A Facial Gesture-Controlled Wheelchair for Individuals with Complete Disabilities

This paper presents a novel face-controlled robotic wheelchair system that leverages computer vision and facial landmark detection to enable intuitive, real-time control of motorized movements. The system integrates OpenCV and dlib for detecting facial features such as the nose and mouth, while a Tkinter-based graphical user interface (GUI) provides seamless interaction for users. Key contributions include the ability to control the system using facial gestures captured by a camera, effective handling of multiple face scenarios by prioritizing the largest detected face, and dynamic real-time speed control via a GUI slider. The integration of an Arduino microcontroller with an L298N motor driver ensures precise control of motors, enabling smooth forward, backward, left, and right movements based on the position of the nose and the state of the mouth. Experimental results demonstrate high responsiveness and accuracy in translating facial gestures into motor commands. Despite its advantages, the system has limitations, such as the lack of obstacle detection, no waterproofing, and the absence of wireless control. Future work will focus on addressing these limitations to enhance the system’s robustness and applicability in real-world scenarios.