oomwoo
Open-source DIY robot vacuum: 3D-printed, ROS2 + 2D LiDAR, fully local.
License
Apache-2.0
Overview
oomwoo is an open-source home robot vacuum you build yourself, made for the Raspberry Pi, ROS2, Home Assistant and 3D-printing communities. It uses an affordable 2D LiDAR to map your home and navigate autonomously, runs fully local with no cloud required, and avoids vendor lock-in. Hardware, software and firmware are all open, designed to be built from parts you source yourself (an optional kit is available, never required).
oomwoo is a round, 3D-printed, ROS2-based robot vacuum built community-first, module by module, that doubles as an affordable ROS2 learning platform. Compute is a Raspberry Pi 5 (4 GB) running ROS2 + LiDAR SLAM + Nav2; a custom JLCPCB I/O board carries an MCU running micro-ROS for motor drivers and sensor front-ends. Sensing centers on a 2D LiDAR turret, with bumper, IR cliff, multizone ToF, IMU and an optional color camera. The design is simulation-first (Gazebo + URDF) so contributors without hardware can still build. The project is at the design/RFC stage: a first working bill of materials is targeted for mid-July 2026 and a fully-costed BoM for end of August 2026.
Why it matters
Commercial robot vacuums are closed, cloud-dependent and locked to one vendor. oomwoo offers a fully open, locally-controlled alternative that the Raspberry Pi and ROS2 communities can build, repair and extend themselves. It also serves as a practical, affordable platform to learn real autonomous navigation.
Problem it solves
It removes vendor lock-in and mandatory cloud connectivity from home robot vacuums, giving builders full control over a local, hackable robot. By specifying common, abundant wear-part sizes and sourceable mechanisms, it makes a capable vacuum buildable and repairable from parts you can find anywhere.
Key takeaways
- Open hardware, software and firmware (Apache-2.0 code) for a DIY robot vacuum
- Raspberry Pi 5 + ROS2/Nav2 + 2D LiDAR for autonomous mapping and navigation
- Cloud-free, local-first operation; Home Assistant integration as a goal
- Print the geometry, source the mechanisms and wear items
- Built in the open via modular RFCs; simulation-first in Gazebo
Who it is for
- ROS2 developers
- Raspberry Pi makers
- 3D-printing hobbyists
- Home Assistant tinkerers
- Robotics learners
Limitations
- Early design/RFC stage — first working BoM targeted ~mid-July 2026, fully-costed BoM end of August 2026
- BoM is an explicit work-in-progress draft with price ranges, not a final validated parts list
- No step-by-step build instructions yet; printable STL files are not published
- Advanced obstacle avoidance (camera + AI) is a later/experimental goal