Brain-computer interfacing method for controlling a robotic arm device and the brain-computer interface device thereof

Brain-Computer Interface Technology for Robot Arm Control via Error-Related Potential Verification and Mode Re-determination

This technology is a brain-computer interface (BCI) method and device for controlling a robot arm that determines the control mode using BCI technology and verifies and re-determines the appropriateness of the control mode in real-time based on error-related potentials.

Existing BCI-based control technologies are limited to detecting simple motor imagery, making it difficult to efficiently control diverse robot movements and challenging to prevent errors when there is a mismatch between the user's intent and the robot's actual movement.

This technology proposes a method that determines and provides feedback on one of three control modes—reaching, grasping/releasing, or wrist rotation—based on abstract features extracted from the user's EEG, and re-determines the mode if the error-related potential exceeds a threshold. It can be used for assistive robots for patients with quadriplegia and in rehabilitation training, significantly improving control reliability by self-correcting mismatches between intent and action.

‍

‍

Key Features:

• Measuring the user's primary EEG to extract abstract features used for determining the control mode
• Determining one of the control modes—reaching, grasping/releasing, or wrist rotation—based on the abstract features
• Providing feedback on the determined control mode to the user and measuring the resulting error-related potential
• A configuration that corrects errors by re-determining the control mode if the measured error-related potential exceeds a threshold

‍

This invention was developed with support from the Ministry of Science and ICT under the project "Development of Non-invasive BCI Integrated Brain-Cognitive Computing SW Platform Technology for Controlling Real-life Devices and AR/VR Devices via Thought" (BCI-General/Sub-project 1) and "Development of BCI-based Brain-Cognitive Computing Technology for Recognizing Human Intent using Deep Learning" (BCI-Sub-project 2).