A Novel Adaptive Iterative Learning Control Approach for Nonlinear Output-Constrained Systems With Input Quantization

In practice, the communication bandwidth and physical limitations are the two main categories of threats that the networked control systems may encounter. Therefore, this paper focuses on adaptive iterative learning control (ILC) of nonlinear strict-feedback systems with input quantization and asymmetric output constraint. Through constructing an error transformation mechanism, the original output-constrained control system is converted into the unconstrained form. Subsequently, a novel adaptive ILC algorithm is established by virtue of the command filtered backstepping technique, in which the uncertain terms of Lyapunov function (LF) are decomposed into the parameter compensation components of the controller and the iteration-convergent lumps via the hyperbolic tangent function. Specially, to accommodate the input-related uncertainties brought by quantizer, the devised ILC law adapts a nested structure, thus achieving the estimation of unknown parameters associated with the quantized bias. The convergence of error along the iteration axis is rigorously proven by the composite energy function (CEF). Finally, the proposed approach is applied to two examples, the results of which illustrate the effectiveness of the scheme.