- Ubuntu 18.04
- ROS Melodic
- RosBag for offline testing
- input
/points_raw (sensor_msgs::PointCloud2) - output
/object_cluster/clusters (object_cluster::ClusterArray)
/object_cluster/detected_bounding_boxs (jsk_recognition_msgs::BoundingBoxArray)
-
Move the project into the ROS workspace (e.g. ~/ros_ws/src/)
-
Build the project in the ROS workspace
cd ros_ws/ catkin_make
-
Setup the parameters in object_cluster.launch
<!-- Common Parameters --> <arg name="sensor_model" default="64" doc="Divide the point cloud into nested circular regions centred at the sensor. 16/32/64"/> <arg name="print_fps" default="true" doc="Print the FPS on the terminal."/> <!-- Crop Parameters --> <arg name="x_axis_min" default="0.0" doc="Backward"/> <arg name="x_axis_max" default="80.0" doc="Forward"/> <arg name="y_axis_min" default="-20.0" doc="Right"/> <arg name="y_axis_max" default="20.0" doc="Left"/> <arg name="z_axis_min" default="-0.70" doc="Downward (e.g. the height of the LIDAR.)"/> <arg name="z_axis_max" default="0.5" doc="Upward"/> <!-- Cluster Parameters --> <arg name="cluster_size_min" default="10" doc="The minimum number of the points to be clustered."/> <arg name="cluster_size_max" default="10000" doc="The maximum number of the points to be clustered."/>
Note: The default parameters work well with OUSTER-64.
-
Run the Object Cluster
-
Source the setup.bash
cd ros_ws source devel/setup.bash
-
Launch the Object Cluster node
roslaunch object_cluster object_cluster.launch
-
-
Play the rosbag for offline testing
rosbag play offline_testing.bag
-
Subscribe the clustered objects messages from the topic
/object_cluster/detected_bounding_boxs
.<!-- Demo --> --- header: seq: 11876 stamp: secs: 1641871314 nsecs: 313279366 frame_id: "velodyne" pose: position: x: 4.80017709732 y: 14.9710798264 z: -0.106718301773 orientation: x: 0.0 y: 0.0 z: 0.0 w: 0.0 dimensions: x: 0.2 y: 0.642765045166 z: 1.15449118614 value: 0.0 label: 0 ---