[ros-users] [planning_environment] getting all the contact locations at the current arm configuration

[Advait Jain]

Gil,

Thanks for the information.

I will give this a method shot and also file a ticket in case you were
thinking of adding this functionality in a cleaner way.

Advait



On Fri, Jan 21, 2011 at 12:31 PM, Gil Jones <gjones at willowgarage.com> wrote:
> Advait,
>
> It's true that the service returns a vector of contact information, and you
> can ask it for more contacts by changing the parameter
> "contacts_to_compute_for_display" in the environment server.  Unfortunately,
> this still doesn't do what you want it to - it will return to you an
> arbitrary set of contacts on the first link to be found in collision, not a
> contact or set of contacts for each link found to be in collision.  The only
> way to get something like the desired functionality in the current system is
> the following: repeatedly make calls to getStateValidity.  Whenever a link
> is found to be in collision, add an OrderedCollisionOperation to your
> request that disables collisions between that link and "all", and continue
> until no collisions are reported.  This should give you the full set of
> links in collision in a given state.
>
> -Gil
>
> --
> E. Gil Jones (gjones at willowgarage.com)
> Research Engineer
> Willow Garage, Inc.
> 68 Willow Road
> Menlo Park, CA 94025
> 650.475.9772
>
>
> On Fri, Jan 21, 2011 at 5:39 AM, Advait Jain <advait at cc.gatech.edu> wrote:
>>
>> The functionality that I need is already implemented!
>>
>> The response of the GetStateValidity service returns the contact state
>> as a vector of planning_environment_msgs/ContactInformation
>>
>> Apologies for missing this earlier.
>>
>>
>>
>> On Thu, Jan 20, 2011 at 9:23 PM, Advait Jain <advait at cc.gatech.edu> wrote:
>> > Hi Gil,
>> >
>> > I think I should be able to modify
>> > planning_environment/src/monitors/planning_monitor.cpp to do what I
>> > want.
>> >
>> > I will attempt this in the next couple of days and then file a feature
>> > request ticket.
>> >
>> > Advait
>> >
>> >
>> > On Thu, Jan 20, 2011 at 9:07 PM, Gil Jones <gjones at willowgarage.com>
>> > wrote:
>> >> Advait,
>> >>
>> >> Given the current APIs it's not possible.  I was considering modifying
>> >> the
>> >> collision space API to produce this information, and I'll probably get
>> >> around to it reasonably soon but it's not a high priority.  Note that
>> >> you
>> >> shouldn't trust the current markers to give you all the links that are
>> >> in
>> >> collision - it will really only give you the first link it found that's
>> >> in
>> >> collision.  If you a file a feature request ticket I'll modify the
>> >> ticket
>> >> when it's ready to go.  And we would of course welcome a patch if you
>> >> decide
>> >> to do it yourself.
>> >>
>> >> --Gil
>> >>
>> >> --
>> >> E. Gil Jones (gjones at willowgarage.com)
>> >> Research Engineer
>> >> Willow Garage, Inc.
>> >> 68 Willow Road
>> >> Menlo Park, CA 94025
>> >> 650.475.9772
>> >>
>> >>
>> >> On Thu, Jan 20, 2011 at 5:37 PM, Advait Jain <advait at cc.gatech.edu>
>> >> wrote:
>> >>>
>> >>> I am interested in getting the locations of all the contacts between
>> >>> the arm (in its current configuration) and the environment.
>> >>>
>> >>> The environment_server node of the planning_environment package almost
>> >>> does the job but it publishes each contact point as a separate Marker.
>> >>>
>> >>> It would be really helpful for me if there were a way to get the
>> >>> entire collision state in one shot. Is this a possibility?
>> >>>
>> >>> Thanks again,
>> >>> Advait
>> >>> _______________________________________________
>> >>> ros-users mailing list
>> >>> ros-users at code.ros.org
>> >>> https://code.ros.org/mailman/listinfo/ros-users
>> >>
>> >>
>> >> _______________________________________________
>> >> ros-users mailing list
>> >> ros-users at code.ros.org
>> >> https://code.ros.org/mailman/listinfo/ros-users
>> >>
>> >>
>> >
>> _______________________________________________
>> ros-users mailing list
>> ros-users at code.ros.org
>> https://code.ros.org/mailman/listinfo/ros-users
>
>
> _______________________________________________
> ros-users mailing list
> ros-users at code.ros.org
> https://code.ros.org/mailman/listinfo/ros-users
>
>