A brain-actuated wheelchair: asynchronous and non-invasive Brain-computer interfaces for continuous control of robots

Area of Science:

• Neuroscience
• Robotics
• Human-Computer Interaction

Background:

• Brain-Computer Interfaces (BCIs) offer potential for assistive technologies.
• Non-invasive BCIs using electroencephalography (EEG) face challenges with robustness and long-term usability.
• Continuous control of complex devices like wheelchairs requires a stable and intuitive BCI system.

Purpose of the Study:

• To evaluate the feasibility and robustness of an asynchronous, non-invasive EEG-based BCI for continuous mental control of a wheelchair.
• To assess the system's performance over time and in varied environmental contexts.
• To determine if users can autonomously operate the BCI without external adaptive algorithm tuning.

Main Methods:

• Two subjects controlled a simulated wheelchair using an asynchronous EEG-based BCI.
• Experiments involved driving along pre-specified and novel complex paths.
• System robustness was tested across multiple sessions with variable time intervals and path segments.

Main Results:

• Subjects achieved high success rates in reaching wheelchair control goals (up to 100%).
• Performance varied with time and context, indicating time and context dependency.
• Subject 1 successfully reached goals in 80% of trials with novel, complex paths.

Conclusions:

• Asynchronous EEG-based BCIs are feasible for rapid user mastery and continuous control of wheelchairs.
• The BCI system demonstrated autonomous operation over extended periods without external tuning.
• Key factors for success include shared control with the wheelchair and stable, user-specific EEG feature selection.