Search

Iterative learning control (ILC) is a design technique which can achieve accurate tracking by learning over repeated task attempts. However, long-term stability remains a critical limitation to widespread application, and to-date robustness analysis has overwhelmingly considered structured uncertainties. This paper substantially expands the scope of existing ILC robustness analysis by addressing unstructured uncertainties, a widely used ILC update class, the presence of a feedback controller, and a general task description that incorporates the most recent expansions in the ILC tracking objective. Gap metric based analysis is applied to ILC by reformulating the finite horizon trial-to-trial feedforward dynamics into an equivalent along-the-trial feedback system, as well as deriving relationships to link their respective gap metric values. The results are used to generate a comprehensive design framework for robust control design of the interacting feedback and ILC loops. This is illustrated via application to rehabilitation engineering, an area where they meet an urgent need for high performance in the presence of significant modeling uncertainty. [ABSTRACT FROM AUTHOR]

Certain control applications require that performance variables are explicitly distinguished from measured variables. The performance variables are not available for real-time feedback. Instead, they are often available after a task. This enables the application of batch-to-batch control strategies such as Iterative Learning Control (ILC) to the performance variables. The aim of this paper is first to show that the pre-existing ILC controllers may not be directly implementable in this setting, and second to develop a new approach that enables the use of different variables for feedback and batch-to-batch control. The analysis reveals that by using pre-existing ILC methods, the ILC and feedback controllers may not be stable in an inferential setting. Therefore, the complete closed-loop system is cast in a 2D framework to analyze stability. Several solution strategies are outlined. The analysis is illustrated through an application example in a printing system. Finally, the developed theory also leads to new results for traditional ILC algorithms in the common situation where the feedback controller contains a pure integrator. [ABSTRACT FROM AUTHOR]

Abstract: The iterative learning control (ILC) is constructed for the discrete-time large scale systems. Each subsystem is affine nonlinear and its observation equation is with noise. Subsystems are nonlinearly connected via the large state vector of the whole system. The possibility of data missing, and communication delay is taken into account. It is proved that ILC given in the paper with probability one converges to the optimal one minimizing the tracking error. The simulation results are consistent with theoretical analysis. [Copyright &y& Elsevier]

Abstract: Design of iterative learning control (ILC) often requires some prior knowledge about a system''s control matrix. In some applications, such as uncalibrated visual servoing, this kind of knowledge may be unavailable so that a stable learning control cannot always be achieved. In this paper, a universal ILC is proposed for a class of multi-input multi-output (MIMO) uncertain nonlinear systems with no prior knowledge about the system control gain matrix. It consists of a gain matrix selector from the unmixing set and a learned compensator in a form of the positive definite discrete matrix kernel, corresponding to rough gain matrix probing and refined uncertainty compensating, respectively. Asymptotic convergence for a trajectory tracking within a finite time interval is achieved through repetitive tracking. Simulations and experiments of uncalibrated visual servoing are carried out in order to verify the validity of the proposed control method. [Copyright &y& Elsevier]

The goal of iterative learning control (ILC) is to improve the accuracy of a system that repeatedly follows a reference trajectory. This paper proves that for each causal linear time-invariant ILC, there is an equivalent feedback that achieves the ultimate ILC error with no iterations. Remarkably, this equivalent feedback depends only on the ILC operators and hence requires no plant knowledge. This equivalence is obtained whether or not the ILC includes current-cycle feedback. If the ILC system is internally stable and converges to zero error, there exists an internally stabilizing feedback that approaches zero error at high gain. Since conventional feedback requires no iterations, there is no reason to use causal ILC. [Copyright &y& Elsevier]

In this paper, a set-point related indirect-type iterative learning control (ILC) design is proposed for industrial batch processes with time-varying uncertainties and external disturbances. Different from the existing robust feedforward ILC methods solely focusing on error convergence along the batch direction, the proposed design has a double-loop control structure to conduct also dynamic control performance in the time direction as required in many engineering applications, where the inner loop is a generalized extended state observer based feedback control structure designed to ensure set-point tracking with robust stability in the time direction, and the outer loop consists of a simple proportional-type learning controller to update only the set-point command such that the tracking performance can be gradually improved along the batch direction. A tractable linear matrix inequality based sufficient condition is established to simultaneously guarantee bounded output tracking error and system input against time- and batch-varying uncertainties. An industrial injection molding process model is used to demonstrate the effectiveness and advantages of the new design in comparison to the recently developed robust feedforward ILC and indirect-type ILC designs. [ABSTRACT FROM AUTHOR]