REAL-TIME EEG-BASED WHEELCHAIR CONTROL SYSTEM FOR ENHANCING MOBILITY

This thesis proposes a real-time wheelchair-driving system using electroencephalography (EEG) signals via a non-invasive Brain-Computer Interface (BCI). Designed for individuals with mobility impairments, the system interprets brain activity to execute commands like “start,” “stop,” “turn left,” and “turn right.” EEG signals are processed in MATLAB and classified using a Support Vector Machine (SVM) for accurate command recognition. The commands are transmitted via UDP to an Arduino controller, which directs motor actions. This integration of EEG processing, machine learning, and embedded control offers a practical assistive solution, enhancing user autonomy through thought-driven mobility.