Hi Andrew,
gazebo_ros_f3d returns applied forces and I just checked with
rosservice call gazebo/apply_body_wrench
and rxplotted the topic, and things look as expected for me.  Can you
describe more detail what your setup is?
thanks,
John

On Wed, Jan 12, 2011 at 4:32 PM, Andy Somerville <
andy.somerville@resquared.com> wrote:

> Hi,
>
> Has anyone had previous success with GazeboRosF3D in cturtle? In
> attempting to use the GazeboRosF3D I've hit a bit of a wall. I believe
> that I'm using it correctly and the topic is even being published to,
> but I get all zeros back in situations that seem to clearly imply
> there should be some forces and torques.
>
> My urdf entry looks like:
>
> <gazebo>
>    <controller:gazebo_ros_f3d name="${side}_FTsensor"
> plugin="libgazebo_ros_f3d.so">
>        <alwaysOn>true</alwaysOn>
>        <updateRate>100.0</updateRate>
>        <bodyName>DarmSim/${side}Arm/WAM/LowerWristFTPalmLink</bodyName>
>        <topicName>DarmSim/${side}Arm/WAM/Wrist/FTSensor</topicName>
>        <frameName>DarmSim/${side}Arm/WAM/LowerWristFTPalmLink</frameName>
>        <interface:position name="${side}_FTsensor_force_iface" />
>    </controller:gazebo_ros_f3d>
> </gazebo>
>
>
> Where side is being filled in by "Left" and "Right", and
> "DarmSim/${side}Arm/WAM/LowerWristFTPalmLink" is the name of a link
> defined above the f3d entry in the xacro.
>
> Running the simulation produces the expected topics:
>
> /DarmSim/LeftArm/WAM/Wrist/FTSensor
> /DarmSim/RightArm/WAM/Wrist/FTSensor
>
> But echoing them while performing an action that should clearly
> produce a force and torque on that link consistently only produce 0s.
>
> I can find virtually no references to GazeboRosF3D being used in
> google searches or in the stacks I have checked out with the one
> exception being in
> "pr2_common/pr2_description/urdf/gripper_v0/gripper.gazebo.xacro" but
> that doesn't seem to be being subsequently used by anything. My next
> step is to dive into the code behind GazeboRosF3D.
>
> Is anyone aware of some obvious factors that could explain the
> behavior I'm seeing?
>
>     Andrew
> _______________________________________________
> ros-users mailing list
> ros-users@code.ros.org
> https://code.ros.org/mailman/listinfo/ros-users
>