Adaptive Weight Fuzzy Iterative Learning Control Closed-Loop FES Control Algorithm for Enhancing Rehabilitation Motion Completion

This study presents a closed-loop feedback functional electrical stimulation (FES) control algorithm, employing adaptive weight fuzzy iterative learning control (AW-FILC), aimed at enhancing the completion rate of rehabilitation training movements for paralyzed limbs of stroke patients. By integrating the advantages of fuzzy logic reasoning and iterative learning control, along with an adaptive weight mechanism and penalty factors, the AW-FILC algorithm dynamically optimizes control gains, thereby significantly boosting the system's robustness and real-time responsiveness. Simulation results demonstrated the algorithm's rapid convergence and low root mean square error (RMSE) across various trajectories, such as triangular waves, step functions, and actual pronation-wrist flexion motion curves, suggesting its real-time performance capabilities. Further validation through human experiments, where healthy subjects simulated wrist extension motor dysfunction, showed that the algorithm enabled rapid response and stable convergence, maintaining a high correlation between actual and target motion trajectories, indicating a high degree of movement completion. Collectively, these findings confirm the AW-FILC algorithm's precise control capabilities and its promising potential for clinical application in stroke rehabilitation tasks.