Adaptive learning control of nonlinear underwater robots: achieving repetitive tracking without parameterization

As marine exploration intensifies, precise repetitive tracking control for non-parameterized nonlinear underwater robots is crucial. This paper introduces an adaptive iterative learning control (ILC) approach based on difference-differential principles. By combining difference and differential links, control parameters are dynamically adjusted to achieve high-precision tracking and robustness against system uncertainties. To facilitate memory conservation, the controller features a compact structure requiring only two adaptive variables for parameter updates. Extensive numerical simulations covering various trajectories, parameters, and stochastic disturbances validate the method's effectiveness. Results demonstrate that the proposed controller significantly outperforms existing adaptive ILC schemes in tracking precision and convergence speed, confirming its efficacy for nonlinear underwater robots.