Gazebo (simulator)

Gazebo (simulator)
Name	Gazebo
Author	Open Source Robotics Foundation
Developer	Open Source Robotics Foundation, Community Contributors
Released	2004
Programming language	C++, Python, ROS plugins
Operating system	Linux, Windows, macOS
License	Apache License 2.0

Gazebo (simulator) is an open-source 3D robotics simulator used for testing robot models, sensors, and controllers in virtual environments. It provides realistic physics simulation, sensor emulation, and a plugin architecture to evaluate algorithms for navigation, manipulation, and perception without physical hardware. Gazebo is widely used alongside robotics frameworks and research institutions to accelerate development and reproducible experiments.

Overview

Gazebo integrates a dynamics engine, rendering engine, and sensor models to simulate complex interactions among robots, environments, and obstacles. It supports multi-robot scenarios, articulated mechanisms, and environmental effects such as lighting and contact dynamics, enabling benchmarking of autonomy stacks developed with frameworks like Robot Operating System, ROS 2, YARP, Player Project, and Microsoft Robotics Developer Studio. The simulator can interface with middleware, control libraries, and datasets from projects at institutions such as MIT, Stanford University, Carnegie Mellon University, ETH Zurich, and University of Tokyo.

History and Development

Gazebo originated in the early 2000s and evolved through contributions from research groups and companies. Initial development drew on work from robotics labs affiliated with Willow Garage, Stanford Artificial Intelligence Laboratory, and contributors associated with the University of Pennsylvania and Georgia Institute of Technology. Over time stewardship transitioned to the Open Source Robotics Foundation and later collaborations with organizations such as Toyota Research Institute, NVIDIA, and Open Robotics. Major milestones include support for advanced physics engines like ODE (physics engine), Bullet Physics Library, and DART (robotics) as well as integration efforts with simulation initiatives from NASA and DARPA.

Architecture and Features

Gazebo's architecture separates world description, physics, rendering, and plugins to provide modularity for researchers and developers. Worlds and robot models are commonly described with formats tied to projects at OpenAI and specifications from the Robot Operating System community, using URDF and SDF influenced by standards promoted by IEEE. Physics backends supported include engines pioneered by groups around Sony Computer Entertainment, Google DeepMind research collaborations, and open-source projects like Bullet Physics Library and ODE (physics engine). Rendering leverages graphics systems related to work by Khronos Group and hardware advances from NVIDIA and Intel Corporation. Sensor plugins emulate devices inspired by products from Velodyne, SICK, Intel Corporation (RealSense), and camera models used in datasets from ImageNet and KITTI. The plugin API allows custom controllers, world plugins, and sensor models developed by users affiliated with labs at University of California, Berkeley, Imperial College London, and Max Planck Institute for Intelligent Systems.

Supported Platforms and Integrations

Gazebo runs on major desktop platforms and integrates with middleware, robotics libraries, and cloud services. Native support targets distributions influenced by Canonical (company) and Debian packaging practices, while Windows and macOS ports reflect contributions from community members linked to Microsoft and Apple Inc.. Integrations include bridging to Robot Operating System, ROS 2, OpenSLAM projects, motion planning stacks like MoveIt, perception toolkits such as OpenCV, and machine learning frameworks from TensorFlow and PyTorch. Cloud and CI integrations have been developed alongside services offered by Amazon Web Services, Google Cloud Platform, and Microsoft Azure for scalable simulation workloads.

Use Cases and Applications

Researchers use Gazebo to validate algorithms in autonomous driving research conducted at Waymo and Cruise, aerial robotics trials with groups such as DJI and AeroVironment, and manipulation research led by teams at Boston Dynamics and ABB. Educational programs at institutions like Massachusetts Institute of Technology, ETH Zurich, and Tsinghua University employ Gazebo for coursework and competitions related to DARPA Robotics Challenge preparation and student teams participating in events sponsored by NASA and European Space Agency. Industry adopters use the simulator for prototyping robotic arms in manufacturing lines designed by Siemens and Bosch, and for logistics automation in warehouses developed by Amazon Robotics and Ocado Technology.

Community and Governance

Gazebo's development is guided by an open governance model with contributions from individual researchers, corporate engineers, and foundations. The project engages contributors connected to organizations such as Open Robotics, Open Source Robotics Foundation, Willow Garage, Toyota Research Institute, and NVIDIA and coordinates through online repositories and collaboration platforms used by communities including GitHub and GitLab. Educational outreach, workshops, and tutorials are frequently presented at conferences like ICRA, IROS, RSS, and ISRR and in collaboration with university labs and corporate research groups. The ecosystem includes third-party plugins, model repositories, and community-maintained packages that continue to expand Gazebo's applicability across robotics domains.

Category:Robotics software