Traditional Iterative Learning Control Research Articles

Inferential Iterative Learning Control: A 2D-system approach

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.

Optimization Control of the Color-Coating Production Process for Model Uncertainty.

Optimized control of the color-coating production process (CCPP) aims at reducing production costs and improving economic efficiency while meeting quality requirements. However, because optimization control of the CCPP is hampered by model uncertainty, a strategy that considers model uncertainty is proposed. Previous work has introduced a mechanistic model of CCPP based on process analysis to simulate the actual production process and generate process data. The partial least squares method is then applied to develop predictive models of film thickness and economic efficiency. To manage the model uncertainty, the robust optimization approach is introduced to improve the feasibility of the optimized solution. Iterative learning control is then utilized to further refine the model uncertainty. The constrained film thickness is transformed into one of the tracked targets to overcome the drawback that traditional iterative learning control cannot address constraints. The goal setting of economic efficiency is updated continuously according to the film thickness setting until this reaches its desired value. Finally, fuzzy parameter adjustment is adopted to ensure that the economic efficiency and film thickness converge rapidly to their optimized values under the constraint conditions. The effectiveness of the proposed optimization control strategy is validated by simulation results.

Iterative learning control for nonlinear dynamic systems with randomly varying trial lengths

SummaryIn this paper, we introduce an iterative learning control (ILC) scheme based on an iteratively moving average operator for nonlinear dynamic systems with randomly varying trial lengths. By using the iteratively moving average operator, the proposed ILC algorithm overcomes the limitation of traditional ILC that all trial lengths must be identical. It is shown that for nonlinear affine and non‐affine systems, the proposed learning algorithm works effectively to nullify the tracking error. In the end, two illustrative examples are presented to demonstrate the performance and the effectiveness of the proposed ILC scheme for nonlinear dynamic systems. Copyright © 2015 John Wiley & Sons, Ltd.

Improvement of trajectory tracking control performance by using ILC

This paper presents an iterative learning control (ILC) approach for tracking problems with specified data points that are desired points at certain time instants. To design ILC systems for such problems, unlike traditional ILC approaches, an algorithm which updates not only the control signal but also the reference trajectory at each trial will be developed. The relationship between the reference trajectory and ILC control in tracking problems where there are specified data points through which the system should pass is investigated as the rate of convergence. In traditional ILC, the desired data is stored in a tracking profile file. Due to the huge size of the data file containing the target points, it is important to reduce the computational cost. Finally, simulation results of the presented technique are mentioned and compared to other related works to confirm the effectiveness of proposed scheme.

A study on the optimal tracking problems with predefined data by using iterative learning control

In this paper, we present an iterative learning control (ILC) framework for tracking problems with predefined data points that are desired points at certain time instants. To design ILC systems for such problems, a new ILC scheme is proposed to produce output curves that pass close to the desired points. Unlike traditional ILC approaches, an algorithm will be developed in which the control signals are generated by solving an optimal ILC problem with respect to the desired sampling points. In another word, it is a direct approach for the multiple points tracking ILC control problem where we do not need to divide the tracking problem into two steps separately as trajectory planning and ILC controller. The strength of the proposed formulation is the methodology to obtain a control signal through learning law only considering the given data points and dynamic system, instead of following the direction of tracking a prior identified trajectory. The key advantage of the proposed approach is to significantly reduce the computational cost. Finally, simulation results will be introduced to confirm the effectiveness of proposed scheme.

Adaptive iterative learning control based on unfalsified strategy for Chylla-Haase reactor

An adaptive iterative learning control based on unfalsified strategy is proposed to solve high precision temperature tracking of the Chylla-Haase reactor, in which iterative learning is the main control method and the unfalsified strategy is adapted to adjust the learning rate adaptively. It is encouraged that the unfalsified control strategy is extended from time domain to iterative domain, and the basic definition and mathematics description of unfalsified control in iterative domain are given. The proposed algorithm is a kind of data-driven method, which does not need an accurate system model. Process data are used to construct fictitious reference signal and switch function in order to handle different process conditions. In addition, the plant data are also used to build the iterative learning control law. Here the learning rate in a different error level is adjusted to ensure the convergent speed and stability, rather than keeping constant in traditional iterative learning control. Furthermore, two important problems in iterative learning control, i.e., the initial control law and convergence analysis, are discussed in detail. The initial input of first iteration is arranged according to a mechanism model, which can assure a good produce quality in the first iteration and a fast convergence speed of tracking error. The convergence condition is given which is obviously relaxed compared with the tradition iterative learning control. Simulation results show that the proposed control algorithm is effective for the Chylla-Haase problem with good performance in both convergent speed and stability.

Forecast Iterative Learning Control with Forgetting Factor and Its Application in Valve-controlled Asymmetric Cylinder

Valve-controlled asymmetric cylinder belongs to typical nonlinear and time-varying system at elastic load, in order to eliminate oscillation result from traditional PD-type iterative learning control, a kind of forecast iterative learning control with a forgetting factor is designed. The system mathematical model is built, and the performance characteristics of valve-controlled asymmetric cylinder system are analyzed. The fundamental of iterative learning control is introduced, the problems of traditional PD-type algorithm are analyzed; a forgetting factor is added in the iterative learning control to inhibit the oscillation, the control effect by different forgetting factor value is analyzed by simulating, a part of forecast system given is increased to improve the system accuracy, because forgetting factor lower the system performance. Both simulations and experiments show that, in the valve-controlled asymmetric cylinder system, using forecast iterative learning control with a forgetting factor will overcome oscillation well with a forgetting factor, and can improve accuracy by forecast iterative.

An improvement on the transient response of tracking for the sampled-data system based on an improved PD-type iterative learning control

An improvement on the transient response of tracking for the sampled-data system based on an improved PD-type iterative learning control (ILC) is proposed in this paper. The developed analog ILC method and the high-gain property tracker design methodology are first combined to significantly reduce learning epochs and overcome the initial condition shift problem and discontinuous reference input in the traditional ILC. Besides, the proposed ILC improves the transient response and decreases the rate of weighting matrices Q to R under the traditional linear quadratic tracker design. First, the off-line observer/Kalman filter identification (OKID) is used to determine the appropriate (low-) order system parameters and state estimator for the physical system with unknown system equation, so that the model-based PD-type ILC can be implemented for practical applications. Then, to improve the transient response and decrease the control effort, the proportional difference type (PD-type) ILC algorithm is combined with the high-gain property linear quadratic tracker (LQT) design to construct the high performance tracker for the model-based sampled-data systems. Furthermore, the discrete-time version high performance tracker design for the unknown stochastic sampled-data system via the iterative learning control method is proposed in this paper based on the Euler method and the digital redesign approach. Finally, some examples are given for illustrating the effectiveness of the proposed method.

Iterative learning control in optimal tracking problems with specified data points

In this paper, we present two iterative learning control (ILC) frameworks for tracking problems with specified data points that are desired points at certain time instants. To design ILC systems for such problems, unlike traditional ILC approaches, we first develop an algorithm in which not only the control signal but also the reference trajectory is updated at each trial. We investigate the relationship between the reference trajectory and ILC tracking control as it relates to the rate of convergence. Second, a new ILC scheme is proposed to produce output curves that pass close to the desired points. Here, the control signals are generated by solving an optimal ILC problem with respect to the desired sampling points. One of the key advantages of the proposed approaches is a significant reduction of the computational cost.

Data-driven Based Integrated Learning Controller Design for Batch Processes

The challenge of optimization control of batch processes is how to combine both discrete-time (batch-axis) information and continuous-time (time-axis) information into an integrated frame when designing optimal controller. By using data-driven technology, a novel integrated learning control system is proposed in this paper. Firstly, an iterative learning controller (ILC) is designed along the direction of batch-axis, and then an adaptive single neuron predictive controller (SNPC) that plays role of feedback controller along the direction of time-axis is devised accordingly. As a result, the integrated control system is very effective to eliminate modeling error and uncertainty, which is superior to traditional ILC. In addition, the self-tuning algorithm of SNPC controller is derived by a rigorous analysis based on the Lyapunov method such that the predicted tracking error convergences asymptotically. Lastly, to verify the efficiency of the proposed control scheme, it is applied to a benchmark batch process. The simulation results show that the proposed method has better stability and robustness compared with the traditional iterative learning control.

10.2478/s11814-009-0322-4

Adaptive iterative learning control based on the measured input-output data is proposed to solve the traditional iterative learning control problem in the batch process. It produces a control law with self-tuning capability by combining a batch-to-batch model estimation procedure with the control design technique. To build the unknown batch operation system, the finite impulse response (FIR) model with the lifted system is constructed for easy construction of a recursive least squares algorithm. It can identify the pattern of the current operation batch. The proposed model reference control method is applied to feedback control of the lifted system. It finds an appropriate control input so that the desired performance of the batch output can track the prescribed finite-time trajectory by iterative trials. Furthermore, on-line tracking control is developed to explore the possible adjustments of the future input trajectories within a batch. This can remove the disturbances in the current batch rather than the next batch trial and keep the product specifications consistent at the end of each batch. To validate the theoretical findings of the proposed strategies, two simulation problems are investigated.

Initial iterative learning control strategy based on CMAC neural networks

In order to avoid the problem of slow converging speed and long time expenditure when traditional iterative learning control system faced a new environment an improved algorithm was proposed to obtain the initial value of the iterative learning control based on CMAC neural networks.Desired control input of iterative learning control that was estimated by CMAC neural networks and feedback PID networks based on the historical control experience worked as the initial control input of the iterative learning control.With the role of open-closed loop P-type iterative learning control algorithm the actual output trajectory of the system could track desired trajectory in accurate requirements using less iteration.The simulation results of the robotic system show the algorithm is feasible and effective.

Iterative learning controller synthesis using FIR models for batch processes

Adaptive iterative learning control based on the measured input-output data is proposed to solve the traditional iterative learning control problem in the batch process. It produces a control law with self-tuning capability by combining a batch-to-batch model estimation procedure with the control design technique. To build the unknown batch operation system, the finite impulse response (FIR) model with the lifted system is constructed for easy construction of a recursive least squares algorithm. It can identify the pattern of the current operation batch. The proposed model reference control method is applied to feedback control of the lifted system. It finds an appropriate control input so that the desired performance of the batch output can track the prescribed finite-time trajectory by iterative trials. Furthermore, on-line tracking control is developed to explore the possible adjustments of the future input trajectories within a batch. This can remove the disturbances in the current batch rather than the next batch trial and keep the product specifications consistent at the end of each batch. To validate the theoretical findings of the proposed strategies, two simulation problems are investigated.

A novel design strategy for iterative learning and repetitive controllers of systems with a high modal density: Theoretical background

This paper discusses the design and application of iterative learning control (ILC) and repetitive control (RC) for high modal density systems. Typical examples of these systems are structural and acoustical systems considered in active structural acoustic control (ASAC) and active noise control (ANC) applications. The application of traditional ILC and RC design techniques, which are based on a parametric system model, on systems with a high modal density has several important drawbacks: the design procedure is complex, the controllers require much computational power and the robustness of the controllers is low. This paper describes a novel strategy to design noncausal ILC and RC filters, which is especially suited for high modal density systems. Since it does not require a parametric system model, the novel strategy avoids several drawbacks of the traditional techniques: no cumbersome parametric model estimation is required; the ILC and RC controllers are robust to small changes of the poles and zeros of the controlled system; and the complexity of the ILC and RC control filters is restricted. A crucial element in the proposed strategy is the noncausal filtering in the ILC and RC controllers, which requires the availability of a trigger signal to announce a new ILC trial or RC period in advance. A numerical validation on a simulation model proves the potential of the developed strategy.

A new neural network-based adaptive ILC for nonlinear discrete-time systems with dead zone scheme

By introducing a dead-zone scheme, a new neural network based adaptive iterative learning control (ILC) (NN-AILC) scheme is presented for nonlinear discrete-time systems, where the NN weights are time-varying. The most distinct contribution of the proposed NN-AILC is the relaxation of the identical conditions of initial state and reference trajectory, which are common requirements in traditional ILC problems. Convergence analysis indicates that the tracking error converges to a bounded ball, whose size is determined by the dead-zone nonlinearity. Computer simulations verify the theoretical results.

Spatial-based iterative learning control for motion control applications

For many motion control applications spatial constraints are often more important than temporal constraints. In recent work, we have developed a spatial control strategy called the e-controller for mobile robot applications. The control strategy is based solely on static path geometry with position (in space) feedback. Motivated by this idea, in this paper, we consider the notion of spatial-based iterative learning control (ILC). Specifically, we consider repetitive operation problems where corrections are made to the control signal from trial to trial. Unlike traditional ILC, however, which updates control signals based on the time elapsed along a trajectory, we instead make updates based on path errors and progress along the path. The idea is demonstrated via simulation for a system with bang–bang velocity control. Experimental results using a high-precision, two-axis gimbal mechanism are presented to show the effectiveness of the strategy.

Model reference adaptive iterative learning control for linear systems

AbstractIn this paper, we propose a model reference adaptive control (MRAC) strategy for continuous‐time single‐input single‐output (SISO) linear time‐invariant (LTI) systems with unknown parameters, performing repetitive tasks. This is achieved through the introduction of a discrete‐type parametric adaptation law in the ‘iteration domain’, which is directly obtained from the continuous‐time parametric adaptation law used in standard MRAC schemes. In fact, at the first iteration, we apply a standard MRAC to the system under consideration, while for the subsequent iterations, the parameters are appropriately updated along the iteration‐axis, in order to enhance the tracking performance from iteration to iteration. This approach is referred to as the model reference adaptive iterative learning control (MRAILC). In the case of systems with relative degree one, we obtain a pointwise convergence of the tracking error to zero, over the whole finite time interval, when the number of iterations tends to infinity. In the general case, i.e. systems with arbitrary relative degree, we show that the tracking error converges to a prescribed small domain around zero, over the whole finite time interval, when the number of iterations tends to infinity. It is worth noting that this approach allows: (1) to extend existing MRAC schemes, in a straightforward manner, to repetitive systems; (2) to avoid the use of the output time derivatives, which are generally required in traditional iterative learning control (ILC) strategies dealing with systems with high relative degree; (3) to handle systems with multiple tracking objectives (i.e. the desired trajectory can be iteration‐varying). Finally, simulation results are carried out to support the theoretical development. Copyright © 2006 John Wiley & Sons, Ltd.

On iterative learning from different tracking tasks in the presence of time-varying uncertainties.

In this paper, we introduce a new iterative learning control (ILC) method, which enables learning from different tracking control tasks. The proposed method overcomes the imitation of traditional ILC in that, the target trajectories of any two consecutive iterations can be completely different. For non-linear systems with time-varying and time-invariant parametric uncertainties, the new learning method works effectively to nullify the tracking error. To facilitate the learning control system design and analysis, in the paper we use a composite energy function (CEF) index, which consists of a positive scalar function and L2 norm of the function approximation error.

An iterative learning controller with initial state learning

In iterative learning control (ILC), a common assumption is that the initial states in each repetitive operation should be inside a given ball centred at the desired initial states which may be unknown. This assumption is critical to the stability analysis, and the size of the ball will directly affect the final output trajectory tracking errors. In this paper, this assumption is removed by using an initial state learning scheme together with the traditional D-type ILC updating law. Both linear and nonlinear time-varying uncertain systems are investigated. Uniform bounds for the final tracking errors are obtained and these bounds are only dependent on the system uncertainties and disturbances, yet independent of the initial errors. Furthermore, the desired initial states can be identified through learning iterations.