基于脑机接口的机械手控制教学实验设计

[Objective] This study aims to innovate the teaching methods of biomedical engineering by introducing a brain-computer interface (BCI)-based robotic arm control experiment into the course “Integrated Design Experiment in Biomedical Engineering”. The experiment focuses on integrating EEG signal acquisition, processing and robotic arm control to systematically cultivate students' innovative, practical and problem-solving abilities. In view of the rapid development of brain-computer interfaces in the field of medical rehabilitation, this experiment focuses on bridging the gap between theoretical teaching and practical application, and provides students with a practical platform for BCI and artificial intelligence technology. [Methods] The experiment is built around four key components: a visual stimulus interface, a data acquisition system, a data analysis system and a robotic arm control system. The non-invasive BCI method of steady-state visual evoked potentials (SSVEP) is chosen to control the robotic arm. EEG signals are collected via a 16-channel EEG system with electrodes placed on the visual cortex area of the brain to ensure optimal signal acquisition. Visual stimuli are presented on the screen through sinusoidal modulation at frequencies of 6.7 Hz, 10 Hz, 12 Hz and 15 Hz to elicit corresponding SSVEP responses. After the data was collected, preprocessing steps such as filtering and downsampling were applied, and then a canonical correlation analysis (CCA) was used to classify the EEG signal based on its correlation with the predefined stimulus frequencies. These classifications were then used to generate control commands for the robotic arm to perform specific tasks, such as writing numbers. Tailored to suit students of different levels of expertise according to the adaptive teaching approach, the experiment encouraged capable students to engage in system design beyond basic BCI development, foster creativity and satisfy the students' willingness to engage more deeply, while the project also allowed students to modify and upgrade the experiment, improving their problem-solving and innovation skills. [Results] 30 undergraduate biomedical engineering students successfully controlled the robotic arm through the experimental process, using an SSVEP-based BCI system to write the numbers “1”, “2”, “3” and “4”. It was found that the classification accuracy of the SSVEP response was highest when using the third harmonic, and a balance was achieved between classification accuracy and computational efficiency. The students also tried different visual stimulation parameters, such as adjusting the frequency and position, to evaluate their impact on system performance. Some students also explored alternative communication options between the BCI and the robotic arm, such as Bluetooth, to overcome the limitations of the USB connection, further expanding the versatility of the system. These findings highlighted the importance of signal quality and classification algorithms in achieving reliable robotic arm control, and through experimentation students recognized the wider potential applications of BCI systems in medical rehabilitation, gaming and assistive technology. [Conclusions] The BCI-based robotic arm control experiment successfully demonstrated the potential for integrating advanced technologies such as brain-computer interfaces into practical biomedical engineering applications. The experiment not only improved the students' technical skills in EEG signal processing, data analysis and robotic control, but also cultivated critical thinking and technology transfer capabilities through open innovation tasks. This research confirms that an issue-based learning approach and multidisciplinary integration experiments in BCI and robotics can significantly improve students' practical skills and understanding of complex biomedical engineering systems. This experiment provides an important reference for the optimization of EEG-BCI applications and effectively improves students' comprehensive ability to participate in the development of cutting-edge technologies.