Skip to content

youliangtan/realsense_gazebo_plugin

 
 

Repository files navigation

Intel RealSense Gazebo ROS plugin and model

Important note

This branch only supports ROS Melodic with Gazebo 9.

To use this plugin with ROS Kinetic and Gazebo 7 you should use the kinetic-devel branch.

Quickstart

Build the plugin

catkin build realsense_gazebo_plugin

Test it by running

roslaunch realsense_gazebo_plugin realsense.launch

Run the unittests

After building the plugin, you can run the unittests

rostest realsense_gazebo_plugin realsense_streams.test

Run the point cloud demo

Using depth_image_proc package, we can generate a point cloud from the depth image by running

# in terminal 1
roslaunch realsense_gazebo_plugin realsense.launch 
# in terminal 2 
roslaunch realsense_gazebo_plugin depth_proc.launch camera_name:=realsense_plugin
# Sanity Check, terminal 3
rostopic echo /realsense_plugin/camera/depth_registered/points

Then open Rviz, and display the /realsense_plugin/camera/depth_registered/points topic, you should see something like this Point cloud in Rviz

Run from URDF

# Terminal 1
roslaunch realsense_gazebo_plugin realsense_urdf.launch
# Terminal 2
roslaunch realsense_gazebo_plugin depth_proc.launch

This will behave the same as realsense.launch mentioned above, with the difference that it spawns the model from a URDF (see urdf folder). You can reuse this to plug the sensor in the robot of your choice.

Dependencies

This requires Gazebo 6 or higher and catkin tools for building.

The package has been tested on ROS melodic on Ubuntu 18.04 with Gazebo 9.

Changes made

  • urdf for r200 model, allign depth cam and rgb cam
  • retuned some launch files stuff
  • changes readme

TODO

  • If facing issue with gtest dep, refer to here
  • support realsense 435 series
  • real fix on 2 cam tf registration misallignment issue

Acknowledgement

This is continuation of work done by guiccbr for Intel Corporation.

Thanks to Danfoa for contributing the URDF integration.

About

Intel RealSense R200 Gazebo ROS plugin and model

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • C++ 94.9%
  • CMake 5.1%