Point-Cloud-Based Labware Detection

Detect the oriented bounding boxs of labware on the table based on point cloud captured in multiple robot poses, with the Cluster Algorithms.

$ rosrun robot_control robot_init.py
$ roslaunch pc_detection pc_slam.launch

Available args:

  <arg name="pose_file" default="slam_points.txt"/>
  <arg name="save_file" default="merged_cloud.pcd"/>
  <arg name="marker_file" default="marker_position"/>

Use your own pose_file and marker_file.

Marker publisher

$ rosrun pc_detection marker_pulish.py