Regular Papers

International Journal of Control, Automation, and Systems 2023; 21(9): 2957-2968

https://doi.org/10.1007/s12555-021-0858-6

© The International Journal of Control, Automation, and Systems

A Gain Adaptive Strategy to Improve Closed-loop Performance of Robust Asymptotic Feedback Linearization

Gualberto Solís-Perales*, Jairo Sánchez-Estrada, and Ricardo Femat

Universidad de Guadalajara

Abstract

This contribution presents a gain adaptation, which allows us to tune a robust asymptotic feedback linearization (RAFL). The gain adaptation allows the RAFL to attenuate the measurement noise sensitivity. The RAFL is considered here because it ensures tracking without prior information about the system’s nonlinearities and parameter bounds. Also, the RAFL only has the system output available for feedback. In this work, the robust tracking problem is faced considering: modeling errors, parametric variations, external perturbations, and noisy output measurement. On one side, the RAFL control faces modeling errors, parametric variations, and external perturbations through an observer that estimates uncertainties using an extra state, which lumps all the unknown nonlinearities and uncertainties. On the other hand, the proposed adaptive gain function allows the observer’s high gain to vary to have a fast observer’s convergence while simultaneously avoiding amplifying the measurement noise in the steadystate. The adaptive gain function provides the RAFL control robustness against noisy measurement. Thereby, the RAFL control with adaptive gain function becomes a robust feedback linearizing against to measurement noise. Finally, the RAFL controller with the adaptive gain function is illustrated by a numerical simulation of a tracking problem for a DC-motor and a chemical oxygen demand regulation in an anaerobic digestion process.

Keywords Adaptive control, adaptive observers, nonlinear control theory, robust control.

Article

Regular Papers

International Journal of Control, Automation, and Systems 2023; 21(9): 2957-2968

Published online September 1, 2023 https://doi.org/10.1007/s12555-021-0858-6

Copyright © The International Journal of Control, Automation, and Systems.

A Gain Adaptive Strategy to Improve Closed-loop Performance of Robust Asymptotic Feedback Linearization

Gualberto Solís-Perales*, Jairo Sánchez-Estrada, and Ricardo Femat

Universidad de Guadalajara

Abstract

This contribution presents a gain adaptation, which allows us to tune a robust asymptotic feedback linearization (RAFL). The gain adaptation allows the RAFL to attenuate the measurement noise sensitivity. The RAFL is considered here because it ensures tracking without prior information about the system’s nonlinearities and parameter bounds. Also, the RAFL only has the system output available for feedback. In this work, the robust tracking problem is faced considering: modeling errors, parametric variations, external perturbations, and noisy output measurement. On one side, the RAFL control faces modeling errors, parametric variations, and external perturbations through an observer that estimates uncertainties using an extra state, which lumps all the unknown nonlinearities and uncertainties. On the other hand, the proposed adaptive gain function allows the observer’s high gain to vary to have a fast observer’s convergence while simultaneously avoiding amplifying the measurement noise in the steadystate. The adaptive gain function provides the RAFL control robustness against noisy measurement. Thereby, the RAFL control with adaptive gain function becomes a robust feedback linearizing against to measurement noise. Finally, the RAFL controller with the adaptive gain function is illustrated by a numerical simulation of a tracking problem for a DC-motor and a chemical oxygen demand regulation in an anaerobic digestion process.

Keywords: Adaptive control, adaptive observers, nonlinear control theory, robust control.

IJCAS
May 2024

Vol. 22, No. 5, pp. 1461~1759

Stats or Metrics

Share this article on

  • line

Related articles in IJCAS

IJCAS

eISSN 2005-4092
pISSN 1598-6446