lanelet2_rviz2 ROS2 package - OSMVisualizer

OSMVisualizer is a ROS 2 node for visualizing OpenStreetMap (OSM) data in RViz2. It processes .osm files specified via launch parameters and publishes markers for map elements, such as nodes, ways, and relations.

Features

• Static File Loading: Reads the .osm filename from a launch file parameter (osm_filename).
• Advanced Visualization in RViz2:
  • Displays map relations as colored line strips:
    • Left ways are shown in lime.
    • Right ways are shown in red.
  • Relation areas are filled using polygon triangulation.
  • Speed limits are visualized with a gradient color scale:
    • Green for lower speeds.
    • Blue for medium speeds.
    • Red for higher speeds.
  • One-way relations are marked with dynamically positioned direction arrows.
  • Parking lots are parking spaces visualized as filled areas.
• Dynamic Parameter Reconfiguration: Modify visualization settings like line_width, frame_id, or speed_color_max during runtime without restarting the node.
• Reusable Data Structures: Efficiently processes nodes, ways, and relations.

Topics

Published

• /osm_relations (visualization_msgs/msg/MarkerArray): Marker array for relations (left and right ways).
• /osm_speed (visualization_msgs/msg/MarkerArray): Marker array for relation areas, visualized with speed gradient colors. --TODO (currently only one publisher)
• /osm_one_way_arrows (visualization_msgs/msg/MarkerArray): Marker array for one-way direction arrows. --TODO

Parameters

Node Parameters

• osm_filename (string): Path to the .osm file to process.
• frame_id (string): The coordinate frame for visualization markers (default: "map_gyor_0").
• speed_color_max (double): Maximum speed limit for gradient visualization (default: 90.0).

Usage

1. Build the Project

Make sure your ROS 2 workspace is set up.

Install the necessary dependencies:

sudo apt install `sudo apt install ros-humble-lanelet2`

Clone this repository into the src directory and build it:

cd ~/ros2_ws/src && git clone https://github.com/jkk-research/lanelet2_rviz2

cd ~/ros2_ws && colcon build --packages-select lanelet2_rviz2 --symlink-install

source ~/ros2_ws/install/setup.bash

2. Run the Node

Launch the OSMVisualizer node with a launch file specifying the .osm filename:

ros2 launch lanelet2_rviz2 visualize_osm.launch.py

ros2 run lanelet2_rviz2 visualize_osm --ros-args -p osm_filename:=/mnt/c/map/lanelet.osm -p frame_id:=map

3. Visualize in RViz2

• Open RViz2:

  rviz2

• Add a MarkerArray display.
• Set the topics to:
  • /osm_relations for line strips representing left and right ways.
  • /osm_speed for filled relation areas with speed gradient.
  • /osm_one_way_arrows for direction arrows.

4. Dynamic Parameter Updates

Update visualization parameters during runtime using the ros2 param set command. Example:

ros2 param set /osm_visualizer line_width 1.5

Or with rqt_reconfigure:

ros2 run rqt_reconfigure rqt_reconfigure

Usage with Pointcloud

Option 1. - pcl_ros

sudo apt install ros-$ROS_DISTRO-pcl-ros

ros2 run pcl_ros pcd_to_pointcloud --ros-args -p file_name:=/home/he/dlio_map2.pcd -p tf_frame:=map_gyor_0 -p publishing_period_ms:=500 --qos-reliability best_effort

Option 2. - pcd_publisher

github.com/jkk-research/pcd_publisher

ros2 run pcd_publisher pcd_publisher --ros-args -p pcd_file_path:=/home/dev/autoware_map/gyor_campus/pointcloud_map.pcd -p topic_name:=/points_campus -p frame_id:=map_gyor_0

Troubleshooting

The software was designed to work in our workflow, which mainly relies on maps authored or edited with the Vector Map Builder of TIER4. It will likely work with Lanelet2 maps authored or written by different software, but if you run into unsupported map errors, try loading and exporting your map in the TIER4 map editor.

Acknowledgments

• Lanelet2 is used for parsing the OSM files.
• Earcut is used for polygon triangulation, enabling visualization of filled areas in RViz2.