Simulate a DIY robot vacuum in Gazebo with mapping and navigation before building real hardware.
Drive a simulated or physical OOMWOO One vacuum around with keyboard teleop while it builds a map.
Use it as a starting robot description for building your own ROS 2 based vacuum robot.
| makerspet/oomwoo-one | 1038lab/comfyui-agnes-ai | andyuneducated/resolve-ai | |
|---|---|---|---|
| Stars | 18 | 18 | 18 |
| Language | Python | Python | Python |
| Setup difficulty | hard | easy | hard |
| Complexity | 4/5 | 2/5 | 4/5 |
| Audience | developer | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires ROS 2 Jazzy and Gazebo installed, connecting a physical robot needs SangamIO set up on the vacuum first.
OOMWOO One is the software description for a real, physical robot vacuum cleaner you can build yourself, part of a broader open source project called OOMWOO. This repository specifically contains the robot's description for ROS 2, the standard software framework used in robotics, along with everything needed to simulate the vacuum inside Gazebo, a 3D physics simulator, before you ever touch real hardware. Inside, you will find a detailed 3D model of the vacuum's physical shape, roughly 349 millimeters across, including its body, its spinning LiDAR sensor turret used for mapping a room, two independently driven wheels, and a caster wheel. There is also configuration for combining odometry and inertial sensor data to estimate the robot's position, tuning files for SLAM, the technique robots use to build a map of a room while tracking their own location within it, and for Nav2, the navigation system that lets the robot plan and follow paths. A separate set of files wires the simulated version of the robot up in Gazebo, including a simulated LiDAR and bumper sensors on the front of the robot, with a linked document explaining some quirks in how the simulated bumpers behave. You can try the whole thing without owning a physical robot: after selecting this robot model in the shared launch configuration, a few ROS 2 launch commands start the simulated world, bring up navigation and mapping, and let you drive the simulated vacuum around with your keyboard. If you do have a real robot, similar commands connect to it over your local network instead, using a companion piece of software called SangamIO that needs to be set up on the vacuum first, and you can then save the map it builds for future use. The README is honest that the robot's physical measurements in this file are approximations meant to be refined against your own actual robot. The project is released under the Apache 2.0 license and has attracted eighteen stars on GitHub, reflecting a small but active hobbyist robotics community around DIY robot vacuums.
A ROS 2 robot description and Gazebo simulation package for OOMWOO One, an open-source robot vacuum you can build yourself.
Mainly Python. The stack also includes ROS 2, Python, Gazebo.
Apache 2.0 lets you use, modify, and distribute the code, including commercially, as long as you keep attribution and license notices.
Setup difficulty is rated hard, with roughly 1h+ to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Don't trust strangers blindly. Verify against the repo.