Zenzu's profile - activity

ROS 2 Cannot find python modules Hi, I have a python program that I would like to integrate into ROS but I have a proble

I have solved it by loading the second robot_description with a different name and creating links between these two mode

Ok I have found the reason. The delay is caused by choosing the fixed frame in Rviz. If the fixed frame is panda_link0 e

Ok I have found the reason. The delay is caused by choosing the fixed frame in Rviz. If the fixed frame is panda_link0 e

I am running desktop Ubuntu and my specs are: 32GB RAM, i7-11700K, Geforce 1050Ti

I am sorry for misunderstanding. I do not have two robots, just two robot descriptions (of course with different names).

I am sorry for misunderstanding. I do not have two robots, just two robot descriptions (of course with different names).

Thank you, it worked to some extent, however, when I have two robot models visualized in Rviz, Panda's model has a huge

Hi, I am using the original franka_ros packages (that I do not want to modify) with my custom packages. My problem is that the original franka_control.launch file loads the panda's urdf as the robot_description parameter, however, I want to add other components within a scene such as table, etc. I have two ideas how can I add these components but I am not sure, which one is better or what is the correct way of doing it.

A) Overwrite the original robot_description with these commands:

  <!-- Overwrite panda description with panda_on_table description -->
  <arg default="$(find franka_mis_description)/urdf/table_mis.xacro" name="model"/>
  <param command="$(find xacro)/xacro $(arg model)" name="robot_description"/>

Do I need to load robot_state_publisher and joint_state_publisher again?

B) Is there a way to extend the loaded robot_description with another urdf file?

Since I only needed to add static elements (table, etc) I have solved it by using @aarsh_t 's advice to load more models

@aarsh_t and for creating a link between these two models, can I just use a static publisher?

@aarsh_t and for creating a ling between these two models, can I just use a static publisher?

Overwrite robot_description Hi, I am using the original franka_ros packages (that I do not want to modify) with my custo

MoveIt! MotionPlanning plugin does not update the start state Hi, I have set the effort_joint_trajectory_controller with

Hello, I am trying to implement Reinforcement Learning within ROS for a KUKA manipulator. Therefore, I need to let run program for a day in order to learn the KUKA manipulator accomplish certain task. However, after few hours the program is killed with this error messages:

ERROR: Failed to fetch current robot state

INFO: Didn't received robot state (joint angles) with recent timestep within 1 second. Check clock synchronization if you are running ROS across multiple machines!

To provide more info: I am using Gazebo simulator with a small cube and the task is to push the object into the goal location. Regarding to Moveit, I am using TrackIK solver and I generate [x,y,z] coordinates from the neural network directly to trajectory_controller using setFromIK function.

I am using:

ros::AsyncSpinner spinner(2);
spinner.start();

so this should not be a problem. I have already tried increase number of threads but its not working. Could someone point me to the right direction? This is the problem why the learning process is failing.

Here is a link to my github with the custom plugin that I used. However, it is in C++. Hope it helps. https://github.com

Here is a link to my github with the code that I used. However, it is in C++. Hope it helps. https://github.com/TomasMer

Here is a link to my github, where is the code that I used. However, it is in C++. Hope it helps. https://github.com/Tom

I am trying to add a box, with which a KUKA robot could interact (push it into the desired position). Besides being able to interact with it, I have to be able to change its position through a node. Could you point me into the right direction, what should be the right way of doing it? My ideas/attempts are:

1. If I create a box in same urdf file as KUKA, the arm avoids the object because of the collision avoidance. I believe that if I could somehow disable this, I could push the object with arm. Will I be able to change the position of the object using tf?
2. The second idea is to insert the object in Gazebo using the Gazebo GUI. It seems to work but I am getting an error, right after I have inserted the object:

[ERROR] [1581331637.324486823]: This robot has a joint named "joint_a1" which is not in the gazebo model.

[FATAL] [1581331637.325261818]: Could not initialize robot simulation interface

So far I have not find any impact of that error. I am able to change its position using /gazebo/set_model_state service. To visualize the object in Rviz, tf broadcaster is needed, right? I think that the actual position of the object I could get from /gazebo/get_model_state service.

3) The third idea was to create two separate robot_models (2 separate urdf files with 2 urdf_spawn nodes). But with this idea I have had no success so far.

Could you point me to the right direction? How can it be solved?

PS: for better understanding I am trying to create this environment in ROS using KUKA: https://gym.openai.com/envs/FetchPush...

Hi, did you solve this issue?

I am sorry for late answer.. the problem was with setting the goal tolerance as well as with configuring std::cout preci

Hi, I would like to ask if it is possible to accomplish the repeatability of 10 micrometers regarding robotic arms using Moveit, ros_control and tf for control. I need such precision for my project. I have tried to move an arm from theposition.x = -0.0528998 to the position.x = -0.0528999, although the move_group.getCurrentPose() and also the tf library provide me with the value of -0.0528155 for both positions. Is there a way how to improve the precision or is it something that cant be accomplished?

The problem was with setting the goal tolerance as well as with configuring std::cout precision. It seems that the preci

I am sorry for late answer.. the problem was with setting the goal tolerance as well as with configuring std::cout preci

We are developing a small robotic arm for lab experiments that consists of handling samples of size under 1mm. But still

We are developing a small robotic arm for lab experiments that consists of handling samples of size under 1mm.