Akademik Veri Yönetim Sistemi
13th IFAC Symposium on Nonlinear Control Systems, NOLCOS 2025, Reykjavik, İzlanda, 23 - 25 Temmuz 2025, cilt.59, ss.370-375, (Tam Metin Bildiri)
Symbiotic control integrates fixed-gain and adaptive learning architectures to mitigate the effect of system uncertainties. Specifically, fixed-gain control alone requires knowledge of uncertainty bounds to achieve predictable closed-loop system behavior, while adaptive learning alone functions without such bounds but may introduce unpredictability in closed-loop system performance. To this end, symbiotic control harnesses the predictability of fixed-gain control and the adaptability of learning mechanisms to create a framework that does not rely on knowledge of system uncertainties. This paper studies symbiotic control of dynamical systems with nonparametric system uncertainties. First, we robustify the nonparametric symbiotic control framework by introducing a low-pass filter to its fixed-gain control architecture. Second, using grid-based optimization, we compute optimal parameters for the proposed fixed-gain architecture that maximizes the delay margin and minimizes the H∞ norm. In particular, when adaptive learning is of, the delay margin is calculated from the loop transfer function and the H∞ norm is derived from a transfer function with uncertainty as input and user-defined states as output. Then, the adaptive learning mechanism is added to the control loop with low gains to improve the closed-loop system performance gradually. Finally, two illustrative numerical examples are also given to demonstrate the efficacy of the proposed nonparametric symbiotic control framework.