Francisco's profile - activity

Yes, those inertia values are huge. To give you a sense of magnitude, these are the inertia values of a cylinder with a

You can use the command line. Here a MWE: # Download the URDF and convert it to SDF $ wget https://gist.githubuserconte

You can use the command line. Here a MWE: # Download the URDF and convert it to SDF $ wget https://gist.githubuserconte

Have you tested gazebo on your 1050 Ti card?

You may as well install the compiled one with:

$ sudo apt-get install ros-hydro-hector-gazebo-plugins

What are you using for your robot model? .urdf or .sdf? Assuming .urdf here you can find an usage example.

Hope this helps

Hi @arocchi, did you solve it?

Hi @nkoenig, It doesn't work. I think is a problem with the sensor tag within a joint because if I put only provideFeedback It's added to the .sdf. I'm working with gazebo 1.9.5

Here you can find 2 mazes. In Testing the installation better execute this command:

roslaunch labrob_worlds crazy_maze.launch show_rviz:=0

The idea behind this implementation is to have a custom world file and a custom model.

Check the repository files and you will see the needed parts for you application.

You can include it in the URDF model of your robot using the plugin tag like this:

<gazebo>
  <plugin name="novatel_gps_sim" filename="libhector_gazebo_ros_gps.so">
    <alwaysOn>1</alwaysOn>
    <updateRate>10.0</updateRate>
    <bodyName>base_footprint</bodyName>
    <topicName>fix</topicName>
    <velocityTopicName>fix_velocity</velocityTopicName>
    <drift>5.0 5.0 5.0</drift>
    <gaussianNoise>0.1 0.1 0.1</gaussianNoise>
    <velocityDrift>0 0 0</velocityDrift>
    <velocityGaussianNoise>0.1 0.1 0.1</velocityGaussianNoise>
  </plugin>
</gazebo>

Here you can find a working example.

You should update (edit) the original question instead of posting answers

It seems that there isn't a plugin to interface gazebo::sensors::RFIDSensor with ROS.

In any case, it may be fairly easy to implement such plugin. Take as example the gazebo_ros_bumper plugin (.cpp and .h). This plugin interfaces gazebo::sensors::ContactSensor with ROS.

Just in case here you can find the list of gazebo sensors and in this wiki you can find the list of gazebo-ros plugins.

I think you are looking for a Skid Steering Drive.

Check labrob.urdf.xacro for a simple example.

Hope this helps.

Thanks Nate! That works, although it takes a while (~ 5 sec.) to attach and detach the object. I guess it's a matter of playing with the attach_steps and detach_steps and maybe to relax the attach condition

I have been trying to use the <gripper> tag in my urdf robot like this:

<gazebo>
    <gripper name="grasping">
        <grasp_check>
            <attach_steps>20</attach_steps>
            <detach_steps>40</detach_steps>
            <min_contact_count>1</min_contact_count>
        </grasp_check>
        <gripper_link>r_finger_link</gripper_link>
        <gripper_link>l_finger_link</gripper_link>
        <palm_link>gripper_base</palm_link>
    </gripper>
</gazebo>

I have checked the corresponding code and it's loaded properly. The problem is that the Gripper::OnContact function is never called, therefore the fixed joint between the gripper and the object within the gripper will never be created.

The contacts do exist as can be seen in the figure. Additionally when I use a ContactSensor like the libgazebo_ros_bumper.so plugin I also detect such contacts.

Has somebody run with the same problem?

I'm working with gazebo-1.9.1, ros-groovy and gazebo_ros_pkg

Maybe late for some people but the solution We used for the VRC Qualifications can be found in this repository: https://bitbucket.org/fsuarez6/atlas-arm-manipulation

arm_kinematics_tools (wiki) is the package used for solving the kinematics.

I have tried the arm_kinematics package but after a while it wasn't enough for my application.

However I can recommend you the arm_navigation stack and specifically the Planning Description Configuration Wizard.

For the atlas_sandia_hands it is:

$ roslaunch planning_environment planning_description_configuration_wizard.launch urdf_package:=atlas_sandia_hands urdf_path:=atlas_sandia_hands.urdf

I think this tutorial will be a good starting point for you.

After that, you may want to continue with the rest of the arm_navigation tutorials.

You may think that this alternative is complex if you only want the inverse kinematics but I think it is worth the shot.

Hope this help