Learning is a unique behaviour characteristic of human beings. By imitating this learning ability, iterative learning control technology can achieve high-precision trajectory tracking tasks with repetitive learning and correcting. Recently, iterative learning control technology has become a hot topic in the fields of intelligent control, due to its wide application value in chemical batch processes, train operation control and medical rehabilitation, etc.Industrial controlled systems usually exhibit model uncertainty, external disturbance,nonlinearity and system fault, as which traditional iterative learning control methods may have some limitations. Thus, this thesis focuses on the research on robust iterative learning control for discrete linear systems and parameterized nonlinear systems. The convergent performance of the corresponding model is also studied and several simulation examples are considered to demonstrate the efficacy of the proposed method. The major work of this thesis is summarized as follows:(1) Robust iterative learning control of a uncertain multiple-input multiple-output discrete system with monotonic convergence is studied. The one-dimensional error state space model in the iteration domain is derived by using two-dimensional model of the considered iterative learning process. Monotonically convergent existence conditions of tracking errors and explicit formula for determining learning gain matrices are obtained by employing bounded real lemma of discrete systems and linear matrix inequality technology. Finally, some numerical simulations are given to verify the obtained theoretical results.(2) Robust finite frequency range iterative learning control of a multiple-input multipleoutput discrete system with monotonic convergence is studied. By utilizing generalized KYP lemma, the finite frequency performance of the iterative learning process is equivalent to solving corresponding linear matrix inequality. Furthermore, monotonically convergent criterion of tracking errors under different relative degree and explicit formula for computing control gain matrices are derived. Finally, some numerical simulations are presented to illustrate the obtained theoretical results.(3) Robust fault-tolerant iterative learning control of a multiple-input multiple-output discrete system with actuator failures is investigated. Based on two-dimensional state space model of the linear repetitive process, an integrated state feedback iterative learning control designing method is proposed. Existence conditions of the closed loop iterative learning process with fault tolerant stability and explicit for-mula for determining control gain matrices are derived by using Lyapunov stability theory. Finally, some numerical simulations are presented to justify the obtained results.(4) Adaptive iterative learning control of a parameterized continuous nonlinear system with unknown bounded disturbance and unknown time-varying input gain is investigated. By analyzing the imperfection of the traditional sliding mode control method, an adaptive iterative learning control scheme is designed with second-order sliding mode technology. The tracking errors can be proved to be converged to an controllable region by using Lyapunov stability theory. Furthermore, some numerical simulations are presented to support the obtained conclusions.(5) Adaptive iterative learning control of a linear increased discrete nonlinear system with unknown time-varying parameter is studied. By using forgetting factor least square method of traditional adaptive control, a forgetting factor least square adaptive iterative learning control scheme is proposed for trajectory tracking in finite time interval. Pointwise convergence in time domain and gradual convergence in iterative domain are proved with Lyapunov stability theory. Moreover, a numerical simulation is presented to illustrate the effectiveness of the results. |