7.2.8. ROS2 Navigation

7.2.8.1. Learning Objectives

In this ROS2 sample, we are demonstrating Omniverse Isaac Sim integrated with ROS2 Nav2.

7.2.8.2. Getting Started

Important

Make sure to source your ROS 2 installation from the terminal before running Isaac Sim. If sourcing ROS 2 is a part of your bashrc then Isaac Sim can be run directly.

Prerequisite

• This ROS2 Navigation sample is only supported on ROS2 Foxy and Humble.

• The Nav2 project is required to run this sample. To install Nav2 refer to the Nav2 installation page.

• Enable the omni.isaac.ros2_bridge Extension in the Extension Manager window by navigating to Window > Extensions.

• This tutorial requires carter_navigation and isaac_ros_navigation_goal ROS2 packages which are provided as part of your Omniverse Isaac Sim download. These ROS2 packages are located inside the appropriate ros2_ws (foxy_ws or humble_ws). They contain the required launch file, navigation parameters, and robot model. Complete ROS & ROS 2 Installation, make sure the ROS2 workspace environment is setup correctly.

7.2.8.3. Nav2 Setup

This block diagram shows the ROS2 messages required for Nav2.

As described above, the following topics and message types being published to Nav2 in this scenario are:

ROS2 Topic	ROS2 Message Type
/tf	tf2_msgs/TFMessage
/odom	nav_msgs/Odometry
/map	nav_msgs/OccupancyGrid
/point_cloud	sensor_msgs/PointCloud
/scan	sensor_msgs/LaserScan (published by an external pointcloud_to_laserscan node)

7.2.8.3.1. Occupancy Map

In this scenario, an occupancy map is required. For this sample, we are generating an occupancy map of the warehouse environment using the Occupancy Map Generator extension within Omniverse Isaac Sim.

1. Go to Isaac Examples -> ROS -> Navigation to load the warehouse scenario.

2. At the upper left corner of the viewport, click on Camera. Select Top from the dropdown menu.

3. Go to Isaac Utils -> Occupancy Map.

4. In the Occupancy Map extension, ensure the Origin is set to X: 0.0, Y: 0.0, Z: 0.0. For the lower bound, set Z: 0.1. For the Upper Bound, set Z: 0.62. Keep in mind, the upper bound Z distance has been set to 0.62 meters to match the vertical distance of the lidar onboard Carter with respect to the ground.

5. Select the warehouse_with_forklifts prim in the stage. In the Occupancy Map extension, click on BOUND SELECTION. The bounds of the occupancy map should be updated to incorporate the selected warehouse_with_forklifts prim. The map parameters should now look similar to the following:

   

   A perimeter will be generated and it should resemble this Top View:

   

6. Remove the Nova_Carter_ROS prim from the stage.

7. Once the setup is complete, click on CALCULATE followed by VISUALIZE IMAGE. A Visualization popup will appear.

8. For Rotate Image, select 180 degrees and for Coordinate Type select ROS Occupancy Map Parameters File (YAML). Click RE-GENERATE IMAGE. The ROS camera and Isaac Sim camera have different coordinates.

9. Occupancy map parameters formatted to YAML will appear in the field below. Copy the full text.

10. Create a YAML file for the occupancy map parameters called carter_warehouse_navigation.yaml and place it in the maps directory which is located in the sample carter_navigation ROS2 package (carter_navigation/maps/carter_warehouse_navigation.yaml).

11. Insert the previously copied text in the carter_warehouse_navigation.yaml file.

12. Back in the visualization tab in Omniverse Isaac Sim, click Save Image. Name the image as carter_warehouse_navigation.png and choose to save in the same directory as the map parameters file.

    The final saved image will look like the following:

    

An occupancy map is now ready to be used with Nav2!

7.2.8.4. Running Nav2

1. Go to Isaac Examples -> ROS -> Navigation to load the warehouse scenario.

2. Click on Play to begin simulation.

3. In a new terminal, run the ROS2 launch file to begin Nav2.

   ros2 launch carter_navigation carter_navigation.launch.py
   

   RViz2 will open and begin loading the occupancy map. If a map does not appear, repeat the previous step.

4. Since the position of the robot is defined in the parameter file carter_navigation_params.yaml, the robot should already be properly localized. If required, the 2D Pose Estimate button can be used to re-set the position of the robot.

5. Click on the Navigation2 Goal button and then click and drag at the desired location point in the map. Nav2 will now generate a trajectory and the robot will start moving towards its destination!

Note

The Carter robot uses the RTX Lidar by default. You can add people assets into the scene and they will be detected by the Lidar when being passed to Nav2.

Note

Some of the ROS2 Image publisher pipelines in the hawk cameras are disabled by default to improve performance. To start publishing images, open the _hawk action graphs found under the robot prim and enable the _camera_render_product nodes. The ROS Camera publisher nodes which are downstream of the render product nodes should be enabled by default and will only start publishing when the render product node is enabled.

7.2.8.5. Sending Goals Programmatically

The isaac_ros_navigation_goal ROS2 package can be used to set goal poses for the robot using a python node. It is able to randomly generate and send goal poses to Nav2. It is also able to send user-defined goal poses if needed.

1. Make any changes to the parameters defined in the launch file found under isaac_ros_navigation_goal/launch as required. Make sure to re-build and source the package/workspace after modifying its contents.

   The parameters are described below:

   • goal_generator_type: Type of the goal generator. Use RandomGoalGenerator to randomly generate goals or use GoalReader for sending user-defined goals in a specific order.

   • map_yaml_path: The path to the occupancy map parameters yaml file. Example file is present at isaac_ros_navigation_goal/assets/carter_warehouse_navigation.yaml. The map image is being used to identify the obstacles in the vicinity of a generated pose. Required if goal generator type is set as RandomGoalGenerator.

   • iteration_count: Number of times goal is to be set.

   • action_server_name: Name of the action server.

   • obstacle_search_distance_in_meters: Distance in meters in which there should not be any obstacle of a generated pose.

   • goal_text_file_path: The path to the text file which contains user-defined static goals. Each line in the file has a single goal pose in the following format: pose.x pose.y orientation.x orientation.y orientation.z orientation.w. Sample file is present at: isaac_ros_navigation_goal/assets/goals.txt. Required if goal generator type is set as GoalReader.

   • initial_pose: If initial_pose is set, it will be published to /initialpose topic and goal poses will be sent to action server after that. Format is [pose.x, pose.y, pose.z, orientation.x, orientation.y, orientation.z, orientation.w].

2. Go to Isaac Examples -> ROS -> Navigation to load the warehouse scenario.

3. Click on Play to begin simulation.

4. In a new terminal, run the ROS2 launch file to begin Nav2.

   ros2 launch carter_navigation carter_navigation.launch.py
   

   RViz2 will open and begin loading the occupancy map. If a map does not appear, repeat the previous step.

5. Run the isaac_ros_navigation_goal launch file, to start sending goals automatically:

   ros2 launch isaac_ros_navigation_goal isaac_ros_navigation_goal.launch.py
   

Note

The package will stop processing (setting goals) once any of the below conditions are met:

1. Number of goals published till now >= iteration_count.

2. If GoalReader is being used then if all the goals from file are published, or if condition (1) is true.

3. A goal is rejected by the action server.

4. In case of RandomGoalGenerator, if a goal was not generated even after running the maximum number of iterations, which is rare but could happen in very dense maps.

To learn more about programmatically sending navigation goals to multiple robots simultaneously see Sending Goals Programmatically for Multiple Robots.

7.2.8.6. Summary

In this tutorial, we covered

1. Occupancy map

2. Running Isaac Sim with Nav2.

3. Running the Isaac ROS2 Navigation Goal package to send nav goals programmatically.

7.2.8.6.1. Next Steps

Continue on to the next tutorial in our ROS2 Tutorials series, Multiple Robot ROS2 Navigation to move multiple navigating robots with ROS2.

7.2.8.6.2. Further Learning

• To learn more about Nav2, refer to the project website: https://navigation.ros.org/index.html

• More about Occupancy Map Generator.

• Explore the inner workings of RTX Lidar sensors by learning How They work, the RTX Lidar Nodes that use them, and how to get RTX Lidar Synthetic Data.