This system uses Steady-State Visual Evoked Potential (SSVEP) as its core technology, combined with the self-developed NM-01 EEG signal acquisition module and a digital twin track simulation system, to create a high-precision, high-real-time BCI car solution. The hardware layer relies on the NM-01 module, which adopts a 24-bit high-precision ADC and high input impedance design (compatible with dry electrodes or fabric electrodes), capable of stably capturing weak EEG signals and providing reliable data for accurate SSVEP decoding. Similar research has achieved an average recognition accuracy of 92.50% in direction and graded speed control via SSVEP.

The software layer innovatively introduces a digital twin track simulation system, which constructs a virtual simulation scenario based on real competition environment parameters (e.g., track width, obstacle layout, etc.), and provides real-time feedback on BCI command recognition results and car motion trajectories, supporting users in conducting intensive training and strategic optimization before competitions. The simulation system can precisely replicate key variables in both laboratory and real-world scenarios, helping users adapt to complex environments and predict competition results in advance. In addition, the system supports secondary development interfaces, allowing developers to expand personalized applications (e.g., multimodal signal fusion control or emergency braking intent detection) based on raw EEG signals or decoded control commands.