Blickfeld ROS2 package (:warning: ROS2 support is experimental )

This package provides a ROS2 node for publishing PointCloud2 messages from Blickfeld LiDAR devices. The driver is not yet available publicly. Please request the sources from the Blickfeld Sales department.

Supported devices

The Blickfeld ROS2 driver supports all currently available Blickfeld LiDARs such as Cube 1 and Cube Range 1.

Supported ROS2 Distributions

The Blickfeld ROS2 driver supports the following ROS2 Distributions

  • Eloquent Elusor

  • Foxy Fitzroy

Dependencies

To install the Blickfeld ROS2 driver, please make sure the following dependencies are installed on your system.

  • blickfeld-scanner-library (BSL) with system-wide protobuf installation. This package has been tested using BSL Version 2.13.0

  • ROS Eloquent Installation with Ubuntu 18.04 / ROS Foxy Installation with Ubuntu 20.04

  • diagnostic_updater can be acquired via your distribution’s package manager, ${ROS_DISTRO} should be your ROS 2 version. (e.g. foxy or eloquent)

    $ sudo apt install ros-${ROS_DISTRO}-diagnostic-updater
    $ sudo apt install ros-${ROS_DISTRO}-diagnostic-msgs
    

    or via

    $ rosdep update
    $ rosdep install --from-paths src --ignore-src -r -y --skip-keys "blickfeld-scanner"
    

    The -skip-keys will instruct rosdep to not check blickfeld-scanner, which is not a ros package.

  • For using xml formatted launch files with ROS2 Eloquent, install the dependent packages as shown below.

    :information_source: In ROS2 Foxy, the support for xml formatted launch files are included by default, and hence not required to be installed

    $ sudo apt install ros-eloquent-launch-xml
    $ sudo apt install ros-eloquent-launch-yaml
    
  • colcon installation: required for building the workspace

:warning: The default DDS of ROS2 Foxy FastRTPS is not compatible with BSL. Hence for ROS2 Foxy, we need to install DDS from other vendors. (e.g. Eclipse Cyclone or RTI Connext)

Build

Before building ensure that your ROS DISTRO is sourced. You will need to run this command on every new shell you open to have access to the ROS 2 commands. ${ROS_DISTRO} should be your ROS 2 version. (e.g. foxy or eloquent)

$ source /opt/ros/${ROS_DISTRO}/setup.bash

To build the ROS2 driver, request the ROS2 driver from the Blickfeld Sales department and decompress the archive into your ROS2 workspace and run.

$ colcon build --symlink-install --cmake-clean-first

Switching to desired DDS (:warning: Necessary only for ROS2 Foxy )

:information_source: In the case of ROS2 Eloquent the default DDS works flawlessly with Blickfeld scanner library. Hence, we can run the node using the standard ros2 run approach. But, in the case of ROS2 Foxy, we need to switch from the default DDS to a compatible one. This can be done using either of the two approaches.

  1. Exporting the RMW_IMPLEMENTATION for the present terminal session

    $ export RMW_IMPLEMENTATION=rmw_cyclonedds_cpp
    
  2. Switching to the desired RMW_IMPLEMENTATION while running by adding a prefix before the ros2 run

    $ RMW_IMPLEMENTATION=rmw_cyclonedds_cpp ros2 run  PKG_NAME NODE_NAME ARGUMENTS
    

Running the Blickfeld ROS node

Before running, ensure that your colcon workspace is sourced. This is achieved by sourcing the setup script in your colcon workspace.

$ source install/setup.bash

You can launch the ROS driver in a DHCP controlled Network by providing the Hostname of the LiDAR device (you can check and set the Hostname in the WebGUI of the device) e.g:

$ ros2 run blickfeld_driver blickfeld_node --ros-args -p host:=cube-0056 --remap __node:=bf_lidar

If you connect the LiDAR directly to your PC or don’t have a DHCP Server you can launch the driver by using the IP configured in the WebGUI of the Device or the Fallback-IP of the Device (make sure to adjust your network-settings accordingly) e.g:

$ ros2 run blickfeld_driver blickfeld_node --ros-args -p host:=192.168.26.26 --remap __node:=bf_lidar

You can also start it with more parameters, e.g.:

$ ros2 run blickfeld_driver blickfeld_node --ros-args -p host:=cube-0056 -p publish_ambient_light:=true -p publish_intensities:=false

To visualize the published data, run rviz2

$ ros2 run rviz2 rviz2

If you prefer a similar way to launch, like you would in ROS(1) (beware of the new “.xml”), you can start the driver like this:

$ ros2 launch -a blickfeld_driver live_scanner.launch.xml host:=cube-0056

Or if you prefer to put the parameters into a yaml, run the python launch file. It reads the two yaml file containing the configurations.

$ ros2 launch blickfeld_driver live_scanner_yaml.launch.py

The yaml files are located in the ros2_blickfeld_driver/config folder of the blickfeld_driver package. There are two yaml files:

  1. driver_config.yaml

    Configures host id, output point cloud topic etc.

  2. blickfeld_scanner.rviz

    Configures the parameters for Rviz visualization, e.g. topic of point cloud, configuration of opened viewers etc.

Parameters

All the following parameters are available in launch and run:

Argument Default Note
host (required) The host name or the IP of the device you want to publish the point clouds from, e.g., cube-0028, 192.168.26.26.
node_name bf_lidar Name of this ROS node.
point_cloud_out $(var node_name)/points_raw Topic to publish the PointCloud2 message to.
remap true Remap this node’s input/output topics to commonly used ones.
If false, canonical names in the form of $(var node_name)/foo_(in/out) are used, depending on whether a topic is an input or an output topic.
lidar_frame_id lidar The ROS TF where the point cloud should be in.
rviz false Start rviz if this argument is true.
use_lidar_timestamp true Set to true if the timestamp in the ROS point cloud message should be generated from the device timestamp; otherwise the timestamp will be the ROS time when the frame was received on ROS.
publish_intensities true Set to true if the PointCloud2 message should contain intensities.
publish_ambient_light false Set to true if the PointCloud2 message should contain the ambient light level.
publish_explicit_range false Set to true if the PointCloud2 message should contain the range field.
publish_no_return_points false Set to true if the PointCloud2 message should contain points in a given range for pulses without a return.
no_return_point_range 1.0 Dummy range for points of pulses without a return.
publish_all_returns false Set to true to publish all the returns for every point. Adds the field return_id (ID for each returned point) to the point cloud.
publish_time_delta false If true, the PointCloud2 message will contain the timestamp field per point to represent the time offset from the start of the frame.
publish_point_id false Add the scanline_id field, the scanline_point_index field (= the point’s number in the scan line), and the point_id (= frame-global point ID) to PointCloud2 message.
If the device is configured to return multiple return points (multiple reflections), all these three IDs will be identical; only the return_id will differ.

Debugging

Add ‘__log_level:=debug’ to see details:

$ ros2 run blickfeld_driver blickfeld_node --ros-args --remap __node:=bf_lidar -p host:=cube-0056 -p publish_ambient_light:=true -p publish_no_return_points:=true __log_level:=debug

ROS2 message content

To look at the ROS message sent by this ROS node, you can use

$ debugging_tools/ros2-topic-echoparsed.py

It shows all the fields present in the message.

License

This package is released under a BSD 3-Clause License (see also https://opensource.org/licenses/BSD-3-Clause)