Development of BCI-Driven Robotic Hand Control System: A Preliminary Study

The sophisticated Brain-Computer Interface (BCI) utilizes discrete or model-based control techniques to manage external devices. A continuous control strategy is vital for achieving optimal and smooth operation. Achieving a continuous estimation of control parameters with minimal latency is crucial for enhancing the effectiveness and acceptance of mind-controlled prostheses, exoskeletons, and robotic arms by users. This research proposes a novel BCI model for controlling a robotic hand, incorporating Task Classification (TC) and Trajectory Estimation (TE) modules. The TC module interprets user intentions based on EEG signals, identifying tasks such as initiating, grasping, releasing, halting, or maintaining the current state. Upon detecting grasping (Task 2), the TE module estimates the 3D trajectory of hand movements, enabling precise control of the robotic hand. The control algorithm translates the estimated trajectory into commands for the robotic hand, facilitating object manipulation and task execution. Real-time feedback loops and error-handling mechanisms enhance user interaction and system robustness. Results demonstrate the efficacy of the integrated TC and TE modules in providing seamless control of the robotic hand, empowering individuals with motor disabilities to perform daily tasks independently. The implications of this work extend to the development of advanced BCIs, facilitating more natural and seamless communication between the human brain and technological devices.