Current Volume 9
This paper presents an eye-controlled wheelchair system designed to restore independent mobility for individuals with severe physical disabilities such as quadriplegia, who are unable to operate traditional joystick-controlled wheelchairs. The proposed system integrates a camera module for real-time eye image capture, followed by image preprocessing, eye detection, and eye tracking modules to isolate and monitor gaze direction. A Convolutional Neural Network (CNN) classifies the user's gaze into three discrete commands—left, right, and center—which are interpreted by the command module to control wheelchair movement accordingly. These commands are transmitted via an Arduino communication module to a motor driver that actuates high-torque DC motors, enabling forward, backward, and stop functions. The system incorporates robust safety mechanisms, including an automatic stop when eyes are centered or closed, ensuring reliable operation in real-world environments. Experimental results demonstrate that the CNN-based classification achieves high accuracy under varying lighting conditions and with minor head movements, providing an intuitive, hands-free, and low-cost assistive solution that significantly enhances user independence and quality of life.
Eye-Controlled Wheelchair, Convolutional Neural Network (CNN), Gaze Classification, Assistive Technology, Real-Time Image Processing, Hands-Free Navigation
IRE Journals:
Karthikeyan P, Siva S, Pandikannan A, Dr. M. Iiayaraja "Smart AI Based Eye Controlled Wheelchair" Iconic Research And Engineering Journals Volume 9 Issue 11 2026 Page 1895-1902 https://doi.org/10.64388/IREV9I11-1717823
IEEE:
Karthikeyan P, Siva S, Pandikannan A, Dr. M. Iiayaraja
"Smart AI Based Eye Controlled Wheelchair" Iconic Research And Engineering Journals, 9(11) https://doi.org/10.64388/IREV9I11-1717823