Abstract

Hand-free control of assistive mobility devices has been developed for severe physical movement disabilities. This paper proposes a human-machine interface (HMI) using piezoelectric sensors to translate face movements for wheelchair control commands. Piezoelectric ceramic buzzer and thin film piezoelectric sensors were observed for facial muscle signal acquisition. We used time and frequency analysis to extract features of the signals during winking and tongue movements for simple classification. We also verified the proposed HMI for control commands of wheelchair steering. Using ceramic buzzer piezoelectric sensors with maximum peak and frequency mean features achieved more than 95% of the average classification accuracy. The proposed system can be used in a powered wheelchair for quadriplegic patients. We will further implement and verify in real-time control.