Data-driven Adaptive Iterative Learning Control for Stochastic Nonlinear Systems

This paper studies the data-driven adaptive iterative learning control (DDAILC) problem for nonlinear stochastic systems. First, the DDAILC framework for nonlinear stochastic systems is established with external random disturbances. Then, to resolve the contradiction between the control accuracy and convergence speed, an event-triggered mechanism is constructed to generate a multistage adaptive learning step sequence, where the sequence retains invariant at the same stage to improve the convergence speed and decreases between two adjacent stages to realize high-precision tracking. Finally, a numerical simulation is provided to demonstrate the validity of the devised event-triggered DDAILC algorithm.