Humanoid Robot Walking in Maze Controlled by SSVEP-BCI Based on Augmented Reality Stimulus

Area of Science:

• Robotics
• Neuroscience
• Human-Computer Interaction

Background:

• Brain-computer interfaces (BCI) offer potential for robot control but face challenges with stimulator portability.
• Augmented reality (AR) presents an opportunity to overcome these limitations by serving as a visual stimulator.

Purpose of the Study:

• To investigate the feasibility of an AR-based steady-state visual evoked potential (SSVEP) BCI system for robot control.
• To develop a portable BCI system for practical robot navigation tasks.

Main Methods:

• An AR system was utilized as the visual stimulator for an SSVEP-BCI.
• A robot maze-walking experiment was conducted using the AR-BCI system.
• Human intentions were decoded using the Filter Bank Canonical Correlation Analysis (FBCCA) algorithm to control robot movement.

Main Results:

• All 12 subjects successfully completed the robot maze-walking task using the AR-BCI system.
• The system demonstrated the feasibility of controlling a robot through a maze via BCI.
• The AR-SSVEP-NAO system proved effective for online robot control.

Conclusions:

• The developed AR-BCI system is feasible for robot control applications.
• This study offers a novel approach for portable BCI systems.
• The findings highlight the potential of AR-BCI for advancing robot technology and BCI practicality.