Iterative learning control for nonlinear random impulsive systems with application to soft robotic actuators

Abstract

Motivated by recent developments in random dynamical systems and iterative learning  control (ILC), this paper investigates a $PID$-type ILC scheme for a class of nonlinear random impulsive differential systems arising in soft robotic actuator dynamics. The proposed  model incorporates  both randomness and impulsive effects, leading to piecewise continuous system trajectories. By employing a weighted Banach space concepts and imposing suitable Lipschitz and boundedness conditions,  sufficient criteria are established to ensure the well-posedness of the system and convergence of the learning process. In particular, the actuator output is shown  to converge to the desired reference trajectory in the mean-square sense. The initial tracking error is estimated using the mean-value theorem. A numerical example involving a soft robotic actuator system is presented to illustrate the applicability and effectiveness of the proposed control strategy.