anjum a écrit :
>>>
>>>> It's great that they agreed on the use case. But am not really sure on
>>>> how to edit the wiki.
>>>>
>>> I do not know neither. We will need to proceed in an other way. Do you
>>> have some online material that you can edit? (on an other server, school,
>>> university...)
>>>
>>
>> Alternatively, we could start a google document. Do you have an account?
>>
>> Martin
>>
>>
>
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Nadeem Anjum <na...@gmail.com>.
Hi Martin,
A Google doc sounds good. I have started one, which can be accessed at:
https://docs.google.com/document/d/1CEOXvb6jFZKBSTM160-4lXFAoFgjROpUC46VlbmGi9g/edit
Thanks,
Nadeem
On Wed, Feb 12, 2014 at 11:46 PM, Martin Desruisseaux <
martin.desruisseaux@geomatys.fr> wrote:
> Le 12/02/14 17:35, Martin Desruisseaux a écrit :
>
> Le 12/02/14 16:06, Nadeem Anjum a écrit :
>>
>>> It's great that they agreed on the use case. But am not really sure on
>>> how to edit the wiki.
>>>
>> I do not know neither. We will need to proceed in an other way. Do you
>> have some online material that you can edit? (on an other server, school,
>> university...)
>>
>
> Alternatively, we could start a google document. Do you have an account?
>
> Martin
>
>
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Martin Desruisseaux <ma...@geomatys.fr>.
Le 12/02/14 17:35, Martin Desruisseaux a écrit :
> Le 12/02/14 16:06, Nadeem Anjum a écrit :
>> It's great that they agreed on the use case. But am not really sure
>> on how to edit the wiki.
> I do not know neither. We will need to proceed in an other way. Do you
> have some online material that you can edit? (on an other server,
> school, university...)
Alternatively, we could start a google document. Do you have an account?
Martin
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Martin Desruisseaux <ma...@geomatys.fr>.
Hello Nadeem
Le 12/02/14 16:06, Nadeem Anjum a écrit :
> It's great that they agreed on the use case. But am not really sure on how
> to edit the wiki.
I do not know neither. We will need to proceed in an other way. Do you
have some online material that you can edit? (on an other server,
school, university...)
Martin
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Nadeem Anjum <na...@gmail.com>.
Hi Martin,
It's great that they agreed on the use case. But am not really sure on how
to edit the wiki.
Thanks,
Nadeem.
------------------------------------
Sent from my Nexus 4
On Feb 11, 2014 10:03 PM, "Martin Desruisseaux" <
martin.desruisseaux@geomatys.fr> wrote:
> Hello Nadeem and all
>
> The Moving Feature SWG chair just replied. He would be interested in
> Nadeem's use case. But because of tight deadline before the next OGC
> meeting, he would like the use case within a week. We could start a wiki
> page for that, but I just looked at the SIS wiki and do not yet see an
> "Edit" button. Does anyone know if there is progress on this issue?
>
> Martin
>
>
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Martin Desruisseaux <ma...@geomatys.fr>.
Hello Nadeem and all
The Moving Feature SWG chair just replied. He would be interested in
Nadeem's use case. But because of tight deadline before the next OGC
meeting, he would like the use case within a week. We could start a wiki
page for that, but I just looked at the SIS wiki and do not yet see an
"Edit" button. Does anyone know if there is progress on this issue?
Martin
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Martin Desruisseaux <ma...@geomatys.fr>.
Hello Suresh
Le 10/02/14 00:55, Suresh Marru a écrit :
> Great to see this. Thanks Martin and Nadeem. Please let me know if you need
> any elaboration of the use case and I will be happy to provide it.
Thanks. I emailed to the Moving Feature list with Nadeem in copy. I'm
worried that we may be late in the process (I should have reacted
sooner...), but we will see what the SWG said.
Martin
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Suresh Marru <sm...@apache.org>.
Great to see this. Thanks Martin and Nadeem. Please let me know if you need
any elaboration of the use case and I will be happy to provide it.
Suresh
On Sun, Feb 9, 2014 at 1:58 PM, Nadeem Anjum <na...@gmail.com>wrote:
> Sure Martin. Let's go ahead with it.
>
> ------------------------------------
> Sent from my Nexus 4
> On Feb 9, 2014 11:56 PM, "Martin Desruisseaux" <
> martin.desruisseaux@geomatys.fr> wrote:
>
> > Hello Nadeem
> >
> > Thank for you quick reply. Your work seems a nice use case. If you agree,
> > I will post an email on the Moving Feature SWG asking if they would like
> > such use case (it will probably depends on their schedule and whether
> they
> > consider having enough use cases or still want more - they have 4 in
> their
> > latest draft, but your work does not seem to duplicate the other use
> > cases). If the SWG said that they would like to add this use case, then
> we
> > could write one or two pages about your work, including screenshots if
> you
> > have some?
> >
> > Does it sound okay?
> >
> > Martin
> >
> >
> > Le 09/02/14 17:50, Nadeem Anjum a écrit :
> >
> >> Hi Martin,
> >>
> >> I am very much present here on the list.
> >>
> >> The GSoC 2013 project involved simulation of single moving criminal
> agent,
> >> to predict crime patterns. The future work of this project could involve
> >> moving victim agents add well as moving police agents.
> >>
> >> So yeah, the project involves a multi-agent moving model and can be
> >> proposed as a use case for the OGC working group.
> >>
> >> ------------------------------------
> >> Sent from my Nexus 4
> >> On Feb 9, 2014 9:59 PM, "Martin Desruisseaux" <
> >> martin.desruisseaux@geomatys.fr> wrote:
> >>
> >> Hello Nadeem Anjum (hopping he is still on the list)
> >>>
> >>> The OGC's Moving Features standard working group is looking for use
> >>> cases.
> >>> I saw some proposals about transportations in a Japanese city,
> hurricane
> >>> trajectory, or a soccer match composed of 23 moving features.
> >>>
> >>> If my memory serve me right, the Google Summer of Code 2013 project was
> >>> about a multi-agent model applied to a security or law enforcement
> >>> problem,
> >>> is that right? If so, I guess that those agents could be considered as
> >>> moving features. Should we propose this GSoC project as a use case for
> >>> the
> >>> OGC working group?
> >>>
> >>> Martin
> >>>
> >>>
> >>>
> >
>
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Nadeem Anjum <na...@gmail.com>.
Sure Martin. Let's go ahead with it.
------------------------------------
Sent from my Nexus 4
On Feb 9, 2014 11:56 PM, "Martin Desruisseaux" <
martin.desruisseaux@geomatys.fr> wrote:
> Hello Nadeem
>
> Thank for you quick reply. Your work seems a nice use case. If you agree,
> I will post an email on the Moving Feature SWG asking if they would like
> such use case (it will probably depends on their schedule and whether they
> consider having enough use cases or still want more - they have 4 in their
> latest draft, but your work does not seem to duplicate the other use
> cases). If the SWG said that they would like to add this use case, then we
> could write one or two pages about your work, including screenshots if you
> have some?
>
> Does it sound okay?
>
> Martin
>
>
> Le 09/02/14 17:50, Nadeem Anjum a écrit :
>
>> Hi Martin,
>>
>> I am very much present here on the list.
>>
>> The GSoC 2013 project involved simulation of single moving criminal agent,
>> to predict crime patterns. The future work of this project could involve
>> moving victim agents add well as moving police agents.
>>
>> So yeah, the project involves a multi-agent moving model and can be
>> proposed as a use case for the OGC working group.
>>
>> ------------------------------------
>> Sent from my Nexus 4
>> On Feb 9, 2014 9:59 PM, "Martin Desruisseaux" <
>> martin.desruisseaux@geomatys.fr> wrote:
>>
>> Hello Nadeem Anjum (hopping he is still on the list)
>>>
>>> The OGC's Moving Features standard working group is looking for use
>>> cases.
>>> I saw some proposals about transportations in a Japanese city, hurricane
>>> trajectory, or a soccer match composed of 23 moving features.
>>>
>>> If my memory serve me right, the Google Summer of Code 2013 project was
>>> about a multi-agent model applied to a security or law enforcement
>>> problem,
>>> is that right? If so, I guess that those agents could be considered as
>>> moving features. Should we propose this GSoC project as a use case for
>>> the
>>> OGC working group?
>>>
>>> Martin
>>>
>>>
>>>
>
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Martin Desruisseaux <ma...@geomatys.fr>.
Hello Nadeem
Thank for you quick reply. Your work seems a nice use case. If you
agree, I will post an email on the Moving Feature SWG asking if they
would like such use case (it will probably depends on their schedule and
whether they consider having enough use cases or still want more - they
have 4 in their latest draft, but your work does not seem to duplicate
the other use cases). If the SWG said that they would like to add this
use case, then we could write one or two pages about your work,
including screenshots if you have some?
Does it sound okay?
Martin
Le 09/02/14 17:50, Nadeem Anjum a écrit :
> Hi Martin,
>
> I am very much present here on the list.
>
> The GSoC 2013 project involved simulation of single moving criminal agent,
> to predict crime patterns. The future work of this project could involve
> moving victim agents add well as moving police agents.
>
> So yeah, the project involves a multi-agent moving model and can be
> proposed as a use case for the OGC working group.
>
> ------------------------------------
> Sent from my Nexus 4
> On Feb 9, 2014 9:59 PM, "Martin Desruisseaux" <
> martin.desruisseaux@geomatys.fr> wrote:
>
>> Hello Nadeem Anjum (hopping he is still on the list)
>>
>> The OGC's Moving Features standard working group is looking for use cases.
>> I saw some proposals about transportations in a Japanese city, hurricane
>> trajectory, or a soccer match composed of 23 moving features.
>>
>> If my memory serve me right, the Google Summer of Code 2013 project was
>> about a multi-agent model applied to a security or law enforcement problem,
>> is that right? If so, I guess that those agents could be considered as
>> moving features. Should we propose this GSoC project as a use case for the
>> OGC working group?
>>
>> Martin
>>
>>
Re: Question about GSoC 2013: use case for OGC's moving features?
Posted by Nadeem Anjum <na...@gmail.com>.
Hi Martin,
I am very much present here on the list.
The GSoC 2013 project involved simulation of single moving criminal agent,
to predict crime patterns. The future work of this project could involve
moving victim agents add well as moving police agents.
So yeah, the project involves a multi-agent moving model and can be
proposed as a use case for the OGC working group.
------------------------------------
Sent from my Nexus 4
On Feb 9, 2014 9:59 PM, "Martin Desruisseaux" <
martin.desruisseaux@geomatys.fr> wrote:
> Hello Nadeem Anjum (hopping he is still on the list)
>
> The OGC's Moving Features standard working group is looking for use cases.
> I saw some proposals about transportations in a Japanese city, hurricane
> trajectory, or a soccer match composed of 23 moving features.
>
> If my memory serve me right, the Google Summer of Code 2013 project was
> about a multi-agent model applied to a security or law enforcement problem,
> is that right? If so, I guess that those agents could be considered as
> moving features. Should we propose this GSoC project as a use case for the
> OGC working group?
>
> Martin
>
>