Design and fNIRS based control of 2-DOF exo-ankle for gait rehabilitation

This study presents the design and control of an ankle-assistive exoskeleton utilizing functional near-infrared spectroscopy (fNIRS). Ankle-related impairments are a major cause of mobility hindrance, often resulting in gait imbalances and abnormal walking patterns. The proposed exoskeleton is designed to provide two degrees of freedom (2-DOF), enabling active dorsiflexion and plantarflexion movements, along with passive eversion and inversion. The mechanical structure incorporates linear actuators and a ball-and-socket joint, which are controlled through brain signals. Furthermore, this study develops and implements an innovative fNIRS-based control system capable of detecting hemodynamic changes in the brain associated with ankle movements. After pre-processing the fNIRS signals, a machine learning model is trained to accurately predict the intended ankle movements. The results demonstrate that fNIRS-based control for ankle-assistive exoskeletons holds significant potential for future applications in gait rehabilitation.