Iterative learning control method for tracking quasi-sinusoidal signals with slowly varying random frequency perturbations

Original iterative learning control (ILC) algorithm requires identical trajectories across iterations, limiting its adaptability. In industrial systems, trajectory frequency perturbations commonly arise due to environmental and operational changes. This study introduces a novel ILC algorithm designed for quasi-sinusoidal signal tracking, which is ubiquitous in industrial applications. First, the problem is formulated mathematically for a linear and time-invariant system. Then, the ILC algorithm for tracking slowly varying quasi-sinusoidal signals (ILCQS) is established using a methodology that combines perturbed target estimation with the iterative learning process. Subsequently, the algorithm’s convergence is theoretically analyzed. Finally, the ILCQS algorithm is simulated, and its performance is compared with that of the original iterative learning control (OILC) algorithm in terms of control error under the similar parameters. Results show that by applying ILCQS, the slowly varying frequency perturbation can be estimated, and a more precise tracking can be achieved. The control error of ILCQS is potentially smaller than that of OILC after adequate iterations.