Disturbances in iterative learning control (ILC) may be amplified if these vary from one iteration to the next, and reducing this amplification typically reduces the convergence speed. The aim of this paper is to resolve this trade-off and achieve fast convergence, robustness and small converged errors in ILC. A nonlinear learning approach is presented that uses the difference in amplitude characteristics of repeating and varying disturbances to adapt the learning gain. Monotonic convergence of the nonlinear ILC algorithm is established, resulting in a systematic design procedure. Application of the proposed algorithm demonstrates both fast convergence and small errors.

Iterative feedback tuning (IFT) enables the tuning of feedback controllers using only measured data to obtain the gradient of a cost criterion. The aim of this paper is to reduce the required number of experiments for MIMO IFT. It is shown that, through a randomization technique, an unbiased gradient estimate can be obtained from a single dedicated experiment, regardless of the size of the MIMO system. The gradient estimate is used in a stochastic gradient descent algorithm. The approach is experimentally validated on a mechatronic system, showing a significantly reduced number of experiments compared to standard IFT.

Iterative learning control (ILC) and repetitive control (RC) can lead to high performance by attenuating repeating disturbances perfectly, yet these approaches may amplify non-repeating disturbances. The aim of this paper is to achieve both perfect, fast attenuation of repeating disturbances and limited amplification of non-repeating disturbances. This is achieved by including a deadzone nonlinearity in the learning filter, which distinguishes disturbances based on their different amplitudes to apply different learning gains. Convergence conditions for nonlinear ILC and RC are developed, which are used in combination with system measurements in a comprehensive design procedure. Experimental implementation demonstrates fast learning and small errors.

Cross-coupled iterative learning control (ILC) can improve the contour tracking performance of manufacturing systems significantly. This paper aims to develop a framework for norm-optimal cross-coupled ILC that enables intuitive tuning of time- and iteration-varying weights of the exact contour error and its tangential counterpart. This leads to an iteration-varying ILC algorithm for which convergence conditions are developed. In addition, a resource-efficient implementation is developed that reduces the computational load significantly and enables the use of long reference signals. The approach is experimentally validated on an industrial flatbed printer.

The increasing complexity of next-generation mechatronic systems leads to different types of periodic disturbances, which require dedicated repetitive control strategies to attenuate. The aim of this paper is to develop a new repetitive control strategy to completely attenuate a periodic disturbance and a user-defined number of relevant higher harmonics with limited memory usage. To this end, a multirate repetitive controller is developed, which combines a buffer at a reduced sampling rate with learning and robustness filters at the original sampling rate of the system. This leads to a linear periodic time-varying system, for which convergence conditions are developed. The method is implemented on an industrial print-belt system, demonstrating that it can match the performance of traditional repetitive control while significantly reducing the memory usage.

The performance of feedforward control depends strongly on its ability to compensate for reproducible disturbances. The aim of this paper is to develop a systematic framework for artificial neural networks (ANN) for feedforward control. The method involves three aspects: a new criterion that emphasizes the closed-loop control objective, inclusion of preview to deal with delays and non-minimum phase dynamics, and enabling the use of an iterative learning algorithm to generate training data in view of addressing generalization errors. The approach is illustrated through simulations and experiments on an industrial flatbed printer.

Model-free iterative learning control (ILC) can lead to high performance by attenuating repeating disturbances completely, using dedicated experiments on the real system to replace the traditional model. The aim of this paper is to develop a fast data-driven method for MIMO ILC that uses random learning in the form of efficient unbiased gradient estimates. This is achieved by developing a stochastic conjugate gradient algorithm, in which the search direction and optimal step size are generated using dedicated experiments. The approach is applied to MIMO automated feedforward tuning. Simulation and experimental results show that the method is superior to earlier stochastic and deterministic methods.

Iterative feedback tuning (IFT) enables the tuning of feedback controllers based on measured data without the need for a parametric model. The aim of this paper is to develop an efficient method for MIMO IFT that reduces the required number of experiments. Using a randomization technique, an unbiased gradient estimate is obtained from a single dedicated experiment, regardless of the size of the MIMO system. This gradient estimate is employed in a stochastic gradient descent algorithm. Simulation examples illustrate that the approach reduces the number of experiments required to converge.

Repetitive control can lead to high performance by attenuating periodic disturbances completely, yet it may amplify non-periodic disturbances. The aim of this paper is to achieve both fast learning and low errors in repetitive control. To this end, a nonlinear learning filter is introduced that distinguishes between periodic and non-periodic errors based on their typical amplitude characteristics to adapt the extent to which they are included in the repetitive controller. Convergence conditions for the nonlinear repetitive control system are derived by casting the resulting closed-loop as a discrete-time convergent system. Simulation results of the proposed approach demonstrate fast learning and small errors.

Piezo stepper actuators are very promising for nanopositioning systems due to their high resolution, high stiffness, fast response, and the ability to position a mover over an infinite stroke by means of motion reminiscent of walking. The aim of this paper is to enhance the waveforms for actuating piezo steppers, by actively compensating for repetitive disturbances that are introduced by the walking behavior. A compensation method is developed to compensate for disturbances in the stepping domain, since disturbances may vary in the time domain if the velocity changes. The results from this procedure are exploited to determine an optimal waveform for the working range of the actuator. A significant improvement in performance is achieved after applying this waveform to a piezo stepper actuator.

Iterative learning control (ILC) involves a trade-off between perfect, fast attenuation of iteration-invariant disturbances and amplification of iteration-varying ones. The aim of this paper is to develop a nonlinear ILC framework that achieves fast convergence, robustness, and low converged error values in ILC. To this end, the method includes a deadzone nonlinearity in the learning update, which uses the difference in amplitude characteristics of repeating and varying disturbances to modify the learning gain for each error sample. A criterion for monotonic convergence of the nonlinear ILC algorithm is provided, which is used in combination with system measurements to select suitable design parameters. The proposed algorithm is validated using simulations, in which fast convergence to low error values is demonstrated.

Parameterized feedforward control is at the basis of many successful control applications with varying references. The aim of this paper is to develop an efficient data-driven approach to learn the feedforward parameters for MIMO systems. To this end, a cost criterion is minimized using a stochastic gradient descent algorithm, in which both the search direction and step size are determined through system experiments. In particular, the search direction is chosen as an unbiased estimate of the gradient which is obtained from a single experiment, regardless of the size of the MIMO system. The approach is illustrated using a simulation example, in which it is shown to be superior to a deterministic method in terms of convergence speed and thus experimental cost.

Data-driven iterative learning control can achieve high performance for systems performing repeating tasks without the need for modeling. The aim of this paper is to develop a fast data-driven method for iterative learning control that is suitable for massive MIMO systems through the use of efficient unbiased gradient estimates. A stochastic conjugate gradient descent algorithm is developed that uses dedicated experiments to determine the conjugate search direction and optimal step size at each iteration. The approach is illustrated on a multivariable example, and it is shown that the method is superior to both the earlier stochastic gradient descent and deterministic conjugate gradient descent methods.