Open joelbudu opened 3 years ago
@joelbudu , I would like to learn if grasp library is used for recognition picking in your scenario? In such a case, the object recognition result detected from RGB image is used and thus expecting organized 2D point cloud where depth image is aligned into color frame. While the ROS2 eloquent branch you're using seems provide 1D unordered point cloud. My suggestion is you raise the issue to the corresponding project.
Thanks for your response - Yes, the grasp library is used for recognition picking in my scenario. I did raise an issue there but was directed to raise it in this project since it's directly connected.
Will it be possible to support the L515 LiDAR camera in the https://github.com/intel/ros2_intel_realsense repository? I believe that may be the best solution at this time.
Joel, realsense-ros is the tree for new feature development and maintenance. You may try to raise the question to that project. For your ref, according to ROS2 interface, "Point clouds organized as 2d images may be produced by camera depth sensors".
I am trying to use this library with the L515 LiDar camera from intel. Since the https://github.com/intel/ros2_intel_realsense repository doesn't have support for that camera I attempted to use the eloquent branch of https://github.com/IntelRealSense/realsense-ros. This throws a Segmentation fault error when the /plan_grasps service is called and crashes the ROS package. I'm suspecting it may be due to a difference in the structure of PointCloud2 messages being delivered.
Please help