Online data-driven incremental life-long learning control for uncertain robotic manipulators via self-evolving interval type-2 fuzzy systems

In this article, a novel self-evolving interval type-2 fuzzy systems based incremental life-long learning controller (SEIT2FS-ILLC) is proposed for robotic manipulators with unmodeled dynamics and external disturbance. The proposed controller structure that combines PID controller and SEIT2FS in par...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Nonlinear dynamics Jg. 113; H. 17; S. 23225 - 23244
Hauptverfasser: Su, Qinyin, Zhao, Tao
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Dordrecht Springer Nature B.V 01.09.2025
Schlagworte:
Controllers
Data processing
Design
Distance learning
Fuzzy logic
Fuzzy sets
Fuzzy systems
Lifelong learning
Manipulators
Mathematical models
Methods
Neural networks
Proportional integral derivative
Robot arms
Robot control
Robotics
Stability
Transient performance
Unmanned aerial vehicles
ISSN:0924-090X, 1573-269X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In this article, a novel self-evolving interval type-2 fuzzy systems based incremental life-long learning controller (SEIT2FS-ILLC) is proposed for robotic manipulators with unmodeled dynamics and external disturbance. The proposed controller structure that combines PID controller and SEIT2FS in parallel, allows for the construction of rule bases from scratch without the need for offline pretraining, while retaining the life-long learning and self-evolving structure abilities of SEIT2FS. Meanwhile, the introduction of FLS as a compensation controller in the proposed controller to improve the transient performance of the SEIT2FS-ILLC. Unlike most previous results, the proposed method does not rely on the mathematical model of the robotic manipulators but only on the input-output data of its operation. Utilizing the Lyapunov stability theory, the stability of the proposed controller is demonstrated. Both numerical simulations and actual robotic manipulators experiment to validate that the proposed method is effective.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0924-090X
1573-269X
DOI:10.1007/s11071-025-11348-0