Motor function disorders profoundly affect physical mobility and emotional well-being, often resulting in dependence on assistive care. Since eye movements remain largely unaffected, they play a vital role in enabling communication and control. Electrooculography (EOG), which records eye movements within the 0.5-40 Hz range, has significant potential for enhancing human-computer interaction (HCI). This study aimed to develop a real-time, low-cost EOG-based assistive device to support individuals with motor impairments, including stroke survivors and amputees. A two-channel EOG-based assistive device was designed and implemented. EOG signals corresponding to specific eye movements were recorded from both able-bodied and disabled participants. The system incorporated a designed amplifier and an ESP32 microcontroller for on-board signal processing. Preprocessing steps included powerline noise removal, smoothing, and mean subtraction. A lightweight, subject-specific threshold-based classifier was employed to achieve efficient and accurate classification suitable for embedded deployment. The proposed system achieved an average accuracy of 100% for healthy participants and 99.2% for disabled participants. The mean response times (RTs) were 2.03 and 2.04 s, while the corresponding information transfer rates (ITRs) were 136.04 and 132.89 bits/min. User feedback highlighted the system's ease of use, safety, and comfort. Classified eye movements and blinks were successfully mapped to wheelchair navigation commands, enabling intuitive mobility control. This portable, real-time, EOG-based system provides a practical and effective mobility solution for individuals with severe motor disabilities. Its low-cost and reliable design makes it especially suitable for deployment in low-resource environments, offering a pathway towards inclusive and accessible assistive technologies.HighlightsDeveloped a two-channel real-time electrooculography (EOG) mobility-assistive device for individuals with motor function disorders, utilising a custom amplifier and ESP32 microcontroller for signal processing and wheelchair control.Classified eye movements and blinks for wheelchair navigation, ensuring ease of use, safety, and comfort, making it suitable for low-resource settings.Achieved 100% accuracy for healthy subjects and 99.2% for disabled subjects, with response times of 2.03s and 2.04s, respectively.
S et al. (2026) studied this question.