Skip to content

CleverHand is an open-source, modular Human-Machine Interface (HMI) system designed for the acquisition of biosignals (EMG, ECG, EEG, etc.) and the delivery of sensory feedback (electrotactile, vibrotactile, visual, etc.).

License

Unknown and 2 other licenses found

Licenses found

Unknown
LICENSE
Unknown
LICENSE-DOCS
Unknown
LICENSE-HARDWARE
Notifications You must be signed in to change notification settings

Aightech/CleverHand

Repository files navigation

CleverHand

CleverHand is an open-source, modular hardware–software platform for Human–Machine Interfaces (HMI), designed to support high-quality biosignal acquisition and multimodal sensory feedback in research, education, and advanced prototyping.

The platform enables the acquisition of physiological signals (EMG, ECG, EEG, IMU, etc.) and the delivery of feedback (electrotactile, vibrotactile, visual), with a strong emphasis on modularity, reproducibility, and data quality.

CleverHand is not a closed product. It is an open infrastructure intended to be:

  • built and modified by researchers,
  • integrated into experimental systems,
  • and reused across multiple projects without redesigning the stack each time.

Diagram

CleverHand system diagram


Project philosophy

  • Open by design All hardware, firmware, and software are open-source. Users may self-manufacture modules or assemble them via standard PCBA services.

  • Research-first, not consumer-first CleverHand prioritises signal integrity, timing control, and experimental flexibility over plug-and-play simplicity.

  • Access without lock-in For users who do not want to manufacture hardware themselves, pre-assembled and tested modules may be offered as an access layer (shop), without restricting openness.

  • Long-term maintainability Stable interfaces, documented modules, and explicit quality-control practices are preferred over rapid feature churn.


System overview

CleverHand devices are built by chaining small, specialised modules that share:

  • a common electrical and communication interface,
  • a unified firmware and data model,
  • and a host-side software stack (GUI, LSL, Python, C++).

Repository structure

  • hardware Hardware design files (KiCad), manufacturing data, and module-level documentation.
  • software
    • driver Host-side api and libraries (Drivers, LSL streaming, Python/C++ APIs).
    • firmware Embedded firmware for inter-module communication and host interfacing.
    • tools Utilities for configuration, testing, and data visualisation:

Key features

Modular architecture

  • 1 to 32 modules per device
  • Sensor, feedback, communication, and interface modules can be freely combined

Position-based addressing

  • Module addresses are inferred from physical position in the chain
  • No hardcoded IDs or manual configuration

High-quality biosignal acquisition

  • Up to 4 kHz sampling rate
  • 24-bit resolution (ADS1293 / ADS1298 based modules)
  • Designed for EMG and other low-amplitude physiological signals

High-density wearable design

  • Typical module size: 20 × 20 mm
  • Integrated gold-plated electrode pads (e.g. 3 × 18 mm)
  • Supports dense spatial sampling

Flexible connectivity

  • Wired: USB, Ethernet
  • Wireless: Bluetooth, Wi-Fi (others possible depending on controller)

Unified software interface

  • Configuration GUI
  • Real-time visualisation
  • LSL streaming
  • Python and C++ APIs

Background and motivation

Details

Motivation

Electromyography (EMG) provides a non-invasive way to measure muscle activation and is widely used in:

  • human–machine interaction,
  • prosthetics and assistive devices,
  • rehabilitation and exoskeleton control,
  • sports science and biomechanics.

Despite its importance, existing EMG systems often fall into two extremes:

  • low-cost, low-performance wearable devices,
  • or high-performance, high-cost laboratory systems with limited portability.

Limitations of existing systems

Wearable EMG devices

Device Price Electrodes Sampling rate Resolution Notes
Myo armband ~200 $ 8 bipolar 200 Hz 8-bit Easy to use, discontinued
MyoWare ~40 $ 3 bipolar 200 Hz 10-bit Suitable for prototyping only
Delsys Trigno Avanti ~2000 $ 16 bipolar 2000 Hz 16-bit High quality, expensive

Non-wearable EMG systems

Device Price Channels Sampling rate Resolution
OT Sessanta Quattro ~10 k$ 64 2000 Hz 24-bit
OT Muovi+ ~25 k$ 4×32 2000 Hz 24-bit
OT Quattrocento ~50 k$ 382 2000 Hz 24-bit

CleverHand approach

CleverHand aims to bridge the gap between these categories by providing:

  • modular, wearable, high-density EMG acquisition,
  • research-grade signal quality,
  • and a system that scales from small experiments to complex multi-site recordings.

Each EMG module integrates an ADS1293 or ADS1298 AFE, supporting 3–8 channels per module, sampled at 4 kHz / 24-bit. Multiple modules can be chained to increase spatial coverage.

Gold-plated electrode pads on the PCB can be used directly or adapted via interface boards (flex electrodes, snap connectors, jack connectors).


Additional capabilities

Feature Description
Visual feedback RGB LEDs per module
IMU Optional inertial sensing modules
Electrotactile Optional stimulation modules
Vibrotactile Optional actuator modules
Attachments Bracelet, mesh, custom mounts
Interfaces GUI, LSL, Python, C++

Project status & contributions

CleverHand is an active research platform. Interfaces and core architecture are stabilising, while individual modules continue to evolve.

Contributions are welcome in the form of:

  • issues and bug reports,
  • hardware or firmware improvements,
  • documentation and experiments,
  • or external use cases and feedback.

If you plan to use CleverHand in a study or project, citations and references are appreciated.


If you want next, I can:

  • add a “Who should / should not use CleverHand” section,
  • split this into a short README + extended docs index,
  • or rewrite it again with a stronger academic / grant-proposal tone.

About

CleverHand is an open-source, modular Human-Machine Interface (HMI) system designed for the acquisition of biosignals (EMG, ECG, EEG, etc.) and the delivery of sensory feedback (electrotactile, vibrotactile, visual, etc.).

Resources

License

Unknown and 2 other licenses found

Licenses found

Unknown
LICENSE
Unknown
LICENSE-DOCS
Unknown
LICENSE-HARDWARE

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published