Laserscan and Pointcloud Combiner

Robotics and Autonomous Systems Group, Research Engineering Facility, Research Infrastructure Queensland University of Technology

Introduction

The Laserscan and Pointcloud Combiner is a ROS 1 node that is designed for combining PointCloud2 and LaserScan input sources by selecting the nearest detection in a ROS application. The resulting detection is published to a LaserScan topic. The package supports at most 1 PointCloud2 source and 2 LaserScan sources. Due to resampling, quantization error may be introduced during the processing of combining multiple sources of different sampling patterns.

ROS Integration

The following figure illustrates the data flow between the Combiner and the ROS 1 infrastructure.

The input to the Combiner may include 1 to 3 sources, with at most one of them a PointCloud2 message topic. The others are LaserScan message topics. The LaserScan topics may be of different sampling patterns (i.e., sampling rate and sample angle range). The PointCloud2 input will pass through a filter that selects data in a specified z-axis range. This enables the selection of objects at different vertical positions and the ignoring of the ground plane.

The Combiner operates at a specified sampling pattern, which are specified as parameters. If the input sources include one LaserScan message topic, then the topic's sampling pattern may be adopted as the default output sampling pattern parameter.

The Combiner combines the input sources through selecting the nearest data point for every output sample. In the case of a PointCloud2 input message topic, the Combiner will convert the pointcloud data into laser scan data by quantization. This involves selecting the nearest point within a quantized angle range and z-axis range. In the case of a LaserScan input message topic, the Combiner will resample and again select the nearest point when there are two or more source data points. If there are more than one input sources, the Combiner will select the nearest point among the input sources for every output sample.

Parameters

The Combiner is customizable through various parameters. The parameters may be specified as private ROS parameters (read through the rospy.get_param() function calls) and as keyword parameters in instantiate an object of the main class LaserscanPointcloudCombiner in the executable program file /nodes/laserscan_combiner.py.

Input ROS Topics

Name	Default	Remarks
pointcloud_topic_name	None	A point cloud input topic (PointCloud2)
laserscan_in_1_topic_name	None	A laser scan input topic (LaserScan)
laserscan_in_2_topic_name	None	A laser scan input topic (LaserScan)

A None value for the above input ROS topic parameters implies that the input is not used.

Output ROS Topics

Name	Default	Remarks
laserscan_out_topic_name	/laser_link/scan	The combined laser scan output topic (LaserScan)
laser_frame_id	combined_laser	The frame id of the simulated source of the laser scan output

Note that laser_frame_id must have been specified in the scene (i.e., the TF tree).

Parameters for the Input PointCloud

Name	Default	Remarks
pointcloud_min_z	-0.2	The lower end of the z-axis range accepted by the Combiner (in meters) in the laser_frame_id frame
pointcloud_max_z	10.0	The upper end of the z-axis range accepted by the Combiner (in meters) in the laser_frame_id frame

Parameters for the Output LaserScan

Name	Default	Remarks
pub_rate	20	The rate of publishing the output LaserScan (Hz)
laser_samples	The input LaserScan or 120	The number of laser scan samples of each LaserScan message
laser_min_angle	The input LaserScan or -2.0	The lower end of the angle range of the output laser scan (radians)
laser_max_angle	The input LaserScan or 2.0	The upper end of the angle range of the output laser scan (radians)
laser_range_min	The input LaserScan or 0.1	The nearest end in the valid range of the output laser scan (meters)
laser_range_max	The input LaserScan or 20	The furthest end in the valid range of the output laser scan (meters)

Note that for the last five parameters, the laserscan_in_1_topic_name takes precedent over laserscan_in_2_topic_name as the default when both are specified as input sources.

Specifying Parameters in a Launch File

A sample launch file default.launch is provided as an example of how to specify the parameters.

<launch>
    <node name="laserscan_combiner" pkg="mgv_laserscan_combiner" type="laserscan_combiner.py" respawn="false" output="screen">
        <param name="pointcloud_topic_name" value="/velodyne_points"/>
        <param name="laserscan_in_1_topic_name" value="/bound_detector/scan"/>
        <!-- the package supports at most 2 laserscan input and 1 pointcloud input -->
        <!-- <param name="laserscan_in_2_topic_name" value="/another_detector/scan"/> -->
        <param name="laserscan_out_topic_name" value="/laser_link/scan"/>
        <param name="laser_samples" value="240" />     
        <!-- <param name="laser_frame_id" value="velodyne" />     
        <param name="laser_min_angle" value="-3.14" />     
        <param name="laser_max_angle" value="3.14" />     
        <param name="laser_range_min" value="1" />     
        <param name="laser_range_max" value="10" />  
        <param name="pointcloud_min_z" value="-0.2" />     
        <param name="pointcloud_max_z" value="10.0" />      
        <param name="pub_rate" value="10" />      -->
    </node>
</launch>

The above example specifies one input PointCloud2 source and one input LaserScan source. It specifies their topic names and the topic name of the output LaserScan. Beside the laser_sample, the default will be used for the other parameter values.

Specifying Parameters in Python

The Combiner may be executed as a ROS node through your own Python program. It is as simple as initializing the program as a ROS node and then instantiate an object of the class LaserscanPointcloudCombiner.

    rospy.init_node(LaserscanPointcloudCombiner.NODE_NAME)
    try:
        rospy.loginfo(f'{LaserscanPointcloudCombiner.__name__}: The node "{LaserscanPointcloudCombiner.NODE_NAME}" is running (ns: {rospy.get_namespace()})')
        robot_agent = LaserscanPointcloudCombiner()
        rospy.spin()
    except rospy.ROSInterruptException as e:
        rospy.logerr(e)

The Combiner parameters may be specified through keywords in the call to the constructor. Note that the Combiner parameters specified as keyword parameters of the constructor can be overruled by ROS parameters.

    rospy.init_node(LaserscanPointcloudCombiner.NODE_NAME)
    try:
        rospy.loginfo(f'{LaserscanPointcloudCombiner.__name__}: The node "{LaserscanPointcloudCombiner.NODE_NAME}" is running (ns: {rospy.get_namespace()})')
        robot_agent = LaserscanPointcloudCombiner(laser_samples=320, laserscan_out_topic_name=`/another_laser_link/scan`)
        rospy.spin()
    except rospy.ROSInterruptException as e:
        rospy.logerr(e)

Installation

This package should be installed under the src folder of a workspace. Use an existing workspace or create a new one as follows.

cd ~
mkdir -p catkin_ws/src

Change directory to the src folder under the workspace.

cd catkin_ws/src

Clone this repository and the package is downloaded as a folder called mgv_laserscan_combiner.

git clone https://github.com/REF-RAS/mgv_laserscan_combiner.git

Build the package and launch the ROS node

cd ~/catkin_ws
catkin_make

roslaunch mgv_laserscan_combiner default.launch

The node will wait for messages from the input sources if they are not already publishing in the ROS environment.

Developer

Dr Andrew Lui, Senior Research Engineer
Robotics and Autonomous Systems, Research Engineering Facility
Research Infrastructure
Queensland University of Technology

First version: 23 Jul 2024
Latest update: 24 Jul 2024