Abstract
Functional electrical stimulation (FES) has demonstrated efficacy in enhancing comfort and mobility for individuals with paralysis, neurological diseases, and compromised muscle function. It facilitates motions that would otherwise be challenging or unattainable for these persons. Achieving optimal performance from FES systems necessitates the formulation of a strong and efficient control approach. To address this need, this paper proposes a novel controller called predictive proportional-integral-filtered derivative (PPID-F), which incorporates a predictive term into a filtered PID controller. The controller parameters are ideally optimized utilizing a recently devised metaheuristic method known as the artemisinin optimizer (AO). In contrast to traditional approaches in the literature, AO is employed to minimize a specifically designed objective function (OF) that integrates the integral of time-weighted absolute error (ITAE) criteria and the system's peak response (y(peak)). Comparative simulation results show that the proposed AO: PPID-F controller achieves the lowest objective function (OF) value among all evaluated methods. Specifically, it reduces the OF value from 0.4862 (AO: PID-F) to 0.2519, resulting in a 48.19% improvement. Compared to other benchmark controllers, the improvement ranges between approximately 48 and 68%. The effectiveness of the proposed controller is subsequently evaluated against six established approaches reported in the literature. The results demonstrate that the AO: PPID-F controller significantly enhances time-domain performance while preserving satisfactory frequency-domain features in comparison to alternative controller architectures developed for the FES system.