ROS2 Navigation

Note

ROS 2 Navigation with Isaac Sim is fully supported on Linux. On Windows, ROS 2 Navigation with Isaac Sim is partially supported and could potentially produce errors.

Learning Objectives

This ROS2 sample demonstrates Omniverse Isaac Sim integrated with ROS2 Nav2.

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 and ROS 2 Installation, make sure the ROS2 workspace environment is setup correctly.

Note

In Windows 10 or 11, depending on your machine’s configuration, RViz2 may not open properly.

Running Nav2

Using the Nova Carter Description Package (Optional)

The Nova Carter description package contains the robot geometry including meshes that can be used to visualize the robot in RViz2. Follow the steps below to configure this description package for Isaac Sim workflows:

  1. Complete the steps outlined in the Isaac ROS Development Environment Setup.

  2. Complete the steps outlined in the Repositories Setup.

  3. In a new terminal navigate to the folder containing launch files in the nova_carter_description package and create a new file called nova_carter_description_isaac_sim.launch.py.

    cd ${ISAAC_ROS_WS}/src/nova_carter/nova_carter_description/launch
gedit nova_carter_description_isaac_sim.launch.py

    Copy the following snippet into the new file and save.

    # SPDX-FileCopyrightText: NVIDIA CORPORATION & AFFILIATES
# Copyright (c) 2024 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
# SPDX-License-Identifier: Apache-2.0

# flake8: noqa: F403,F405
import isaac_ros_launch_utils as lu
from isaac_ros_launch_utils.all_types import *
from launch import LaunchDescription
from launch_ros.actions import Node


def generate_launch_description() -> LaunchDescription:
    args = lu.ArgumentContainer()
    args.add_arg('calibrated_urdf_file', default='/etc/nova/calibration/isaac_calibration.urdf')

    return LaunchDescription([
        lu.add_robot_description(nominals_package='nova_carter_description',
                                nominals_file='urdf/nova_carter.urdf.xacro',
                                robot_calibration_path=args.calibrated_urdf_file),
    ])

  4. Inside the Docker container build all the dependencies for nova_carter_description package.

    cd /workspaces/isaac_ros-dev
colcon build --packages-up-to nova_carter_description

  5. In the Docker container source the isaac_ros-dev workspace and run the custom launch file that you just created above to start publishing the robot description to the /robot_descritption topic.

    source /workspaces/isaac_ros-dev/install/setup.bash
ros2 launch nova_carter_description nova_carter_description_isaac_sim.launch.py

  6. Outside of the container (on the host), open a new terminal and ensure ROS is installed and that the Isaac Sim Workspace is sourced. Additionally, source the nova_carter_description package workspace in a new terminal.

    source ${ISAAC_ROS_WS}/install/setup.bash

  7. In the same terminal run the ~carter_navigation~ launch file:

    ros2 launch carter_navigation carter_navigation.launch.py

  8. Verify that the robot model is automatically loaded in the scene in Rviz.

Sending Goals Programmatically

Note

The isaac_ros_navigation_goal package is fully supported on Linux. On Windows, running this package could potentially produce errors.

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 -> ROS2 -> Navigation -> Carter 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 opens and begins 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 stops 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.

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.

Next Steps

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

Further Learning