Hostname: page-component-7b7567c8c-7pwfn Total loading time: 0 Render date: 2026-11-16T11:15:46.539Z Has data issue: false hasContentIssue false

Adaptive impedance control of uncertain robot manipulators with saturation effect based on dynamic surface technique and self-recurrent wavelet neural networks

Published online by Cambridge University Press:  05 October 2018

Mohammad Hossein Hamedani
Affiliation:
Department of Electrical and Computer Engineering, Isfahan University of Technology, Isfahan 84156-83111, Iran. E-mails: mh.hamedani@ec.iut.ac.ir, sheikh@cc.iut.ac.ir
Maryam Zekri*
Affiliation:
Department of Electrical and Computer Engineering, Isfahan University of Technology, Isfahan 84156-83111, Iran. E-mails: mh.hamedani@ec.iut.ac.ir, sheikh@cc.iut.ac.ir
Farid Sheikholeslam
Affiliation:
Department of Electrical and Computer Engineering, Isfahan University of Technology, Isfahan 84156-83111, Iran. E-mails: mh.hamedani@ec.iut.ac.ir, sheikh@cc.iut.ac.ir
*
*Corresponding author. E-mail: mzekri@cc.iut.ac.ir
Rights & Permissions [Opens in a new window]

Summary

Core share and HTML view are not available for this content. However, as you have access to this content, a full PDF is available via the 'Save PDF' action button.

Saturation nonlinearities, among the known challenges in control engineering, are ubiquitous in robotic systems and can lead to stability and performance degradation. In this paper, an adaptive dynamic surface impedance (ADSI) control approach is developed for an n-link robotic manipulator by employing self-recurrent wavelet neural networks (SRWNNs) in order to overcome the saturation effect. The proposed control approach is inspired by the theory of dynamic surface control (DSC) and SRWNNs. As a novel application of the dynamic surface method to obtain a simple structure, the target impedance is formulated in the state–space, and effective dynamic surfaces are defined to track the desired impedance behavior. In fact, DSC is used to force the robot manipulator to track the desired impedance, while the robot interacts with an environment. In addition, SRWNNs are applied to approximate the parametric uncertainties and external disturbances in the robot dynamical model. Self-feedback neurons are embedded as memory units in SRWNNs to model the sudden dynamic jumps of the environment. Using Lyapunov's method, an ADSI controller is designed, and adaptation laws are induced to guarantee the stability of the closed-loop system. Finally, simulations are conducted to verify the proper performance of the proposed approach for removing the saturation effect and tracking the target impedance. It is worth noting that the simulation results indicate the robustness of the controller against uncertainties and external disturbances.

Information

Type
Articles
Copyright
Copyright © Cambridge University Press 2018