Skip to main content

O3DE: Creating realistic simulations with open-source game engine

UB2.147 | Day 2 | 14:10 - 14:35 | Speakers: Jan Hanca

O3DE: Creating realistic simulations with open-source game engine
A picture of a devroom at FOSDEM 2024
Open in browser

Notes

Abstract

A variety of simulation platforms are available to support research and development in robotics and biomechanics, with popular open-source options like Gazebo, Webots, and MuJoCo, as well as proprietary alternatives such as NVIDIA Isaac Lab, which provide end-to-end solutions for simulation and machine learning. Additionally, tools integrating game engines like Unity and Unreal into robotic simulations have been developed to leverage their advanced rendering capabilities.

O3DE (Open 3D Engine) is a versatile, open-source, real-time 3D engine designed for creating high-fidelity games, robotic simulations, and immersive 3D environments. Its modular architecture supports seamless integration with robotics systems through ROS 2 middleware, while its AAA-quality renderer produces photorealistic visuals ideal for machine learning training and evaluation.

Governed by the Open 3D Foundation, part of the Linux Foundation, O3DE’s codebase is freely available on GitHub under the Apache 2.0 and MIT licenses, making it a powerful and accessible tool for advancing robotics simulation and development.

Talk Plan (20 Minutes):

  1. Introduction to O3DE (3 minutes)

    • Provide an overview of O3DE.
    • Highlight its strengths and weaknesses.
  2. Getting Started with O3DE (5 minutes)

    • Explain the installation process and how to build O3DE from source.
    • Share available resources, including robotic project templates and ready-to-test projects (e.g., open-source demos from ROSCon 2022, 2023, and 2024).
  3. O3DE and the ROS 2 Ecosystem (5 minutes)

    • Discuss O3DE's integration with the ROS 2 ecosystem.
    • List built-in implementations, such as ROS 2 sensors, joint control, and robot control features.
    • Provide examples of use cases supported by the existing codebase, including importing robots from Gazebo.
  4. C++ Code Examples for ROS 2 (5 minutes)

    • Demonstrate C++ code snippets for implementing ROS 2 topic subscribers and publishers.
  5. Conclusion (2 minutes)

    • Share information on contributing to O3DE.
    • Highlight communication channels and ways to get involved with the community.

Resources: O3DE website O3DE repository O3DE repository with ROS2 integration and templates ROSCon2021 Demo using O3DE ROSCon2022 Demo using O3DE ROSCon2023 Demo using O3DE ROSCon2024 Demos using O3DE

Attachments

Speakers

Jan Hanca

Notice: The placeholder video image is licensed under CC BY-SA 4.0. The original image can be found hereChanges made to the image are: Cropped the image to a new ratio, part of the image was cut off.