malhar's profile - activity

Hi, thanks for confirming.

Hi,

Currently Gazebo displays ( in the left panel as shown below ) the orientation ( Roll , Pitch and Yaw ) of a model in radians. Please see image if required.

I have a few related questions -

• Is it possibe to convert the Roll pitch and yaw to degrees (Mainly for convenience) ?
• Is it possible to change the representation to a quaternion ( q_x, q_y, q_z, q_w ) ?

Thanks!

Hi, thanks for confirming.

Hi, While @smorad's answer helped , the mentioned topic (/gazebo/model_states) did not provide timing values. Also it p

Hi,

I am currently working on a drone related project and using gazebo for simulation

In order to measure accuracy of my localisation, I need to know the ground truth values. Is there some way in Gazebo to continuously track/store the actual/true/ground truth 6D pose, ie the rotation and translation w.r.t. to a fixed frame like the world frame.

I wish to create something like the following -

Thanks a lot !

Hi, While @smorad's answer helped , the mentioned topic (/gazebo/model_states) did not provide timing values. Also it p

Hi yes I am using Gazebo + ROS , but I needed a topic that provides both - the ground truth and timing values.

Should I try using the /ground_truth/state topic instead ?

Hi, I tried using it and it's almost what I want but unfortunately I've run into an issue. The topic does not have any t

I found the solution by some trial and error. Update: What I did - Go to folder ~/.gazebo/models Make a copy of the

I found the solution by some trial and error. Update: What I did - Go to folder ~/.gazebo/models Make a copy of the

I found the solution by some trial and error. Update: What I did - Go to folder ~/.gazebo/models Make a copy of the

I found the solution by some trial and error. Update: What I did - Go to folder ~/.gazebo/models Make a copy of the

Hi,

Update:

I have also tried rosrun gazebo --verbose and I get a warning - [Wrn] [msgs.cc:1655] Conversion of sensor type[depth] not suppported. I believe this may be due to <sensor name="camera" type="depth"> in XML below.

Question: I have been following this tutorial to simulate a Kinect in Gazebo and then publish image,pointcloud etc to ROS topics. However, rostopic list does not show the required topics in the output.

I have also looked at the following questions on this forum - this and this. The issue in the first one seems to be same as mine. The person suggested to use URDF files instead (which I will try but would still like to know why below approach is not working ).

Here is the relevant XML code for the model.sdf file in the kinect directory that I have modified according to the tutorial mentioned above -

  //some xml 

  <sensor name="camera" type="depth">
                    <alwaysOn>true</alwaysOn>
            <update_rate>20</update_rate>
            <camera>
              <horizontal_fov>1.047198</horizontal_fov>
              <image>
                <width>640</width>
                <height>480</height>
                <format>R8G8B8</format>
              </image>
              <clip>
                <near>0.05</near>
                <far>3</far>
              </clip>
            </camera>
                <plugin name="camera_plugin" filename="libgazebo_ros_openni_kinect.so">
              <baseline>0.2</baseline>
              <alwaysOn>true</alwaysOn>
              <!-- Keep this zero, update_rate in the parent <sensor> tag
                will control the frame rate. -->
              <updateRate>0.0</updateRate>
              <cameraName>camera_ir</cameraName>
              <imageTopicName>/camera/depth/image_raw</imageTopicName>
              <cameraInfoTopicName>/camera/depth/camera_info</cameraInfoTopicName>
              <depthImageTopicName>/camera/depth/image_raw</depthImageTopicName>
              <depthImageInfoTopicName>/camera/depth/camera_info</depthImageInfoTopicName>
              <pointCloudTopicName>/camera/depth/points</pointCloudTopicName>
              <frameName>camera_link</frameName>
              <pointCloudCutoff>0.05</pointCloudCutoff>
              <distortionK1>0</distortionK1>
              <distortionK2>0</distortionK2>
              <distortionK3>0</distortionK3>
              <distortionT1>0</distortionT1>
              <distortionT2>0</distortionT2>
              <CxPrime>0</CxPrime>
              <Cx>0</Cx>
              <Cy>0</Cy>
              <focalLength>0</focalLength>
              <hackBaseline>0</hackBaseline>
            </plugin>
    </sensor>

   // some more xml

I would like to mention the following -

1. Note that in the XML code - The plugin is defined as a sensor plugin (I am aware that model and sensor plugins are different things in Gazebo and it can lead to an error if the right one is not used.)

2. I have been using the command - roslaunch gazebo_ros empty_world.launch to open the world.I then manually add the kinect model shown above.

3. I have also tried rosrun gazebo --verbose and I get a warning - [Wrn] [msgs.cc:1655] Conversion of sensor type[depth] not suppported.

4. Also, output of rostopic list after launching gazebo and adding the kinect ( Sorry unable to format this as code for some reason ) -

   /clock /gazebo/link_states /gazebo/model_states /gazebo/parameter_descriptions /gazebo/parameter_updates /gazebo/set_link_state /gazebo/set_model_state /rosout /rosout_agg

5. I have checked rostopic echo /clock , which means the Gazebo simulation is running. o I need to install that plugin libgazebo_ros_openni_kinect.so ? This seems more like an issue on the Gazebo side and hence I have posted here ( instead of ROS forum. )

Thanks a lot !