Search

This paper presents a systematic method to design a high-order repetitive model reference controller (HO-RMRC) for a discrete-time linear system that is subject to uncertain periodic references and disturbances. The controller design method we propose is established by combining two controller design schemes: high-order repetitive control (HORC) and model reference control (MRC). Here, the HORC is used to suppress uncertain periodic disturbances significantly and the MRC is used to track time-varying references without introducing new transient responses. The stability analysis of the resulting closed-loop system is provided and describes how admissible learning gains are determined to achieve stability and a desirable convergence rate. The efficacy of the proposed control method is demonstrated via numerical simulations where the performance of the resulting HO-RMRC is compared with those of the HORC and the MRC as standalone controllers. It is evident that the HO-RMRC outperforms the HORC and the MRC in simultaneously achieving perfect reference tracking and disturbance cancellation/attenuation. [ABSTRACT FROM AUTHOR]

This paper presents a new approach to kinematic modeling and trajectory tracking control system (TTCS) design of an autonomous two-wheeled robot. The robot kinematic modeling is carried out based on a vector diagram. It transforms the robot trajectory tracking problem into a stabilization problem of non-linear posture error dynamics. A solution is presented by applying a state feedback control. The posture error dynamics is linearized using the Taylor series and linear quadratic regulator (LQR) is applied to design the state feedback control. It results in a closed-loop system where all of the eigenvalues are real negative numbers. Performance evaluation of the TTCS is presented through numerical simulations in computer. The results show that the robot is able to track a reference trajectory without oscillation regardless of the robot initial posture. [ABSTRACT FROM AUTHOR]

A new method to construct a discrete-time variable-structure repetitive controller for a class of linear systems perturbed by multiple-period exogenous signals is presented. The proposed control scheme combines the features of the discrete-time multiple-period repetitive control (MP-RC) and variable-structure control (VSC) techniques. The MP-RC part is assigned to simultaneously track and reject periodic signals consisting of multiple uncorrelated fundamental frequencies. The VSC part is then integrated to provide a fast transient response and robustness against plant parameter variations. Stability and robustness analyses are also elaborated to ensure that the resulting closed-loop system satisfies the desired control objectives. Moreover, it is shown through an example that the repetitive control system constructed using the proposed control method can effectively track a sinusoidal reference signal despite the presence of a multiple-period disturbance. [ABSTRACT FROM AUTHOR]

SUMMARY This paper presents a new method to construct a decentralized nonlinear robust H ∞ controller for a class of large-scale nonlinear uncertain systems. The admissible uncertainties and nonlinearities in the system satisfy integral quadratic constraints and global Lipschitz conditions, respectively. The decentralized controller, which is required to be stable, is capable of exploiting known nonlinearities and interconnections between subsystems without treating them as uncertainties. Instead, additional uncertainties are introduced because of the discrepancies between nondecentralized and decentralized nonlinear output feedback controllers. The H ∞ control objective is to achieve an absolutely stable closed-loop system with a specified disturbance attenuation level. A solution to this control problem involves stabilizing solutions to algebraic Riccati equations parametrized by scaling constants corresponding to the uncertainties and nonlinearities. This formulation is nonconvex; hence, an evolutionary optimization method is applied to solve the control problem considered. Copyright © 2012 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]