Pareto iterative learning control: Optimized control for multiple performance objectives

Lim, In-Gyu; Barton, Kira

doi:10.1016/j.conengprac.2014.01.011

Cited by 14 publications

(5 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…

where

and

represent the tracking error and the control signal at the kth iteration, respectively. Lim et al [104] proposed a Pareto learning control framework that integrated multiple indicators into a design framework to improve the multiple performance indicators of the system. Axelsson et al [105] modified the objective function in the optimization problem, extended the norm-optimal ILC algorithm for linear systems, and applied the Kalman filter to linear time-invariant systems.…”

Section: Optimal Controllers and Rectifying Algorithmsmentioning

confidence: 99%

A survey of methods for handling initial state shifts in iterative learning control

Chen,

Lu,

2023

Heliyon

View full text Add to dashboard Cite

“…

where

and

Section: Optimal Controllers and Rectifying Algorithmsmentioning

confidence: 99%

A survey of methods for handling initial state shifts in iterative learning control

Chen,

Lu,

2023

Heliyon

View full text Add to dashboard Cite

“…Dai Rong et al [14] proposed an iterative learning identification algorithm based on timevarying neural network, which improved the convergence speed of the algorithm and the identification accuracy of nonlinear time-varying system. Barton et al [15] proposed the Pareto iterative learning control method to discuss the optimization of multiple performance objectives. In terms of the ILC norm optimization theory, Son et al [16] developed an ILC optimization algorithm based on pointto-point robust monotone convergence.…”

Section: Introductionmentioning

confidence: 99%

Multi-Axis Motion Control Based on Time-Varying Norm Optimal Cross-Coupled Iterative Learning

Hou

et al. 2020

IEEE Access

View full text Add to dashboard Cite

In the process of multi-axis contour tracking control, the traditional time-invariant method could lead to a significant error in contour tracking due to the existence of two different motion conditions, namely single-axis independent motion and multi-axis coupled motion. In order to tackle this issue, a timevarying weighting matrix has been developed considering the trajectory and time-varying random disturbance. In this paper, a time-varying control method for multi-axis motion based on norm optimal cross-coupling iterative learning is proposed. Compared to the time-invariant control method, the simulation and experiment results demonstrate that the proposed method can effectively reduce the contour error improving the multi-axis control precision. INDEX TERMS Time-varying weighted matrix, Iterative learning, Norm optimal, cross coupling, Multiaxis motion control.

show abstract

“…Aiming to address the undesirable large transient behavior in the iteration domain, the explicit optimal cost functions were introduced and minimized to design the norm optimal ILC (NOILC) algorithms , and guarantee the monotonic error convergence along the iteration direction. The lifted representation with supervector approach is often used in the design of NOILC . The linear system is reformulated to map the input vector to the output vector with the system matrix, where the initial states are set as straightforward zeros .…”

Section: Introductionmentioning

confidence: 99%

“…More recently a unified data‐driven design framework has been proposed for optimality‐based general iterative learning control. Different from the NOILCs and MPC‐based ILCs , the proposed approach in does not use any process model information in the controller design but the I/O measurements. However, it is still based on a lifting technique and the control law includes a matrix operation, which leads to a time‐consuming computation because of the increasing matrix dimension with the data points.…”

Section: Introductionmentioning

confidence: 99%

Computationally‐Light Non‐Lifted Data‐Driven Norm‐Optimal Iterative Learning Control

Chi

Hou

Jin

et al. 2017

Asian Journal of Control

View full text Add to dashboard Cite

Computational complexity and model dependence are two significant limitations on lifted norm optimal iterative learning control (NOILC). To overcome these two issues and retain monotonic convergence in iteration, this paper proposes a computationally-efficient non-lifted NOILC strategy for nonlinear discrete-time systems via a data-driven approach. First, an iteration-dependent linear representation of the controlled nonlinear process is introduced by using a dynamical linearization method in the iteration direction. The non-lifted NOILC is then proposed by utilizing the input and output measurements only, instead of relying on an explicit model of the plant. The computational complexity is reduced by avoiding matrix operation in the learning law. This greatly facilitates its practical application potential. The proposed control law executes in real-time and utilizes more control information at previous time instants within the same iteration, which can help improve the control performance. The effectiveness of the non-lifted data-driven NOILC is demonstrated by rigorous analysis along with a simulation on a batch chemical reaction process.

show abstract

Pareto iterative learning control: Optimized control for multiple performance objectives

Cited by 14 publications

References 25 publications

A survey of methods for handling initial state shifts in iterative learning control

A survey of methods for handling initial state shifts in iterative learning control

Multi-Axis Motion Control Based on Time-Varying Norm Optimal Cross-Coupled Iterative Learning

Computationally‐Light Non‐Lifted Data‐Driven Norm‐Optimal Iterative Learning Control

Contact Info

Product

Resources

About