Sampled-Data Adaptive Iterative Learning Control for Uncertain Nonlinear Systems
Hui Yu, Deyuan Meng, Ronghu Chi, Kaiquan Cai
Abstract
In the realm of data-driven adaptive iterative learning control (AILC), the emphasis in designing and analyzing control schemes mainly concentrates on discrete-time systems, while fewer results are developed for the more common continuous-time plants. To overcome this limitation, a practical sampled-data AILC (SDAILC) is developed for continuous-time nonaffine nonlinear plants. A sampled-data iterative dynamic linearization (SDIDL) method is devised to build the dynamic connection between input and output (I/O) data throughout different iterations. On this basis, the SDAILC method, including a sampled-data parameter estimation algorithm and a learning control law, is proposed by utilizing optimization-based design. In SDAILC, the sampling period is treated as a parameter to compensate for its influence on the control performance, and an error feedback is naturally involved, improving the robustness against uncertainties and the closed-loop stability of the plant. Notably, SDAILC is a data-driven approach independent of model information. The validity of SDAILC is proved mathematically and demonstrated by simulations.