Our Projects - EEG Controlled Hand-Exoskeleton

Background

This project demonstrates a prototype brain-computer interface (BCI) system that enables a robotic hand exoskeleton to perform finger movements using interpreted brain signals. The system combines EEG signal processing, machine learning, robotics, and embedded systems to create a low-cost assistive technology designed to support rehabilitation for patients with hand paralysis.

Key Features

• Brain-computer interface using EEG signals
• Machine learning classification of motor imagery
• Tendon-inspired robotic hand exoskeleton
• Arduino-controlled actuation system
• Interactive graphical control interface

The aim of this project was to develop a functional prototype robotic hand exoskeleton controlled by brain signals.

The system translates EEG signals into commands that trigger finger movements through a robotic mechanism.

The goal was to demonstrate a proof-of-concept neurorehabilitation device that is affordable and scalable.

Results

Testing across 10 trials showed:

• Finger movement accuracy: ~65%
• Servo actuation time: ~2.5 seconds per action
• EEG classification runtime: ~5 minutes per dataset

The system successfully demonstrated proof-of-concept brain-controlled robotic movement, though improvements are needed for real-time operation and increased reliability.

Impact

This project demonstrates the potential of combining machine learning, neuroscience, and robotics to develop assistive technologies for rehabilitation. The prototype provides a foundation for future commercial neurorehabilitation devices.