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 with flexible PCB and pressure matrix | OpenMuscle-FlexGrid |
| LASK5 | Handheld labeler wand for training ML models on gesture inputs | OpenMuscle-LASK5 |
| Band (OM-12) | Original OpenMuscle armband with 12 sensors (legacy design) | OpenMuscle-Band |
| Software/Firmware | Shared firmware for all devices, includes async scanning, model hooks, and examples | OpenMuscle-Software |
📚 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:
| Milestone | Description | ETA |
|---|---|---|
| FlexGrid v1.0 | Finalize PCB, BOM, and test data quality | ⬜ Q4 2025 |
| LASK5 v5.0 | Usability testing and power optimization | ⬜ Q4 2025 |
| Documentation Hub | Flesh out full guides and build flow | ⬜ Q4 2025 |
| Public ML Datasets | Upload labeled data sessions | ⬜ TBD |
We welcome pull requests, design files, firmware improvements, and documentation edits.
Start by reading the contribution guide:
📄 [CONTRIBUTING.md](./CONTRIBUTING.md)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)