Comparative Bioinspired Optimization for Delay-Compensated SMC in Dominant Dead-Time Systems

This work aims to develop and evaluate a tuning methodology for a Smith Predictor-based Sliding Mode Controller (SMC) optimized with five bioinspired algorithms to improve robustness and performance in time-delay systems under uncertainty. The central hypothesis is that combining the predictive capabilities of the Smith Predictor with the robustness of SMC and the optimization power of bioinspired algorithms can produce a controller that maintains high performance. Experimental tests on the TCLab platform, which add significant delays, confirm this. The results show that the proposed strategy effectively manages reference changes, outperforming traditional tuning methods. The main contributions include integrating intelligent optimization with robust predictive control and validating the methodology in a real-time delayed process.