Welcome to the central repository for all OpenMuscle devices, software, and documentation.
OpenMuscle is a modular, open-source prosthetic sensing platform using pressure and muscle activity to drive machine learning–based control.
This repository serves as the starting point for users, contributors, and researchers exploring the OpenMuscle ecosystem.
OpenMuscle enables wearable biosensing devices using custom sensors, open firmware, and open hardware. The system is designed for researchers, developers, and prosthetics hackers interested in:
- Real-time muscle signal acquisition
- Machine learning model training (classification/regression)
- Experimental prosthetic and robotic control
- 3D-printable enclosures and modular design
| Project | Description | Repository |
|---|---|---|
| FlexGrid | 60-sensor wearable: flexible sensor PCB plus rigid controller PCB | OpenMuscle-FlexGrid |
| FlexGrid Firmware | MicroPython firmware for the FlexGrid controller (ESP32-S3) | FlexGridV3-Firmware |
| LASK5 | Handheld labeler wand for capturing ground-truth gesture labels | OpenMuscle-LASK5 |
| Software | PC application: data capture, ML training and inference, web UI | OpenMuscle-Software |
| AR / XR | WebXR client and the future native headset app | OpenMuscle-AR |
| Library | Shared KiCad symbols, footprints, and 3D/CAD parts | OpenMuscle-Library |
| Band (OM-12) | Original 12-sensor armband (legacy design) | OpenMuscle-Band |
📚 Full build guides, videos, and tutorials will be hosted here or at https://openmuscle.org.
This repository will grow into a full documentation hub with instructions for:
- Building each device
- Flashing and configuring firmware
- Training ML models with labeled data
- Mechanical assembly (with 3D-printable STL files)
- Example use cases and integration ideas
📌 Check each device repo for its specific README and wiring guide.
🧪 Want to build a device or contribute?
- Choose a device (e.g. FlexGrid or LASK5) and visit its repo.
- Follow the hardware build guide in the
/Docsfolder. - Flash the firmware from OpenMuscle-Software.
- [Optional] Integrate the data into your own ML workflow.
🧠 OpenMuscle uses a modular architecture:
- Devices like FlexGrid and LASK5 communicate using MicroPython firmware.
- Sensors send signals over ESP-NOW or Wi-Fi protocols.
- A Python or serial-based pipeline can log, visualize, or process signals.
- Data can be used in supervised ML pipelines for gesture classification or finger prediction.
🔧 A full system diagram will be added here soon.
📷 Placeholder for an image/diagram:
The full roadmap, including near-term milestones, mid-term ecosystem plans, the ten-year vision, and the disclosed business structure, lives in ROADMAP.md.
Quick summary of where we are (June 2026):
- Now shipping: FlexGrid V4 dev kits (10 units, first 3 deployed)
- In progress: OpenMuscle-Hub housekeeping, LASK5 repo cleanup, public training dataset, VR/XR client extraction
- Planned: Nonprofit + for-profit incorporation, native VR app, mobile/Android app, V5 hardware with bridge PCB
See ROADMAP.md for the full picture.
We welcome pull requests, design files, firmware improvements, and documentation edits.
Start by reading the org-wide contribution guide:
📄 CONTRIBUTING.md
File issues on the repo that owns the work. Not sure which one? Open an issue here on the Hub and a maintainer will route it. You can also join upcoming GitHub Discussions to suggest ideas or improvements.
✅ Beginners welcome! Look for "good first issue" tags in the Software repo.
- GitHub Discussions
- Discord Server (invite link TBD)
- Monthly Community Builds & AMAs
Want to feature your build or fork? Submit to the Show & Tell section in Discussions!
All documentation and text in this repo is licensed under the Creative Commons Attribution-ShareAlike 4.0 International License.
Any example code snippets are licensed under the MIT License.
📄 See [LICENSE.md](./LICENSE.md) for full details.OpenMuscle is developed by @TURFPTAx and the broader open-source community.
This project is part of an ongoing effort to make prosthetics more affordable, transparent, and hackable.
🌐 More at [openmuscle.org](https://openmuscle.org)