Continuous Super-Twisting Observer-Based Super-Twisting Control of Euler-Lagrange Systems

Authors: Cao, G., Shi, H., Zhang, X., Zhu, Z., Liu, Y. and Yu, H.

Journal: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Volume: 13013 LNAI

Pages: 455-465

eISSN: 1611-3349

ISBN: 9783030890940

ISSN: 0302-9743

DOI: 10.1007/978-3-030-89095-7_44

Abstract:

Higher-order observer-based robust controllers are commonly applied to Euler-Lagrange (EL) systems in presence of unavoidable uncertainties and unmeasurable states. However, the controllers are constructed by an assumption of differential system uncertainties, which constrains the physical applications disrupted by non-differentiable noises, such as discontinuous signals. This paper proposed a continuous observer-based sliding mode controller by utilizing the super-twisting algorithm (STA) and an integral sliding mode surface, and the condition of uncertainties is relaxed to be bounded. First, an improved STA-based observer is employed to estimate system states and adaptively compensate system uncertainties. Second, a novel integral sliding mode surface is employed to design a continuous STA-based controller for EL systems in presence of uncertainties without knowledge of upper bound in prior. The proposed controller relaxes the differential uncertainties to be bounded, achieve continuous inputs and improve system robustness. Finally, numerical simulations verify the effectiveness of the proposed controller.

Source: Scopus