As I mentioned last time, NVIDIA Isaac Sim, powered by Omniverse, is a scalable robotics simulation application and synthetic data generation tool that powers photorealistic, physically-accurate virtual environments to develop, test, and manage AI-based robots. This post is the part two of Isaac Sim’s post series, to read the part one, click here.

This time I will explain how to simulate a Scout V2 robot with a stereo camera and a 2D LiDAR. You will be able to view the sensors and the odometry on Rviz.

Demonstration

Simulation setup

First, you need to open the terminal and create a ROS2 workspace

1

$ mkdir -p scout_ws/src

go to the src folder and clone the repository

1
2

$ cd scout_ws/src
$ git clone https://github.com/leonlime/scout_v2_isaac.git

build the package

1
2
3

$ cd ..
$ source /opt/ros/foxy/setup.bash
$ colcon build

Running the simulation

Open the terminal, navigate to the workspace you created before and source ROS2. Don’t forget, always before opening Isaac Sim, the first thing to do is start a terminal and source your native ROS2

1
2
3

$ cd scout_ws
$ source /opt/ros/foxy/setup.bash
$ . install/setup.bash

Inside the Isaac Sim, open the scout_v2_simulation.usd file

Click on the play button on the left panel, go back to the terminal and check the ROS2 topics

1
2
3
4
5
6
7
8
9
10
11
12

$ ros2 topic list
/cmd_vel
/laser_scan
/left/camera_info
/left/rgb
/odom
/parameter_events
/point_cloud
/right/camera_info
/right/rgb
/rosout
/tf

To visualize the camera images, LiDAR and odometry, open Rviz

1

$ ros2 launch scout_v2_ros2 scout_v2_rviz.launch.py

You can navigate using teleop twist command

1

$ ros2 run teleop_twist_keyboard teleop_twist_keyboard

Thanks to use this tutorial and don’t forget, if something is wrong you can contact me on my e-mail or Github account.