Ask Your Question

Revision history [back]

click to hide/show revision 1
initial version

gazebo links spawn in wrong position - rviz working fine

Hi, I have a problem with gazebo spawning the links of my URDF model not correctly. All parts clumb up in the origin of the world, except of some links which are connected with fixed joints.

Here is the link to the repository of the projekt: github

About ROS and OS:

        ROS_VERSION=1
        ROS_PYTHON_VERSION=2
        ROS_DISTRO=melodic
Gazebo 9
    Linux VERSION="18.04.3 LTS (Bionic Beaver)"

1: It looks like the links are spawning in their origins connected to the world and are not connected by the joints gazebo fail ,gazebo fail2

2: looking good in Rviz, joint_state_publisher working and all joints move as they should. RViz all good

So when I launch boroscope_gazebo.launch I get this :

[ INFO] [1579263007.980516751, 0.195000000]: Starting gazebo_ros_control plugin in namespace: /

[ INFO] [1579261386.832915905, 0.095000000]: gazebo_ros_control plugin is waiting for model URDF in parameter [robot_description] on the ROS param server.

[ERROR] [1579261752.725236087, 0.216000000]: No p gain specified for pid.  Namespace: /gazebo_ros_control/pid_gains/horizontal_x

I get the pid gain error for every actuated joint, but should not be a Problem ? I already tried it with pid gains in the controller but there was no change to report.

I searched a lot for this problem and could not find the right answer. On my search I worked on:

-gazebo_ros_control

-wrong robot namespace through the .launch files

-I calculated the inertia of the links with solid works and on a second try with meshlab

  • transmissions and plugins in the .xacro of the gazebo package

  • Joint nodes/parameters

URDF example:

  <link name="measuring_system">
    <inertial>
      <mass value="1.5" />
      <origin xyz="0.0180561 -0.0098927 -0.0546269" rpy="0.0 0.0 0.0"/>
      <inertia ixx="0.00000147864" ixy="0" ixz="0.00000028363" iyy="0.00000166953" iyz="0.00000021929" izz="0.00000065700" />
    </inertial>
    <visual>
      <geometry>
        <mesh filename="package://boroscope_support/meshes/visual/measuring_system_hull.stl"/>
      </geometry>
      <origin xyz="0.0 0.0 0.0" rpy=" 0.0 0.0 0.0"/>
      <material name="blue"/>
    </visual>
    <collision>
      <geometry>
        <mesh filename="package://boroscope_support/meshes/collision/measuring_system_hull.stl"/>
      </geometry>
      <origin xyz="0.0 0.0 0.0" rpy=" 0.0 0.0 0.0"/>
    </collision>
  </link>

  <joint name="boro_theta" type="revolute">
    <parent link="bgs80cc"/>
    <child link="measuring_system"/>
    <axis xyz="0 0 1"/>
    <origin xyz="0.1007 0.0 0.01" rpy="1.5707963268 0.0 0.0"/>
    <limit effort="10" velocity="0.1" lower="-0.6" upper="0.6"/>
    <dynamics damping="0.7" friction="0.0"/>
  </joint>

Model launched in gazebo: rostopic list:

/arm_controller/command
/arm_controller/follow_joint_trajectory/cancel
/arm_controller/follow_joint_trajectory/feedback
/arm_controller/follow_joint_trajectory/goal
/arm_controller/follow_joint_trajectory/result
/arm_controller/follow_joint_trajectory/status
/arm_controller/state
/clock
/gazebo/link_states
/gazebo/model_states
/gazebo/parameter_descriptions
/gazebo/parameter_updates
/gazebo/set_link_state
/gazebo/set_model_state
/joint_states
/rosout
/rosout_agg
/tf
/tf_static

rosservice list:

/arm_controller/query_state
/boroscope_controller_spawner/get_loggers
/boroscope_controller_spawner/set_logger_level
/controller_manager/list_controller_types
/controller_manager/list_controllers
/controller_manager/load_controller
/controller_manager/reload_controller_libraries
/controller_manager/switch_controller
/controller_manager/unload_controller
/gazebo/apply_body_wrench
/gazebo/apply_joint_effort
/gazebo/clear_body_wrenches
/gazebo/clear_joint_forces
/gazebo/delete_light
/gazebo/delete_model
/gazebo/get_joint_properties
/gazebo/get_light_properties
/gazebo/get_link_properties
/gazebo/get_link_state
/gazebo/get_loggers
/gazebo/get_model_properties
/gazebo/get_model_state
/gazebo/get_physics_properties
/gazebo/get_world_properties
/gazebo/pause_physics
/gazebo/reset_simulation
/gazebo/reset_world
/gazebo/set_joint_properties
/gazebo/set_light_properties
/gazebo/set_link_properties
/gazebo/set_link_state
/gazebo/set_logger_level
/gazebo/set_model_configuration
/gazebo/set_model_state
/gazebo/set_parameters
/gazebo/set_physics_properties
/gazebo/spawn_sdf_model
/gazebo/spawn_urdf_model
/gazebo/unpause_physics
/gazebo_gui/get_loggers
/gazebo_gui/set_logger_level
/joint_state_controller_spawner/get_loggers
/joint_state_controller_spawner/set_logger_level
/robot_state_publisher/get_loggers
/robot_state_publisher/set_logger_level
/rosout/get_loggers
/rosout/set_logger_level

I am new to programming and ROS. After a few weeks of research I dont know how to solve the problem. For more information just tell me and I will provide it.

Thanks in advance

ROI3000

gazebo links spawn in wrong position - rviz working fine

Hi, I have a problem with gazebo spawning the links of my URDF model not correctly. All parts clumb up in the origin of the world, except of some links which are connected with fixed joints.

Here is the link to the repository of the projekt: github

About ROS and OS:

        ROS_VERSION=1
        ROS_PYTHON_VERSION=2
        ROS_DISTRO=melodic
Gazebo 9
    Linux VERSION="18.04.3 LTS (Bionic Beaver)"

1: It looks like the links are spawning in their origins connected to the world and are not connected by the joints gazebo fail ,gazebo fail2

2: looking good in Rviz, joint_state_publisher working and all joints move as they should. RViz all good

So when I launch boroscope_gazebo.launch I get this :

[ INFO] [1579263007.980516751, 0.195000000]: Starting gazebo_ros_control plugin in namespace: /

[ INFO] [1579261386.832915905, 0.095000000]: gazebo_ros_control plugin is waiting for model URDF in parameter [robot_description] on the ROS param server.

[ERROR] [1579261752.725236087, 0.216000000]: No p gain specified for pid.  Namespace: /gazebo_ros_control/pid_gains/horizontal_x

I get the pid gain error for every actuated joint, but should not be a Problem ? I already tried it with pid gains in the controller but there was no change to report.

I searched a lot for this problem and could not find the right answer. On my search I worked on:

-gazebo_ros_control

-wrong robot namespace through the .launch files

-I calculated the inertia of the links with solid works and on a second try with meshlab

  • transmissions and plugins in the .xacro of the gazebo package

  • Joint nodes/parameters

URDF example:

  <link name="measuring_system">
    <inertial>
      <mass value="1.5" />
      <origin xyz="0.0180561 -0.0098927 -0.0546269" rpy="0.0 0.0 0.0"/>
      <inertia ixx="0.00000147864" ixy="0" ixz="0.00000028363" iyy="0.00000166953" iyz="0.00000021929" izz="0.00000065700" />
    </inertial>
    <visual>
      <geometry>
        <mesh filename="package://boroscope_support/meshes/visual/measuring_system_hull.stl"/>
      </geometry>
      <origin xyz="0.0 0.0 0.0" rpy=" 0.0 0.0 0.0"/>
      <material name="blue"/>
    </visual>
    <collision>
      <geometry>
        <mesh filename="package://boroscope_support/meshes/collision/measuring_system_hull.stl"/>
      </geometry>
      <origin xyz="0.0 0.0 0.0" rpy=" 0.0 0.0 0.0"/>
    </collision>
  </link>

  <joint name="boro_theta" type="revolute">
    <parent link="bgs80cc"/>
    <child link="measuring_system"/>
    <axis xyz="0 0 1"/>
    <origin xyz="0.1007 0.0 0.01" rpy="1.5707963268 0.0 0.0"/>
    <limit effort="10" velocity="0.1" lower="-0.6" upper="0.6"/>
    <dynamics damping="0.7" friction="0.0"/>
  </joint>

Model launched in gazebo: rostopic list:

/arm_controller/command
/arm_controller/follow_joint_trajectory/cancel
/arm_controller/follow_joint_trajectory/feedback
/arm_controller/follow_joint_trajectory/goal
/arm_controller/follow_joint_trajectory/result
/arm_controller/follow_joint_trajectory/status
/arm_controller/state
/clock
/gazebo/link_states
/gazebo/model_states
/gazebo/parameter_descriptions
/gazebo/parameter_updates
/gazebo/set_link_state
/gazebo/set_model_state
/joint_states
/rosout
/rosout_agg
/tf
/tf_static

rosservice list:

/arm_controller/query_state
/boroscope_controller_spawner/get_loggers
/boroscope_controller_spawner/set_logger_level
/controller_manager/list_controller_types
/controller_manager/list_controllers
/controller_manager/load_controller
/controller_manager/reload_controller_libraries
/controller_manager/switch_controller
/controller_manager/unload_controller
/gazebo/apply_body_wrench
/gazebo/apply_joint_effort
/gazebo/clear_body_wrenches
/gazebo/clear_joint_forces
/gazebo/delete_light
/gazebo/delete_model
/gazebo/get_joint_properties
/gazebo/get_light_properties
/gazebo/get_link_properties
/gazebo/get_link_state
/gazebo/get_loggers
/gazebo/get_model_properties
/gazebo/get_model_state
/gazebo/get_physics_properties
/gazebo/get_world_properties
/gazebo/pause_physics
/gazebo/reset_simulation
/gazebo/reset_world
/gazebo/set_joint_properties
/gazebo/set_light_properties
/gazebo/set_link_properties
/gazebo/set_link_state
/gazebo/set_logger_level
/gazebo/set_model_configuration
/gazebo/set_model_state
/gazebo/set_parameters
/gazebo/set_physics_properties
/gazebo/spawn_sdf_model
/gazebo/spawn_urdf_model
/gazebo/unpause_physics
/gazebo_gui/get_loggers
/gazebo_gui/set_logger_level
/joint_state_controller_spawner/get_loggers
/joint_state_controller_spawner/set_logger_level
/robot_state_publisher/get_loggers
/robot_state_publisher/set_logger_level
/rosout/get_loggers
/rosout/set_logger_level

I am new to programming and ROS. After a few weeks of research I dont know how to solve the problem. For more information just tell me and I will provide it.

Thanks in advance

ROI3000

gazebo links spawn in wrong position - rviz working fineneed tips for inertia

Hi, I have a problem with gazebo spawning the links of my URDF model not correctly. All parts clumb up in the origin correctly.

So figured out that it is still a problem with the inertia.

-I tried to fix it but could not make it stable. When adding link by link after the 5th part everything is collapsing. I removed the big table from the URDF, than there was a change but in the end it still collapses.

-I used the inertia calculated by Mashlab and I tried it with the inertia from SolidWorks, but there was no improvement. Using different masses for the links didnt change anything as well.

Question 1: should I build the complete testbench in solid works and than use the urdf_exporter? By now I build the URDF link by link...

Question 2: If I change the structure of the world, except of some links which are connected with fixed joints. URDF in Gazebo but leave the old structure in MoveIt, would there be problems with the transformations ?

Here is the link to the repository of the projekt: github

About ROS and OS:

       ROS_VERSION=1
        ROS_PYTHON_VERSION=2
        ROS_DISTRO=melodic
Gazebo 9
    Linux VERSION="18.04.3 LTS (Bionic Beaver)"
and the where the problem is URDF

I would be very grateful I somebody could help me with this issue or giving tips on how to calculate better inertia.

1: It looks like the links are spawning in their origins connected to the world and are not connected by the joints wrong inertia gazebo fail ,gazebo fail2

2: looking good in Rviz, joint_state_publisher working and all joints move as they should. Rviz, RViz all good

So when I launch boroscope_gazebo.launch I get this :

[ INFO] [1579263007.980516751, 0.195000000]: Starting gazebo_ros_control plugin in namespace: /

[ INFO] [1579261386.832915905, 0.095000000]: gazebo_ros_control plugin is waiting for model URDF in parameter [robot_description] on the ROS param server.

[ERROR] [1579261752.725236087, 0.216000000]: No p gain specified for pid.  Namespace: /gazebo_ros_control/pid_gains/horizontal_x

I get the pid gain error for every actuated joint, but should not be a Problem ? I already tried it with pid gains in the controller but there was no change to report.

I searched a lot for this problem and could not find the right answer. On my search I worked on:

-gazebo_ros_control

-wrong robot namespace through the .launch files

-I calculated the inertia of the links with solid works and on a second try with meshlab

  • transmissions and plugins in the .xacro of the gazebo package

  • Joint nodes/parameters

URDF example:

  <link name="measuring_system">
    <inertial>
      <mass value="1.5" />
      <origin xyz="0.0180561 -0.0098927 -0.0546269" rpy="0.0 0.0 0.0"/>
      <inertia ixx="0.00000147864" ixy="0" ixz="0.00000028363" iyy="0.00000166953" iyz="0.00000021929" izz="0.00000065700" />
    </inertial>
    <visual>
      <geometry>
        <mesh filename="package://boroscope_support/meshes/visual/measuring_system_hull.stl"/>
      </geometry>
      <origin xyz="0.0 0.0 0.0" rpy=" 0.0 0.0 0.0"/>
      <material name="blue"/>
    </visual>
    <collision>
      <geometry>
        <mesh filename="package://boroscope_support/meshes/collision/measuring_system_hull.stl"/>
      </geometry>
      <origin xyz="0.0 0.0 0.0" rpy=" 0.0 0.0 0.0"/>
    </collision>
  </link>

  <joint name="boro_theta" type="revolute">
    <parent link="bgs80cc"/>
    <child link="measuring_system"/>
    <axis xyz="0 0 1"/>
    <origin xyz="0.1007 0.0 0.01" rpy="1.5707963268 0.0 0.0"/>
    <limit effort="10" velocity="0.1" lower="-0.6" upper="0.6"/>
    <dynamics damping="0.7" friction="0.0"/>
  </joint>

Model launched in gazebo: rostopic list:

/arm_controller/command
/arm_controller/follow_joint_trajectory/cancel
/arm_controller/follow_joint_trajectory/feedback
/arm_controller/follow_joint_trajectory/goal
/arm_controller/follow_joint_trajectory/result
/arm_controller/follow_joint_trajectory/status
/arm_controller/state
/clock
/gazebo/link_states
/gazebo/model_states
/gazebo/parameter_descriptions
/gazebo/parameter_updates
/gazebo/set_link_state
/gazebo/set_model_state
/joint_states
/rosout
/rosout_agg
/tf
/tf_static

rosservice list:

/arm_controller/query_state
/boroscope_controller_spawner/get_loggers
/boroscope_controller_spawner/set_logger_level
/controller_manager/list_controller_types
/controller_manager/list_controllers
/controller_manager/load_controller
/controller_manager/reload_controller_libraries
/controller_manager/switch_controller
/controller_manager/unload_controller
/gazebo/apply_body_wrench
/gazebo/apply_joint_effort
/gazebo/clear_body_wrenches
/gazebo/clear_joint_forces
/gazebo/delete_light
/gazebo/delete_model
/gazebo/get_joint_properties
/gazebo/get_light_properties
/gazebo/get_link_properties
/gazebo/get_link_state
/gazebo/get_loggers
/gazebo/get_model_properties
/gazebo/get_model_state
/gazebo/get_physics_properties
/gazebo/get_world_properties
/gazebo/pause_physics
/gazebo/reset_simulation
/gazebo/reset_world
/gazebo/set_joint_properties
/gazebo/set_light_properties
/gazebo/set_link_properties
/gazebo/set_link_state
/gazebo/set_logger_level
/gazebo/set_model_configuration
/gazebo/set_model_state
/gazebo/set_parameters
/gazebo/set_physics_properties
/gazebo/spawn_sdf_model
/gazebo/spawn_urdf_model
/gazebo/unpause_physics
/gazebo_gui/get_loggers
/gazebo_gui/set_logger_level
/joint_state_controller_spawner/get_loggers
/joint_state_controller_spawner/set_logger_level
/robot_state_publisher/get_loggers
/robot_state_publisher/set_logger_level
/rosout/get_loggers
/rosout/set_logger_level

I am new to programming and ROS. After a few weeks of research I dont know how to solve the problem. For more information just tell me and I will provide it.

Thanks in advance

ROI3000

gazebo links spawn in wrong position - need tips for inertia

Hi, I have a problem with gazebo spawning the links of my URDF model not correctly.

So figured out that it is still a problem with the inertia.

-I tried to fix it but could not make it stable. When adding link by link after the 5th part everything is collapsing. I removed the big table from the URDF, than there was a change but in the end it still collapses.

-I used the inertia calculated by Mashlab and I tried it with the inertia from SolidWorks, but there was no improvement. Using different masses for the links didnt change anything as well.

Question 1: should I build the complete testbench in solid works and than use the urdf_exporter? By now I build the URDF link by link...

Question 2: If I change the structure of the URDF in Gazebo but leave the old structure in MoveIt, would there be problems with the transformations ?

Here is the link to the repository of the projekt: github and the where the problem is URDF

1: wrong inertia gazebo fail ,gazebo fail2

2: looking good in Rviz, RViz all good

I would be very grateful I somebody could help me with this issue or giving tips on how to calculate better inertia.

1: wrong inertia gazebo fail ,gazebo fail2

2: looking good in Rviz, RViz all good

Thanks in advance

ROI3000

gazebo links spawn in wrong position - need tips for inertia

Hi, I have a problem with gazebo spawning the links of my URDF model not correctly.

So figured out that it is still a problem with the inertia.

-I tried to fix it but could not make it stable. When adding link by link after the 5th part everything is collapsing. I removed the big table from the URDF, than there was a change but in the end it still collapses.

-I used the inertia calculated by Mashlab and I tried it with the inertia from SolidWorks, but there was no improvement. Using different masses for the links didnt change anything as well.

Question 1: should I build the complete testbench in solid works and than use the urdf_exporter? By now I build the URDF link by link...

Question 2: If I change the structure of the URDF in Gazebo but leave the old structure in MoveIt, would there be problems with the transformations ?

Here is the link to the repository of the projekt: github and the where the problem is URDF

1: wrong inertia gazebo fail ,gazebo fail2

2: looking good in Rviz, RViz all good

I would be very grateful I somebody could help me with this issue or giving tips on how to calculate better inertia.

Thanks in advance

ROI3000

gazebo links spawn in wrong position URDF collapsing - need tips for inertia

Hi, I have a problem with gazebo spawning the links of my URDF model not correctly.

So figured out that it is still a problem with the inertia.

-I tried to fix it but could not make it stable. When adding link by link after the 5th part everything is collapsing. I removed the big table from the URDF, than there was a change but in the end it still collapses.

-I used the inertia calculated by Mashlab and I tried it with the inertia from SolidWorks, but there was no improvement. Using different masses for the links didnt change anything as well.

Question 1: should I build the complete testbench in solid works and than use the urdf_exporter? By now I build the URDF link by link...

Question 2: If I change the structure of the URDF in Gazebo but leave the old structure in MoveIt, would there be problems with the transformations ?

Here is the link to the repository of the projekt: github and the where the problem is URDF

1: wrong inertia gazebo fail ,gazebo fail2

2: looking good in Rviz, RViz all good

I would be very grateful I somebody could help me with this issue or giving tips on how to calculate better inertia.

Thanks in advance

ROI3000

gazebo URDF collapsing - need tips for inertiainertia (solved)

Hi, I have a problem with gazebo spawning the links of my URDF model not correctly.

So figured out that it is still a problem with the inertia.

-I tried to fix it but could not make it stable. When adding link by link after the 5th part everything is collapsing. I removed the big table from the URDF, than there was a change but in the end it still collapses.

-I used the inertia calculated by Mashlab and I tried it with the inertia from SolidWorks, but there was no improvement. Using different masses for the links didnt change anything as well.

Question 1: should I build the complete testbench in solid works and than use the urdf_exporter? By now I build the URDF link by link...

Question 2: If I change the structure of the URDF in Gazebo but leave the old structure in MoveIt, would there be problems with the transformations ?

Here

Question 1: should I build the complete testbench in solid works and than use the urdf_exporter? By now I build the URDF link by link...

'-->I didnt go that way but: I reduced fixed Joints to zero (except base_link). Making assemblies in SolidWorks and using the inertia from the urdf_exporter made it nearly stable when spawning. The model is the link to the repository of the projekt: github and the where the problem is URDF

1: wrong inertia gazebo fail ,gazebo fail2

2: looking good in Rviz, RViz all good

I would be very grateful I somebody could help me with this issue or giving tips still a little shaky, I will keep working on how to calculate better inertia.

Thanks in advance that.

ROI3000