- Carl Larsson
- Pontus Svensson
- Viktor Eriksson
This project is part of the ELA411 course, focusing on the development of a Brain-Machine Interface (BMI) system. Our group is tasked with creating a standalone BMI system.
The main objective is to develop a system capable of recording and acquiring neurophysiological data online using EEG and EMG electrodes. This data will then be decoded to extract navigation commands for controlling a robot in a simulated environment.
- Equipment: Ganglion board with EEG gold-cup and EMG/EKG snap electrodes.
- Functionality: Using EMG and EEG data to drive and navigate a robot in a simulated environment.
- Control Commands: Decoding of EMG data for directional control (right vs. left) and EEG data for movement initiation (drive vs. don’t drive).
- Setup: Ganglion board setup is used for recording EEG and EMG activity.
- Data Streaming: A communication interface is developed for real-time streaming of EMG and EEG signals with associated timestamps to a computer.
- Electrode Placement: Strategic placement of EMG and EEG electrodes is critical for accurate navigation variable decoding.
- Methodology: Based on a scientific literature study, we will develop methods for decoding navigation commands from neurophysiological data.
- Signal Processing: The acquired signals will undergo noise and artifact reduction processes, followed by feature extraction.
- AI Implementation: AI methods will be employed to decode the navigation variables from the processed data.
- Simulation Environment: ROS2 controlled Turtlebot3 robot in Gazebo.
- Demonstration Scenario: A repeatable scenario demonstrating the system's functionality across various users.
- The system will undergo verification and validation in accordance with the course instructions.
This project is subject to continuous updates and improvements based on project progress and feedback from our instructors.